3

High-Priority Research Areas and Topics

This chapter describes 10 high-priority research areas, each of which contains one to three high-priority research topics. Five of the research areas focus on disciplines; the other five focus on systems:

Disciplines

Systems

As discussed in the Chapter 4 section “Interrelationship Among Goals, Research Areas, and Research Topics,” successful completion of each of the research topics will help achieve several of the goals discussed in Chapter 2. Likewise, as discussed throughout this chapter, the success of each research area is tied to the success of multiple other research areas.¹

___________________

¹ All but two of the research areas have three research topics. The Digital Twins and Their Supporting Infrastructure research area has one research topic, and the Gas Turbines in Pipeline Applications research area has two research topics. The distribution of the research topics is not intended to indicate that these research areas or their research topics (1) are of less importance than the other research areas or (2) should receive either more or less resources than the others. Rather, the committee concluded that in most research areas there are multiple research topics of comparable importance, whereas in the Digital Twins and Their Supporting Infrastructure research area, there is a one overriding research topic. All 10 research areas are of comparable priority. The order in which they appear was selected to facilitate the explanation of each research area because in some cases the details associated with one research area provide a foundation for understanding other research areas.

RESEARCH AREA 1: COMBUSTION

Overview

The combustor has important impacts on emissions, controls the range of alternative fuels that can be used, and has important influences on the operational limits of a gas turbine.² This section describes three research topics that are needed to enable optimal performance and minimal emissions with a range of alternative fuels (e.g., low carbon fuels) and oxidizers.

Early gas turbine combustion research and development focused on fuel injection technologies that would (1) produce stable flames that did not blow off, (2) ensure complete combustion, and (3) produced acceptable levels of harmful emissions. Over the past two decades, increased restrictions on oxides of nitrogen (NO_x) emissions have led to fundamental shifts in combustion design approaches, increasingly pushing systems toward premixed designs and redirecting significant research and development by government and industry accordingly. This has led to significant research to understand flame stability, emissions, turbulent combustion, and combustion chemistry. More recently, greater interest in improving the efficiency of gas turbine technologies has pushed research and development to understand these issues at increasingly severe pressures and temperatures. In addition, there has been growing interest in cycles that approximate constant volume combustion, such as pulse-detonation or rotating detonation engines. This has motivated work in detonation limits, detonation wave dynamics, and the transition from deflagration (in which combustion propagates at subsonic speeds) to detonation (in which combustion propagates at supersonic speeds). Last, interest in decarbonizing gas turbine systems and developing systems that enable carbon capture has motivated interest in alternative fuels (e.g., hydrogen and biofuels) and oxyfuel combustion.

This research area includes three research topics:

• Fundamental Combustion Properties
• Combustion Concepts to Reduce Harmful Emissions at Elevated Temperatures and Pressures
• Operational and Performance Limits on Combustors

Research Topic 1.1
Fundamental Combustion Properties

Summary

Combustion systems must meet NO_x, carbon monoxide (CO), and particulate emissions specifications, while enabling sufficient load turndown and operability (i.e., by avoiding key limiting phenomena such as flashback, blowoff, detonation limits, autoignition, and combustion instability). For a variety of fundamental reasons, combustors have inherent trade-offs among emissions, turndown and operability, and they are generally optimized for

___________________

² A. Amato, M. Cheng, and T. Lieuwen, 2013, Stationary gas turbine combustion: Technology needs and policy considerations, Combustion and Flame 160:1311-1314, doi:10.1016/j.combustflame.2013.05.001.

a narrow fuel composition specification. Research is needed to better understand fundamental combustion properties and fluid mechanics (e.g., turbulent flame speeds and flame response to flow disturbances) at pressures and temperatures of interest), as well as to develop combustion paradigms that are less sensitive to fuel composition. Moreover, many of these issues involve the coupling between combustion and fluid mechanics (e.g., instabilities of swirling or transverse jets in the presence of combustion) at conditions of interest. While research around general questions of this nature is commonplace in the combustion community, there is little activity or data at conditions of interest for gas turbines because relevant facilities are expensive to maintain and operate and they require specialized staff to manage operational safety issues. For example, the higher temperatures and pressures associated with some techniques for improving efficiency are pushing combustion physics into regimes of combustion where fuel consumption occurs via autoignition. In contrast, fuel consumption and heat release generally occur through flame propagation in facilities that operate at lower pressures and temperatures. Similarly, large-scale computations at conditions of interest become increasingly cost prohibitive, due to the very high Reynolds numbers encountered at these conditions.

Background

A variety of quantities are needed by designers in development of gas turbine combustors. Properties of interest include the following:

• Flame stability and blowoff: extinction stretch rates, edge flame speeds and limit conditions, flame speeds, and detonation limits.
• Combustion instability: forced flame response, receptivity of reacting free shear flows to disturbances, and turbulent flame speed.
• Flame location and position: extinction stretch rates and turbulent flame speed in flame propagation and autoignition regimes.
• Emissions: high-pressure kinetics.
• Detonation strength: factors influencing the strength and speed of detonations.

As noted previously, acquiring these types of data at conditions of interest is quite challenging.

Benefit

This research topic could accelerate ongoing development by providing designers with a better knowledge of critical combustion processes that influence key design decisions, such as where the flame is located, limit behaviors, and emissions. The ability to accurately predict these behaviors rather than having to measure them during tests would reduce uncertainties and enable designers to explore fundamentally new designs more boldly, rather than rely on derivatives with known behaviors.

This research topic could advance the ability to predict relevant knowledge of these properties at conditions of interest by 2030.³ Providing designers with a priori knowledge of combustion properties will significantly reduce risk and enable them to evaluate designs during the design stage.

The technical risk of this research topic is medium because required experiments are complex and difficult and because there is little existing research infrastructure to support them. Requisite data do not exist because the data are expensive to acquire and relatively few laboratories have the high-pressure capabilities to perform them. First principles computations are similarly very expensive. Nonetheless, there is no fundamental reason that this topic cannot be satisfactorily addressed.

___________________

³ A projected technology readiness level (TRL) has not been specified for this research topic because it deals with improving scientific knowledge, not technology development.

Research Topic 1.2
Combustion Concepts to Reduce Harmful Emissions at Elevated Temperatures and Pressures

Summary

Combustors that can operate at higher temperatures and pressures will increase gas turbine efficiency. This is true of gas turbines using either a constant pressure combustion thermodynamic cycle (e.g., a Brayton cycle, which is used by conventional gas turbines) or a pressure gain combustion (PGC) thermodynamic cycle.⁴ Unfortunately, the NO_x formation rate increases exponentially with temperature, and it also increases with pressure. Moreover, potential strategies to reduce NO_x emissions may increase CO or particulate emissions. This challenge is relatively new, as conditions in earlier gas turbine designs made it possible to improve efficiency without increasing NO_x though tailored mixing of air and fuel. As a result, little work has been done to address this challenge. This challenge is also unique to gas turbine applications, as there are essentially no other technology platforms with combustion at these elevated temperatures and pressures.

Background

Historically, reducing NO_x levels in gas turbines with turbine inlet temperatures less than about 1,500°C/2,800°F has been achieved through premixing of the fuel and air. More recently, some aircraft engine original equipment manufacturers (OEMs), however, have used staged combustion schemes such as rich-burn, quick-quench, lean burn (RQL). In either case, advances in materials and cooling technologies have led to higher turbine inlet temperatures during which significant NO_x production occurs, even with perfect premixing. Moreover, gas turbines with PGC may use nonpremixed combustion concepts and high local temperatures, followed by potentially rapid cooling or quenching across the post-detonation expansion wave. This approach can lead to higher levels of NO_x, particulates, CO, and unburned hydrocarbons.

New combustion paradigms are needed to enable low NO_x emissions when using high turbine inlet temperatures and/or PGC systems. NO_x production is proportional to the NO_x formation rate and combustor residence time. Thus, NO_x production can be reduced by developing combustors that (1) minimize NO_x formation rates, and/or (2) reduce residence time while maintaining sufficient turndown and operability characteristics. For example, theoretical work has demonstrated that distributed fuel injection can reduce NO_x emissions by an order of magnitude relative to conventional dry, low NO_x combustor approaches, which are typically used in high-efficiency ground-based gas turbines.⁵ This reduction will occur, however, only if mixing from the distributed fuel injection occurs quickly relative to reactions. Otherwise, distributed fuel injection can actually increase NO_x emissions relative to conventional approaches. Research is needed to develop workable systems whose performance approaches the theoretical limits.

Benefit

This research topic could accelerate ongoing research by developing new paradigms for combustion with lower emissions. Current designs are essentially derivatives of older ones, with technology focused on tailored premixing and, more recently, axial staging of fuel. New research could focus on ways to further extend current capabilities by developing advanced derivatives of current designs and fundamentally new approaches.

This research topic could advance relevant technology from TRL 1 to TRL 6 by 2030. Without additional research, it is likely that the design community will focus on derivatives of current designs to the exclusion of new combustion designs.

___________________

⁴ For more information on gas turbine thermodynamic cycles, see Box 1.1 in Chapter 1 and the section “Research Area 6: Unconventional Thermodynamic Cycles” later in this chapter.

⁵ E. Goh., M. Sirignano, V. Nair, B. Emerson, T. Lieuwen, and J. Seitzman, 2019, Prediction of minimum achievable NO_x levels for fuel-staged combustors, Combustion and Flame 200:276-285, doi:10.1016/j.combustflame.2018.11.027.

This research topic has medium technical risk because historical experience in fielding low NO_x combustion systems has shown significant field problems with gas turbine operability, durability, and emissions. Nonetheless, theoretical work has demonstrated that order of magnitude reductions are possible relative to current combustors with single-point fuel injection, and so progress toward these limits is achievable. Moreover, these concepts can be incrementally deployed, moving from less to more aggressive, enabling successive reductions in NO_x emissions.

PGC increases technical risk, as these systems involve impulsive shock loading on components, raising concerns for life of and wear on components that must operate continuously for several years without replacement. These technologies will find early adoption in applications, such as missiles or rockets, that do not require a long life or a long time between inspections. Experience with these applications will provide crucial insights into their appropriateness for power generation, aviation, and oil and gas applications.

This research topic applies to all gas turbine applications with turbine inlet temperatures higher than about 1,500°C/2,800°F in markets with NO_x restrictions. Depending on the geographic region of the world, these NO_x restrictions affect power generation, aviation, and oil and gas applications differently. For example, power generation is typically subject to the most stringent regulations. NO_x limits tend to be the most stringent in the United States and the European Union, and less stringent in other regions.

Research Topic 1.3
Operational and Performance Limits on Combustors

Summary

Combustion systems influence the operational limits of gas turbines and combined cycle plants overall because of key limiting phenomena such as blowoff, combustion instability, flashback, and autoignition. In pressure gain systems, they also strongly influence limits on thermal efficiency and performance. These are all unsteady or transient phenomena that involve flame extinction, flame stability, and flame dynamics in unsteady flows, and hydrodynamic stability of reacting flows. These phenomena have been heavily researched in the gas turbine combustion community historically, as they have caused major field problems. Nonetheless, the ability to predict these phenomena is still sufficiently immature that unexpected behaviors in the field or during tests still routinely occur. Moreover, addressing these phenomena is critically important, as associated operational limits strongly influence the overall operational limits and life of the entire gas turbine. Similarly, PGC systems, such as rotating detonation engines, have been demonstrated as a workable concept in numerous facilities, but the performance of these systems generally falls well below theoretical.⁶

Background

Experience over recent decades has shown that key limiting phenomena have surfaced in the field in unexpected ways, causing damage, delays, and restricting the range of fuels that a gas turbine can use. Because of the need to balance emissions, turndown, fuel flexibility, and operational limits, difficult trade-offs have been made in the design stage and surprises routinely occur. For example, combustion instability is one of the highest risks in development of new gas turbines for power generation and aviation, because combustion instability often does not present itself until late in the development process or even after delivery to a customer. Improving the ability to predict the occurrence of combustion instabilities would reduce overall gas turbine development risk. Blowoff is a particular safety concern for aviation, so minimizing blowoff risk opens the design space for aircraft gas turbine designers.

___________________

⁶ While PGC systems can theoretically achieve thermal efficiencies that are superior to Brayton cycles, demonstrated PGC system efficiencies have not achieved this level of performance for several reasons. In particular, wave strengths are lower than Chapman–Jouget values, and fuel/oxidizer injection systems have high pressure drops.

Benefit

This research topic could accelerate ongoing research by enabling expanded operational envelopes in terms of turndown, transient response, and operating conditions. In particular, the ability to develop combustion paradigms that are fundamentally less prone to key limiting phenomena or sensitive to small changes in operating conditions, designs, or fuel composition would dramatically reduce risk and cost.

This research topic could advance relevant technology from TRL 2 to TRL 7 by 2030. These topics are closely coupled to understanding the fundamental combustion properties outlined in Box 3.1. In many instances, such as the occurrence of combustion instabilities, a relatively small uncertainty in something like flame position

(that, in turn, is influenced by the turbulent flame speed) means that stability boundaries cannot be accurately predicted.

This research topic has high technical risk because current design paradigms have inherent trade-offs between emissions, turndown, and operational limits. Better understanding of fundamental combustion properties will enable better predictive capabilities to prevent surprises and enable designers to identify optimal conditions. Better yet would be new combustion approaches that are less sensitive to uncertainties and variations in operating conditions and fuel composition.

This research topic applies to power generation, aviation, and oil and gas turbine applications because the combustor inherently limits operational performance in all three applications.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.1. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrows (with an arrowhead at each end) show where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the combustion research area are not shown.

RESEARCH AREA 2: STRUCTURAL MATERIALS AND COATINGS

Overview

From the start of development and introduction of the gas turbine into use, the development and introduction of advanced materials has been critical. During World War II, early jet engines were severely limited in performance due to the limitation of hot gas path materials. The English Gloster aircraft powered by the Whittle jet engine that first achieved flight in 1941 had the advantage of using nickel alloy turbine blades that could be designed without cooling. On the other hand, the famous German Messerschmitt Me262 powered by the Junkers Jumo jet engine was limited to using steel alloy blades due to the lack of access to nickel alloys. This necessitated the complex and costly design of hollow blades that were internally cooled.

To allow higher hot gas path temperature and increased thermal efficiency of the gas turbine, significant progress has been made in the development of high-temperature structural materials, coatings, and manufacturing processes. Early gas turbine engines had a thermal efficiency of approximately 30 percent, compared to current gas turbine engines that reach approximately 50 to 55 percent. To attain current goals for thermal efficiency and fuel burn, advanced high-temperature technologies are required to enable higher temperatures in the hot gas path.

This research area includes three research topics:

• CMC Performance and Affordability
• Physics-Based Lifing Models⁷
• Advanced Alloy Technologies

The scope of all three research topics includes both static and rotating components. Advances in these and related topics would enable substantial improvements in gas turbine efficiency and durability. For example, one related technology of particular importance is the development of environmental barrier coatings.⁸

Research Topic 2.1
CMC Performance and Affordability

Summary

Widespread utilization of components that incorporate very high-temperature CMCs (that are suitable for surface temperatures up to and exceeding 1,500°C/2,700°F) is hindered by the very high cost of fibers and the high cost of current composite manufacturing processes. Research to address this issue includes optimized CMC

___________________

⁷ A lifing model is essentially a lifetime prediction model that predicts either total time to failure or the number of operational cycles to failure for a particular component under a presumed set of operating or environmental conditions, operational modes, loading or power level, and so on.

⁸ An environmental barrier coating is a coating system that is applied to a CMC to provide protection from oxidation, moisture-induced degradation, and calcium, magnesium, and alumina silicate (CMAS)-induced degradation. It also provides thermal insulation. The goal would be to develop environmental barrier coatings that are suitable for the same conditions as the underlying CMCs. (See research topic 2.1, CMC Performance and Affordability.)

manufacturing processes, process modeling, and understanding to effectively increase production rates. CMC components are manufactured using ceramic fibers in a ceramic matrix. The composite requires a high-strength, refractory fiber that can sustain thermal excursions. High-strength, creep-resistant, oxygen-free SiC fibers provide these properties. SiC fiber is manufactured by a limited number of suppliers, each with unique processing and chemistry. This drives the need to optimize and balance chemical compositions, material properties, and cost.

CMC development and maturation investment activities have been ongoing for more than 30 years, with General Electric’s private investment exceeding $1.5 billion in the past decade. Early development and investment were supported by the Department of Energy (DOE), Department of Defense (DoD), and NASA. There is renewed interest in this field now that significant investment has been made in fiber manufacturing capacity. This increased capacity expands the ability to support broader applications of CMCs. Given this situation, it is important to address the limitations of current processes for manufacturing SiC fibers and take advantage of higher quality fibers that provide greater capabilities at lower costs.⁹ These advanced higher quality CMC fibers would enable higher temperatures in hot gas path, particularly for combustor liners and for rotating and static airfoils within the turbine module. Higher operating temperatures would reduce the amount of cooling air that must be diverted from thrust and enable fuel to be combusted more completely. The expected benefits include a reduction in fuel burn and CO₂ of up to 2 percent,¹⁰ improvements in gas turbine efficiency, improved thrust (for aviation applications), and a reduction in the cost of electricity (for power generation applications).¹¹

Background

Advanced materials such as CMCs have played a key role in advancing gas turbine performance and efficiency.

SiC fiber undergoes two coating steps to apply thin coatings on each filament in the tow bundle using chemical vapor deposition processes. The coated fiber is then drum-wound to create a unidirectional tape material. Research is needed to create an improved SiC fiber that possesses improved strength retention, modulus, and creep resistance properties with the key characteristics: stoichiometric carbon-to-silicon ratio, protected grain boundaries, low residual oxygen content, and no foreign phases.

___________________

⁹ J. Steibel, 2019, Ceramic matric composites taking flight at GE Aviation, American Ceramic Society Bulletin 98:30-33, https://ceramics.org/wp-content/uploads/2019/03/April-2019_Feature.pdf.

¹⁰ Ibid.

¹¹ As noted in the preceding section on the combustion research area, higher turbine inlet temperatures tend to increase the production of NO_x. Research to address this issue is described in the section “Research Topic 1.2, Combustion Concepts to Reduce Harmful Emissions at Elevated Temperatures and Pressures.”

¹² J. Steibel, 2019, Ceramic matric composites taking flight at GE Aviation, American Ceramic Society Bulletin 98:30-33, https://ceramics.org/wp-content/uploads/2019/03/April-2019_Feature.pdf.

Benefit

As stated in the summary above, the benefits of higher quality CMCs include improved gas turbine efficiency as a result of higher operating temperatures and reduced cooling air requirements, as well as more efficient combustion, thereby reducing CO₂ emissions.¹³ Reducing the costs of higher quality CMCs will enable them to be adopted earlier across a broader range applications.

This research topic could accelerate ongoing research as follows:

• Develop an optimum SiC fiber using cost-effective polymer processing routes that produce fiber with excellent strength retention, modulus, and creep-resistance properties with the key characteristics: stoichiometric ratio or carbon to silicon, protected grain boundaries, low residual oxygen content, and no foreign phases.
• Target productivity improvements related to in-process controls, statistical monitoring, digitization, and automation.
• Provide building blocks to enable an additional 170°C/300°F of temperature capability beyond current CMC technology, thus enabling an increase of 450°C/800°F over today’s nickel-based materials.

Recent investments in SiC fiber manufacturing capacity (including U.S. Government Title III¹⁴) within the United States allows domestic control over productivity improvements and provides facilities to focus process development and demonstrate product improvements. This research would enable a unified processing approach to advance from TRL 3 to TRL 6 or more by 2030 because the proposed program would highly leverage proven mature processing technology for SiC fiber.

This research topic has medium technical risk because of the limited engineering and manufacturing resources with appropriate materials expertise to support high-quality SiC production. The unique processing systems used by current SiC fiber manufacturers introduce variation in the characteristics of the SiC fiber. A large percentage of the current SiC fiber manufacturing capacity is committed to meeting production volumes supporting current applications, which limits the availability of facilities for assessing potential improvements to manufacturing processes without interrupting highly valued production time. The lack of a dedicated research facility that includes a pilot production line inhibits process optimization studies of key fiber process variables.

___________________

¹³ G. Gardiner, 2017, The next generation of ceramic matrix composites, CompositesWorld, April, http://bit.ly/2qvpcVV.

¹⁴ U.S. Government Title III—Expansion of Productive Capacity and Supply: Authorizes appropriate incentives to create, expand, or preserve domestic industrial manufacturing capabilities for industrial resources, technologies, and materials needed to meet national security requirements (includes homeland security).

This research topic applies to both power generation and aviation gas turbine applications because increasing the maximum operating temperature broadens the use of CMCs in hot gas path components. CMCs offer the potential to increase current gas path temperatures by up to 450°C/800°F. This applies to both aviation and power generation gas turbine hot gas path components of combustor liners, rotating and static airfoils and shrouds in the high-pressure turbine, and with potential use in the low-pressure turbine (and, perhaps, oil and gas applications), if costs are reduced sufficiently.

Research Topic 2.2
Physics-Based Lifing Models

Summary

This research topic would develop models of environmental degradation of materials in the hot section of the main gas path. These models will be particularly useful to guide the development of advanced materials, manufacturing technologies, and digital twin infrastructure.¹⁵

Models to be developed will address the following damage modes:

• Hot corrosion and oxidation of metallic structural materials and coatings;
• Molten salt-induced high-temperature stress corrosion cracking;
• Hot corrosion, oxidation, moisture-induced degradation, calcium-magnesium-alumino-silicate (CMAS)induced degradation and erosion of thermal and environmental barrier coatings; and
• Long-term retention of CMC mechanical properties at extreme turbine operating temperatures.

Physics-based lifing models will be validated through laboratory testing that reliably and accurately predicts field performance.

This research topic is critical for attaining and maintaining the thermal efficiency goals of future advanced gas turbines over their service life. Relevant research is currently lacking. The environmental degradation modes and the models relating to these modes are unique to the main gas path of gas turbines for all three applications.

Background

The hot section components of a gas turbine experience extreme conditions, including exposure to high stresses and corrosive environments and materials, while operating at a significant fraction of their melting point. In order to maximize thermal efficiency, these components are pushed to extreme temperatures that challenge state-of-the-art materials, including nickel-based single-crystal superalloys and advanced thermal barrier coating systems.¹⁶ This presents numerous challenges to gas turbine manufacturers, as the increased operating temperatures that drive higher thermal efficiency also accelerate environmental damage. Further, these damage modes are diverse in rate and mechanism, and include high-temperature oxidation, hot corrosion (deposit-induced high-temperature corrosion, commonly caused by sulfate-rich deposits), particle erosion, spallation of thermal barrier coatings, and CMAS degradation of thermal barrier coatings systems.

___________________

¹⁵ A digital twin is a virtual representation of an operational gas turbine. For more information on manufacturing technologies and digital twins, see the sections “Research Area 3: Additive Manufacturing for Gas Turbines” and “Research Area 9: Digital Twins and Their Supporting Infrastructure,” below.

¹⁶ The term “thermal barrier coating” refers to the coating system, which typically consists of metallic and ceramic coatings applied to a structural alloy. The metallic coating provides oxidation protection, and it bonds the ceramic coating, which provides thermal insulation, to the underlying structure.

The increase in life-cycle cost associated with these environmental damage modes can be measured in millions of dollars annually for a large power generation gas turbine. In an aircraft gas turbine engine, failure of a high-pressure turbine blade can result in an unscheduled engine removal, potential disruption to service for the airline, and, in the worst case, flight safety risk. Analogous costs are incurred upon failure of a gas turbine in service for power generation or oil and gas applications. Figure 3.3 shows an example of a high-pressure turbine blade in an industrial gas turbine that has failed during service as a result of high-temperature corrosion.

The ability to reliably predict which damage modes will be active for a given application and the rate at which they will progress will enable (1) efficient development and optimization of advanced materials that are resistant to these damage modes, and (2) implementation of an advanced digital twin infrastructure.¹⁷

As discussed in the section “Research Area 9: Digital Twins and Their Supporting Infrastructure,” a digital twin infrastructure supports digital twins by storing and retrieving very large quantities of data and processing the data using validated lifing models of turbine operation and related processes to develop and maintain the digital twin. For example, the infrastructure requires a combination of sophisticated sensor technology and location-specific, integrated, probabilistic lifing models for the various degradation modes mentioned above, among others. The models will need to account for all aspects of gas turbine operations, such as mission profile, ambient temperature, environmental contaminants, and cycle time. Input data that need to be captured include all relevant gas stream temperatures; thermal gradients; gas composition in the combustion and secondary flow paths; and particulate matter ingestion rate, size distribution, composition, and deposition rate on various surfaces.

The ability to generate physics-based predictions is a function of the availability of all relevant inputs, many of which are currently lacking. These predictions are of primary importance for the development of advanced materials and for implementation of advanced digital twins. It is expected that physics-based lifing models will

___________________

¹⁷ See the section “Research Area 9: Digital Twins and Their Supporting Infrastructure,” below.

be the most crucial but also the most challenging aspect of developing advanced digital twins and infrastructure. To be effective, physics-based models will need to represent the complex environment distress mechanisms that occur in service. These degradation modes can occur from the combined effects of oxidation, hot corrosion, moisture, and particulate deposits. Computational means to predict the distress observed in service do not exist, and so this information must be obtained by experimentation, testing, and characterization of hardware returning from service. Semi-empirical and machine-learning enhanced models may supplement development and implementation. In addition, Research Topic 5.2 Coordinated Experimental Research (see below) is focused on experimental research to validate numerical simulations, and the scope of that research encompasses some phenomena that are relevant to the development of physics-based lifing models.

The challenges associated with developing advanced lifing models for conventional materials are especially difficult to overcome for materials such as CMCs. Operation of CMCs and their associated environmental barrier coatings at temperatures up to the goal temperatures of 1,500°C/2,700°F and beyond will almost certainly aggravate hot corrosion and the other damage modes listed above.¹⁸ Life shortfalls could reduce gas turbine service intervals for components with CMCs by orders of magnitude relative to their nickel-based counterparts. Additional degradation modes will very likely arise as CMC components accumulate more time in service, creating currently unforeseen durability challenges. A concerted effort is required to understand the mechanistic aspects of these degradation modes and develop quantitative descriptions of them for incorporation into lifing models.

Benefit

This research topic could accelerate ongoing research by (1) enabling the design of future damage-resistant materials and (2) reducing life-cycle cost by giving operators information on precisely when gas turbines need to be taken out of service for maintenance. Achieving the first goal would improve turbine thermal efficiency by enabling operation at peak materials temperature limits. Achieving the second goal would improve life-cycle cost by potentially eliminating unscheduled gas turbine downtime or extending gas turbine operating time. In addition, component performance feedback from an advanced digital twin infrastructure provides a path for continuous improvement to the physics-based life models. Useful models that aid in the prediction of environmental distress in the gas turbine are nonexistent.

This research topic could advance relevant technology from TRL 3 to TRL 6 by 2030. The need to improve thermal efficiency in gas turbines will drive increased operating temperatures and exacerbate the impact of environmental distress. Developing physics-based models and a digital twin infrastructure will reduce the funding level and development time needed to bring to market the advances that could result from this research topic. Expertise and capability are in place in government laboratories, universities, and industry to make the projected advances. Given the current situation of gas turbine operation being limited by environmental degradation and the understanding that environmental distress will become more life limiting with projected increases in main gas path temperatures, this research will become more important.

This research topic has medium technical risk because advances have been made in the basic understanding of the physics-based mechanisms of many of the modes of environmental degradation.¹⁹ Understanding the mechanisms of environmental distress is the first step in establishing physics-based models. One risk is the appearance of unknown degradation modes at higher turbine operating temperatures.

This research topic would significantly reduce life-cycle costs and improve thermal efficiency for power generation, aviation, and oil and gas applications.

___________________

¹⁸ N.P. Padture, 2019, Environmental degradation of high-temperature protective coatings for ceramic-matrix composites in gas-turbine engines,” npj Materials Degradation 3(11), doi:10.1038/s41529-019-0075-4.

¹⁹ Many chemical reactions occur in the gas path, and the rate at which those reactions take place typically increases as temperatures rise. Mechanisms of particular interest at higher temperatures include oxidation and the formation of new phases of materials in the gas stream that can accelerate degradation of gas turbine coatings and structural materials.

Research Topic 2.3
Advanced Alloy Technologies

Summary

The high temperatures required in the main gas path to meet gas turbine efficiency goals will require the use of CMCs as well as metallic components that can be used at temperatures exceeding those of current designs. Advanced, high-temperature metallic components enable the use of CMCs. Current research related to metallic components that could be used in the main gas path and allow the use of CMCs to their full potential is very limited.

This research topic would develop new alloy concepts such as high-entropy alloys, higher-temperature titanium alloys, cobalt-based superalloys, and refractory alloys.²⁰ Variants of these classes of alloys would need to be developed to enable their use in additive manufacturing. Also, new advanced metallic component designs including hybrid turbine disks²¹ consisting of multimicrostructures and single-crystal rim/multicrystalline bore concepts will be needed. Disks with single-crystal rims and a polycrystalline bore are probably not viable for large power generation turbines. Hybrid disks with steel bores and superalloy rims may be possible, but it would be difficult to accommodate the different coefficients of thermal expansion of the steel and superalloys. Combining new physics-based tools with emerging tools based on artificial intelligence (AI) and machine learning would enable alloys such as those mentioned above to be examined in much more detail, at lower cost, and in less time that has been possible using traditional development processes and tools.^22,23 Application of this technology to very high temperature materials outside of gas turbines is limited.

Background

Nickel-based superalloys have been used in gas turbines for more than 50 years. For much of this time, the aerospace industry has been able to rely on continued development of nickel alloy classes that are produced by directional solidification (e.g., CMSX-4, René N5, and PWA 1484) or by wrought processing paths (e.g., Waspaloy, IN718, and U720). These classes of alloys have reached their operational temperature limits with current gas turbine designs. As new requirements have been established to increase efficiency, decrease weight, and reduce emissions, it has become necessary to increase main gas path temperature. The increase in temperature drives the need for alloy technologies that can operate at higher temperatures and still meet design life requirements.

As the gas turbine industry utilizes CMCs for specific components to meet the higher operating temperature requirements of gas turbines with improved efficiency, it will become increasingly important to consider metallic gas turbine components that will also be required to operate at higher temperatures. In order to take full advantage of the higher temperature capability of CMCs, the temperature of the entire hot gas path would increase. Most of the components in the hot gas path are currently fabricated from alloys. Advanced high-temperature alloy technology development has generally been focused on (1) directional solidification casting technology, (2) nickel-based superalloy chemical composition development for turbine blades and vanes, and (3) powder metallurgy-based superalloys for disk (rotor) applications. Research and development to increase temperature capability has reached diminishing returns. Additive manufacturing and the ability to manufacture components with advanced cooling

___________________

²⁰ High-entropy alloys are typically made with five or more elements and consist of a solid solution. Refractory alloys retain favorable mechanical properties at very high temperatures. Examples include tungsten, niobium, and molybdenum.

²¹ The gas turbine disk is also referred to as a rotor.

²² R. Ramprasad, R. Batra, G. Pilania, A. Mannodi-Kanakkithodi, and C. Kim, 2017, Machine learning in materials informatics: Recent applications and prospects, npj Computational Materials 3(54), doi:10.1038/s41524-017-0056-5.

²³ There are some common elements to the development of new tools relevant to this research topic and those addressed in the section “Research Area 3: Additive Manufacturing for Gas Turbines,” below.

configurations will require the development of suitable new alloys. Issues with cracking, hot tearing, and the unique grain structures associated with the emerging suite of processes will need to be understood and addressed. Future development required to allow CMC components to reach their full potential in terms of temperature capability will need to focus on increasing the temperature capability of a large number of the metallic components in the main gas path for which CMCs are unsuitable. These components include blades and vanes, rotors, cases, shafts, seals, and bearing materials.

The most advanced high-temperature materials, such as single-crystal airfoils and advanced corrosion and thermal barrier coatings, are used in the high-pressure turbine and combustor, which are the highest temperature sections of the gas turbine. As the temperature of the main gas path of the gas turbine increases, advanced high-temperature material technologies will be needed in additional areas of the gas turbine. The application of coatings to protect components such as cases and disks from corrosion will need to function for thousands of hours of operation without reducing the base alloy mechanical properties.

Gas turbine components have been largely limited to the production of components comprising a single alloy composition and single manufacturing process. A hybrid component is manufactured from materials that have dissimilar properties. The ability to use hybrid structures increases design options and flexibility by providing a component with more optimum, location-specific properties. Hybrid turbine disks, for example, are being developed that comprise alloys that have distinctly different alloy compositions and are produced by distinctly different manufacturing processes than each other. The hybrid turbine disk shown in Figure 3.4 illustrates a turbine disk bore region that consists of a polycrystalline alloy produced by forging one alloy that has itself

been produced by a powder metallurgy process. The rim of the turbine disk consists of an alloy that is produced by a single-crystal casting process. The separately manufactured sections of the turbine disk are then bound to each other using a joining process such as inertia bonding. The polycrystalline alloy bore would provide low-temperature burst strength where that property is required, and the single-crystal alloy rim would provide high-temperature creep strength where that property is required.

Some alloy technologies have the potential to exceed the temperature capability of the latest generation of nickel-based superalloys. Refractory alloys have the potential to provide both superior high-temperature mechanical properties and ductility. Niobium-based and molybdenum-based refractory alloys strengthened with intermetallic compounds have been developed. Thus far, however, high-temperature oxidation and significant processing challenges have prevented their use in gas turbines. New refractory-based high-entropy alloys, which contain equiatomic mixtures of multiple refractory elements, demonstrate unusually high strength at temperatures above 1,100°C/2,000°F, but research on this class of materials is still at TRL 1. New manufacturing technologies, such as additive manufacturing, may be used to fabricate unique microstructures that address previously identified shortfalls in properties for refractory alloys. Because cobalt-based alloys have such a high melting point, they also have the potential to be used at temperatures exceeding those of current nickel-based alloys.

Improving the capabilities of lower temperature alloy systems, such as those based on titanium, magnesium, and aluminum, would also be beneficial for fan and compressor components. These materials offer strength-to-density ratios at high temperatures that exceed the capability of current state-of-the-art high-temperature composite materials. Development of processing models and integrated computational materials engineering tools²⁴ would further enable the use of materials in these classes while improving product yield and lowering purchase costs.

Current approaches to material development are largely experiential driven. The ability to establish new materials computationally using first principles calculations, physics-based models, and AI methodologies would allow engineers to design and mature materials that are optimized for specific applications much faster than is currently possible.

Benefit

This research topic could accelerate ongoing research by providing the technology that is required to operate gas turbines at main gas path temperatures required to meet thermal efficiency goals. In addition, the successful development and utilization of advanced, high-temperature alloy technologies will be essential in meeting life-cycle cost goals by increasing gas turbine service life. Current gas turbine designs are limited by the maximum temperature at which available materials can be used and still meet life requirements.

This research topic comprises many individual technologies related to the development of high-temperature alloy technologies. Expertise and capability are in place in government laboratories, universities, and industry to make substantial advances by 2030. Elements of this research topic could advance some relevant technology from TRL 1 to TRL 9 by 2030. There is low technical risk in advancing the technologies applicable to hybrid disks to progress from the current TRL 3 to TRL 9 by 2030. The basic manufacturing processes required to make this technology successful are known. For technologies related to high-entropy alloys, much fundamental work needs to be done, and it is more realistic to predict that this technology could advance from TRL 1 to TRL 4 by 2030. Work in this area faces medium technical risk.

This research topic applies to gas turbine applications for power generation, aviation, and the oil and gas industry because of the similarity in the materials used and turbine component designs.

___________________

²⁴ Integrated computational materials engineering is the integration of materials information, captured in computational tools, with engineering product performance analysis and manufacturing-process simulation. National Research Council, 2008, Integrated Computational Materials Engineering: A Transformational Discipline for Improved Competitiveness and National Security, The National Academies Press, Washington, D.C., p. 9, doi:10.17226/12199.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.5. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrows (with an arrowhead at each end) show where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the structural materials and coatings research area are not shown.

RESEARCH AREA 3: ADDITIVE MANUFACTURING FOR GAS TURBINES

Overview

As noted in Chapter 1, additive manufacturing is a global technology trend that will benefit a wide array of industrial applications. This research area will develop turbine-specific design and manufacturing approaches for three-dimensional (3D)-printed turbine components. Technology opportunities exist in integrated design of novel, cooled components, new high-temperature alloys, and morphological control of microstructure for tailored properties. These advances are quite challenging because they must be implemented and ultimately qualified in the extreme conditions within a gas turbine. Key benefits for gas turbines are reduced weight, reduced part count, access to new design spaces, and reduced development time.

Additive manufacturing has evolved dramatically over the past 30 years, and the rate of change continues to increase. Additive manufacturing emerged commercially in 1987 with stereolithography from 3D Systems, a process that uses a laser to solidify thin layers of a liquid polymer that is sensitive to ultraviolet light. The SLA-1 became the first commercially available additive manufacturing system in the world, and in 1988 Asahi Denka Kogyo, Ltd., introduced the first epoxy resin for stereolithography machines. Selective laser sintering became available in 1992, using heat from a laser to fuse powder materials. In 1997, AeroMet developed a process called laser additive manufacturing that used a high-power laser to weld powdered titanium alloys, followed quickly by Optomec, which commercialized its laser powder forming system for fabricating metal parts based on technology developed at Sandia National Labs. Extrude Hone (now ExOne) introduced another additive manufacturing process in 1999 with a system based on 3D inkjet printing technology from the Massachusetts Institute of Technology to build metal parts. In 2001, Generis GmbH commercialized a system that used an inkjet printing technique to fuse sand to produce sand cores and molds for metal castings. By 2006, direct metal laser melting had evolved to include stainless steel and cobalt-chrome materials; and Arcam was distributing electron beam melting systems for metal powders. Concept Laser soon followed with the M2 system for processing reactive materials such as aluminum and titanium for direct metal laser melting.²⁵

Standardization took a big step forward in 2009 when the ASTM International Committee F42 on Additive Manufacturing Technologies published standard terminology for the industry. In 2011, a cooperative agreement was announced between the ASTM International Committee F42 and the ISO Technical Committee 261 on Additive Manufacturing to reduce duplication of effort. In 2012, General Electric (GE) Aviation purchased Morris Technologies and Rapid Quality Manufacturing, which were pioneers in additive manufacturing for aerospace applications. Key acquisitions followed in 2016 of Arcam AB (Sweden) and Concept Laser (Germany), which are both machine manufacturers.

Direct metals processing technologies have garnered significant interest and growth based on the possibility of novel designs, combined with mechanical properties that are nearly equivalent to wrought alloys. The adoption of metal-based additive manufacturing has continued to accelerate, with the biomedical and aerospace communities leading the way. Feature resolution and process controls have continually improved, with applications touching on a broad range of industries, including dental and medical, industrial, aerospace, jewelry, and even sand-casting molds.

Designers of industrial products generally leverage a full range of material and manufacturing options at their disposal for the development of cost-effective components with optimal performance characteristics. Additive manufacturing presents pathways to previously inaccessible design spaces, motivating the development of a new suite of responsive design tools that couple with manufacturing simulation in order to maximize the advantages of additive manufacturing. The digital artifacts of this new design-for-manufacturing process²⁶ can be captured in a digital thread infrastructure,²⁷ which enables information sharing across the digital infrastructure to improve the performance of the product design and manufacturing process.²⁸

While there are many potential avenues for additive manufacturing, the design of new high-temperature materials compatible with these layer-by-layer manufacturing approaches promises improved system performance enabled by combining new materials with innovative component designs. Controlling local grain structure promises further optimization.

Advances in integrated sensing, autonomous analysis of sensor data, and process correction will be required to enable high-quality, high-manufacturing yields, and rapid feedback to the design process. Challenges for additive manufacturing of particular relevance to gas turbines include the ability to successfully manage potentially damaging process-induced phenomena such as increased distortion and cracking, print defects (e.g., lack of fusion

___________________

²⁵ T. Wohlers and T. Gornet, 2013, “History of Additive Manufacturing,” Wohlers Report,http://wohlersassociates.com/history2013.pdf.

²⁶ A design-for-manufacturing process is an integrated engineering process in which manufacturing capabilities and constraints are considered during component and system design in order to optimize performance while minimizing manufacturing risk and cost.

²⁷ See the section “Research Area 9: Digital Twins and Their Supporting Infrastructure,” below, for more information.

²⁸ M. Helu and T. Hedberg, Jr., 2016, “Data Infrastructure and Management for the Digital Thread in Manufacturing,” IMTS 2016 Conference (#IMTS17), September 13, 2016, Chicago, Ill., http://bit.ly/2Cj9FLz.

and gas porosity), and lower mechanical performance due to varied bulk and surface structure quality. This will require mastery of the complex physics of these printing processes, including high and rapidly varying temperature gradients, complex melt pool fluid flow phenomena, widely varying solidification morphologies, and residual stresses during the cooling process.

This research area includes three research topics:

• Integrated Design and Additive Manufacturing
• Additive Manufacturing of High-Temperature Structural Materials
• Integration of Sensors, Machine Learning, and Process Analytics

Research Topic 3.1
Integrated Design and Additive Manufacturing

Summary

Expanding additive design aids and enhancing design practices will enable turbine-specific benefits in terms of reduced life-cycle costs by adapting general advances in the state of the art of additive manufacturing to issues specific to gas turbines, in part doing the following:

• Optimizing the use of cooling air through the use of creative geometries that (1) are enabled by additive manufacturing and (2) would improve heat extraction and improve the control of clearances between parts, thereby improving gas turbine efficiency.
• Enabling the design of new, higher performance, high-temperature materials as well as graded chemistries and microstructures to align with key component stress conditions within gas turbine components to improve component durability.

Background

Current design methodologies and design practices for product development have been optimized based on conventional manufacturing processes using primarily subtractive manufacturing techniques to post-process wrought or cast components. These methodologies do not allow the creative geometries and design options that have been enabled by additive manufacturing and that have begun to revolutionize much of the gas turbine design and manufacturing paradigm. The process of building parts incrementally, layer by layer, reduces costs and weight, enables innovative designs, and challenges the order and speed of the traditional hardware development cycle.²⁹

Gas turbines are already very complex machines, but their performance could in many cases be improved by the ability to incorporate parts with even higher levels of complexity, to the point that the parts are impossible to manufacture using conventional processes. Additive manufacturing also offers a unique ability to substantially reduce the costs and cycle time of producing complex development hardware by enabling prototype hardware designs to be manufactured, tested, revised, and remanufactured much more quickly and inexpensively than is currently possible.³⁰

___________________

²⁹ NASA Marshall Space Flight Center, 2017, MSFC Technical Standard (MSFC-STD-3716), Standard for Additively Manufactured Spaceflight Hardware by Laser Powder Bed Fusion in Metals, https://standards.nasa.gov/standard/msfc/msfc-std-3716.

³⁰ Ibid.

Benefit

This research topic could accelerate ongoing research in gas turbines by greatly enhancing the ability to design gas turbine components to improve their performance, affordability, and manufacturability. This would reduce uncertainty in going from design to manufacture, leading to high-fidelity, location-specific designs for higher performance components with less rework. Additionally, advances in model-based engineering tools are essential to take full advantage of the design benefits offered by additive manufacturing. Specific benefits will include the following:

• Complex geometries. Advanced design aids would enable the development of novel component shapes and geometries based on a systems approach to design, independent of prior industrial constraints and limitations imposed by post-processing utilizing historical subtractive manufacturing techniques. This is of particular importance to gas turbine design because it provides opportunities to lower part count, reduce weight, and decrease cooling air requirements, thus increasing efficiency.
• Distortion control. Advanced design tools that anticipate distortion effects from rapid solidification and heat extraction could be used to guide the sizing and placement of support structures during component build. These tools would also provide guidance on geometry compensation, which could significantly reduce the number of build iterations required to demonstrate dimensional control. This is of particular importance to gas turbine design due to extremely stringent dimensional tolerance requirements.
• Tailored microstructures within a single design. An ability to implement multiple materials or grain structures would allow for location-specific designs, thereby reducing weight and providing for an optimal distribution of stresses for improved durability and lifetime.
• Reduced risk of using new materials for an existing application or using existing materials for new applications. Integrating digital definitions of materials and processes with digital design protocols will allow for more efficient exploration of design spaces and early validation of component performance.
• Digital capability. Optimization of CAD build files will improve productivity in loading the files to additive manufacturing machines, decrease delays during build as information is managed for the layer-by-layer build method, manage residual stresses in components as they are manufactured, enable manufacturing at remote locations and the capture nondestructive evaluation results due to improved transportability of the files, and provide a digital “fingerprint” of each built component.

This research topic could advance relevant technology from TRL 4 to TRL 6 by 2030 by improving component durability, increasing turbine efficiency, and reducing life-cycle costs.

The research topic has medium technical risk primarily because of the broad spectrum of additive manufacturing processes and the need to control the process on a layer-by-layer basis. Each additive process poses unique challenges and restrictions to the design community. Rapid solidification for laser powder beds, sintered powder removal, and build zone heat control for electron beam processes, and dimensional control during sintering and densification of binder jet additive processing all pose unique challenges for designers of additive manufacturing parts. As a result, design aids and design practices need to be customized as a function of the additive process type. Regardless, the payoff is significant for each process that is targeted for use in high-temperature turbine applications.

Advances in model integration across disciplines are essential to more rapid, accurate, and complex design and manufacturing. These problems are challenging, but engineering has been on this trajectory for several years already.

This research topic applies pervasively to gas turbines for power generation, aviation, and oil and gas applications because of the similarity in turbine component designs and the processes used to manufacture components.

Research Topic 3.2
Additive Manufacturing of High-Temperature Structural Materials

Summary

The gas turbine industry will drive the development of new high-temperature structural materials that can be used with additive manufacturing; and advances in additive manufacturing equipment will be required to process what is envisaged to be more refractory materials than are currently used today. Greater coupling of computational and characterization tools is required to quickly identify new material compositions designed for additive manufacturing. These tools would support modeling of key phenomena and determine which combinations of processing and composition modifications can mitigate the driving forces for crack formation. Multiple conventional energy sources as well as new types of sources, such as femtosecond lasers, will be required to deliver energy in an ever more managed fashion to speed build rates and control distortion. To ensure affordability of components produced with these new material compositions and additive techniques, this research topic would take advantage of the global technology trend in additive manufacturing (see Chapter 1), particularly with regard to the development of new, lower cost methods for powder production.

The largest gains in the performance of gas turbines are likely to be achieved in the hottest sections of the gas turbine. One approach for raising the operating temperature limits and increasing the durability of parts is through use of materials processed to a directionally solidified or single-crystal form. The development of new compositions, additive manufacturing methods, and models that enable the growth and repair of directionally solidified or single-crystal components would be game changing for both design and manufacturing.

Rapid qualification and certification methodologies will also be needed to accommodate the wide range of additive manufacturing techniques, structural material compositions, and gas turbine applications in order to cost-effectively produce components with higher temperature limits and durability.

Background

The harsh operating environments in gas turbines require a unique set of highly engineered properties. Limited efforts are already under way to expand the compositional range of structural materials specifically designed for additive manufacturing that can meet these harsh conditions.³¹ The relatively few structural material compositions being employed in additively manufactured applications today, such as Alloy 718 and Ti-6Al-4V, were originally designed to be processed by conventional manufacturing means. The highly tailorable nature of the local processing conditions available in additive manufacturing will enable custom structural material compositions for additive manufacturing. It will be possible for compositions to be modified to enable higher yield through lower incidence of cracking phenomena and deleterious residual stress. Thus, new strategies and methodologies for applying and managing energy will also be needed to accommodate new compositions. Additionally, new structural material compositions will benefit from advances in powder processing techniques aimed at producing higher temperature and higher quality powder in large quantities. The stringent performance requirements of gas turbines have consistently driven the design of structural materials to give at least equal weight to the precise tailoring of microstructures in order to maximize performance. While novel processing of directionally solidified and single-crystal materials using advanced casting techniques is now an industry standard, precise control of microstructures in additively manufactured components remains a challenge. A new combined test and simulation-based analysis approach to part qualification has been elusive, but it is essential to the rapid and cost-effective employment of high-temperature materials for additive manufacturing.

Benefit

This research topic could accelerate ongoing research in this area by focusing on the development of new additive manufacturing equipment capabilities and configurations. These new capabilities would enable the processing of new, higher temperature structural material compositions as well as novel methods for controlling microstructures. Additionally, these capabilities could enable the development of additive processes

___________________

³¹ Information on advanced high-temperature alloys and the state of their development is highly valued intellectual property, and target alloys will likely not be publicly available until process, structure, and properties are demonstrated in a gas turbine environment.

for producing columnar-grained or single-crystal components, which would have significant impacts on the industry by enabling more efficient designs for cooling air flow while providing potential paths to repair of conventionally processed single-crystal components. Accelerating the development of rapid material and process certification and qualification methods will enable much more efficient and widespread application of additive manufacturing.

This research topic could advance some elements of relevant technologies from TRL 3 to TRL 6 by 2030. Advances in additive manufacturing equipment capabilities have a high probability of meeting the requirements for producing advanced structural material compositions in a production setting. New production paths to affordable, high-quality powder are also likely to be invented and introduced, although perhaps only on a small scale by 2030. The most challenging aspect of this research topic will be developing the ability to tailor microstructures during the manufacturing process to produce columnar-grained or single-crystal parts. However, advances in risk assessment methodologies are likely to support the application of rapid qualification and certification methods.

This research topic has medium technical risk because global advances in additive manufacturing are rapidly advancing the equipment design concepts and the understanding of fundamental processes that underpin this topic.

This research topic applies to power generation, aviation, and oil and gas applications because of similarities in relevant materials and component designs.

Research Topic 3.3
Integration of Sensors, Machine Learning, and Process Analytics

Summary

The gas turbine industry already has great interest and robust research efforts under way in precision manufacturing and process controls because there are few other engineering applications of additive manufacturing that require such a high level of performance, reliability, and safety. This is especially challenging given the demanding operational environments within a gas turbine. Advances in precision manufacturing process control for gas turbine components will easily transition to other industries, resulting in higher product yields. Physics-based process models and AI systems will need to be integrated with factory operational technology for real-time, intelligent control and for instantaneous feedback on manufacturing processes. This will require in situ sensing of material states and autonomous methods for real-time defect identification and repair during manufacture. A digital record of the manufacturing process will need to be automatically placed in a digital thread infrastructure for further engineering use in the component life cycle.

Background

Additive manufacturing is an extraordinarily complex means of processing materials into a useful engineering component. Several additive manufacturing processes are being used or explored for their utility in the gas turbine industry, including powder bed fusion (using either electron beam or laser heating sources) and directed energy deposition (using blown powder or wire fed processes).³² A typical metal additive manufacturing process could require consideration of more than 150 parameters. While some of the parameters may not affect

___________________

³² A source of energy is needed to fuse the powdered material that is used as a feedstock in most additive manufacturing processes.

product outcomes, and some will be dependent on other variables, still many variables are left to be interpreted in a very complex and dynamic processing environment. With a goal of correcting defects in situ, analysis of process feedback on critical parameters will need to be immediate (i.e., within the current build layer). Complete, physics-based models for each of these methods is several years off, and even once developed, they will very likely be too complex to be integrated as part of a process control system. Overcoming this issue will require the development of physics-informed, reduced-order models augmented by machine learning that can be “trained” to monitor and autonomously take control to correct additive manufacturing processes in real time. The unique complexity of additive manufacturing will need more than a binary pass/fail criterion to establish part pedigree through the product life cycle. A digital representation of the process will therefore be essential to inform material review boards, to improve additive manufacturing processes, and to identify potential in-service issues in a timely fashion.

Benefit

This research topic could accelerate ongoing research in this area by improving manufacturing quality, increasing product yields, reducing property distributions, and enabling more rapid qualification of additive manufacturing processes. Additional benefits include the creation of a robust path to continuous process improvements while assisting scientists seeking to improve physics-based models of manufacturing processes. Having a complete digital record of the as-built end state of critical components will make it possible to rapidly identify emergent performance issues and potential root causes for failure throughout a system’s life cycle.

This research topic could advance relevant technology from TRL 4 to TRL 7 by 2030. Many individual tools are now available and are beginning to be integrated. This will be a continuously evolving technology as additional advances in physics-based models, in situ sensing, and AI are made and integrated into process control systems. At present, these process control systems are quite rudimentary but are expected to make substantial gains given the current rate of progress being made in each of the supporting technologies. As noted in Chapter 1, global technology trends will ensure continued advances in autonomous systems, physics-based models, and AI apart from research that may be conducted as part of this research topic.

This research topic has medium technical risk due to the difficulty in developing useful physics-based models and relevant in situ sensing methodologies given the sheer complexity of additive manufacturing, especially with regard to the more challenging aspects of gas turbine applications. However, advances in materials characterization techniques to inform model development and relevant global technology trends will mitigate the risks associated with the complexity.

This research topic applies to gas turbines for power generation, aviation, and oil and gas applications because of the similarity in turbine component designs and the processes used to manufacture components.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.6. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrow (with an arrowhead at each end) shows where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the research area on additive manufacturing for gas turbines are not shown.

RESEARCH AREA 4: THERMAL MANAGEMENT

Overview

Given the importance of reducing CO₂ emissions, reducing fuel usage by increasing gas turbine efficiencies continues to remain of high interest to the gas turbine industry. As discussed in Chapter 1, thermal efficiencies of gas turbines and, by extension, fuel burn, are directly related to the turbine inlet temperature. The adiabatic efficiencies of compressors and turbines are also strong drivers of thermal efficiency. Two different approaches are typically employed in order to improve gas turbine thermal efficiency: (1) increase turbine inlet temperatures while maintaining the same cooling flow requirements, or (2) maintain the same turbine inlet temperatures while decreasing the cooling flow level. Both approaches may reduce turbine life unless thermal management schemes increase overall cooling effectiveness. Increasing overall cooling effectiveness at a constant turbine inlet temperature reduces turbine component metal temperature and increases component life. Alternatively, increasing the overall cooling effectiveness while increasing turbine inlet temperature may yield constant turbine component metal temperature, and by extension, turbine component life.

Turbine inlet temperatures over time track well with the overall cooling effectiveness levels produced by advancing film cooling, as shown in Figure 3.7. Increasing cooling effectiveness reduces airfoil temperatures, thereby allowing higher turbine inlet temperatures.

Cooling concepts for the hot section of the gas turbine, which include the combustor and the turbine modules, were envisioned from the very beginning of gas turbine research and development. Blade cooling, however, did not appear in operational equipment until the 1960s. Blade cooling is achieved by extracting air from the compressor prior to entering the combustor and then routing the extracted air into components located in the combustor and turbine modules to cool those components. In the 1960s, blade cooling technology enabled turbine inlet temperatures to increase from 900°C/1,650°F to 1,000°C/1,850°F for commercial transport engines at takeoff. Subsequent development of high-temperature turbine materials, coatings, and cooling technologies have enabled turbine inlet temperatures of commercial transport engines at takeoff to increase from 1,000°C/1,850°F in the 1960s to more than 1,400°C/2,550°F in the 1990s.

Modern gas turbines benefit from decades of impactful research on high-temperature alloys, advanced coatings, and improved cooling technologies. Combining state-of-the-art nickel-based superalloys with the application of more advanced cooling technologies, such as micro-channel cooling (also referred to as double-wall airfoils), has allowed for continual increases in turbine inlet temperatures. Concurrently, increases in thermal efficiencies have

been realized through increases in cycle pressure ratios, which are between 40 and 60 in today’s gas turbines. As cycle pressure ratios increase, the temperature of the air extracted from the compressor for cooling the combustor and turbine modules also increases. Since increased compressor discharge air temperature is a limiting factor in effectively cooling combustor and turbine hardware, developing thermal management techniques that enable higher coolant temperatures are integral to meeting hot section durability requirements. State-of-the-art cooling flow requirements for components in the combustor and turbine modules reach as high as 25 percent of the air flow entering the compressor, depending on the application.

Limitations exist in the current cooling strategies, however, from manufacturing constraints. For example, the shapes of film-cooling holes, which are placed in the turbine airfoils after casting, are limited by both the laser drilling and electro-discharge machining processes. Past research has also identified some key subjects in which our fundamental physical understanding is lacking, thereby limiting the ability to advance gas turbine designs. One such subject is full conjugate heat transfer analyses, in which a single model examines heat transfer involving both fluids and solids in a particular system. Because existing models are not able to accurately capture the complex, 3D thermal energy exchange between cooling film flows and the main gas path, the optimization of combustor and turbine cooling configurations is limited.

Greater understanding is also needed regarding the effects of particle-laden flows entering the gas turbine from the external operating environment.³³ Because of air pollution in some regions of the globe, particle-laden

___________________

³³ Particle-laden flows affect the operation of gas turbines because the inlet supply air carries with it small particles that exist in the surrounding environment. For power generation as well as oil and gas applications, filters remove the large particles without significant pressure penalty; however, small particles can still exist in the supply air. For propulsion applications, filters are not feasible due to the pressure drop penalty. In the case of propulsion, while on the ground the inlet supply air to the gas turbine can contain surrounding dirt and sand, while in the air, the inlet supply air can contain volcanic ash and other particulate matter found in the atmosphere.

flows (with a mixture of solid particles, melted or partially melted particles, and hot gases) may enter the turbine and impact the components in the main gas path as well as the secondary coolant flow path.

This research area includes three research topics:

• Innovative Cooling
• Full Conjugate Heat Transfer Models
• Fundamental Physics and Modeling in Particle-Laden Flows

Research Topic 4.1
Innovative Cooling

Summary

Research to develop innovative cooling strategies needs to address the high temperatures and high mechanical stresses that turbine components experience as well as specific material properties and manufacturing methods. Turbine aerodynamic performance and durability requirements drive increased geometric complexity for cooling the combustor walls as well as in the vane and blade hardware.

Effective turbine thermal management in future gas turbines is related to the technological capability to manufacture geometrically complex components comprising high-temperature materials. Because of increases in the cycle pressure ratio, the air extracted from the compressor outlet to cool components in the combustor and turbine modules is at a higher temperature than in earlier generations of gas turbines. As a result, in some cases innovations will be needed to cool the coolant air.

Background

Turbine airfoils for power generation, propulsion, and oil and gas applications are typically manufactured using investment casting. Several steps are required for this process, starting with pouring wax into metal molds in the shape of the airfoil. Once each wax shape has set, it is removed from the mold and repeatedly immersed in a ceramic slurry bath, forming a ceramic coating that is then heated to further harden the ceramic and melt the wax. The actual airfoil is formed by pouring molten metal into the hollow space left behind from the melted wax. The internal air-cooling passages within each blade are also formed during this stage of production by inserting ceramic cores into the wax pattern.³⁴ Additional complexity arises because airfoils are generally made by directionally solidifying the molten metal to align the grain boundaries, thereby allowing for higher stresses. After the blades are further machined, film-cooling holes are placed in the external walls of the airfoils that lead to the internal passages that supply the coolant. Once the cooling holes are manufactured, a thermal and environmental coating is applied to the external surfaces of the airfoils to improve resistance to corrosion and oxidation as well as insulate the airfoil from the hot main gas path flow. Combustor walls are typically constructed from relatively thin sheets of high-temperature metal in a double-wall configuration where the external surfaces are sprayed with thermal and environmental coatings. The cooling strategies used for the combustor walls are similar to those used for vanes and blades. Research for both the turbine and combustor modules are typically categorized as either internal cooling (for surfaces inside the vane blade or between the combustor liner double-wall) or external cooling (for surfaces exposed to the hot gas path).

Internal cooling strategies could achieve high convective heat transfer coefficients through the use of highly turbulent flows and large surface areas, but only if geometric constraints can be overcome and if enough pressure

___________________

³⁴ J. Moxon, 1985, How Jet Engines Are Made, Threshold Books, London.

is available to appropriately drive the flow. Typical internal cooling strategies for the leading edges of airfoils include impinging jets on the backside of the inside airfoil surface. For the main body of the airfoil, serpentine channels that contain ribs are used to increase surface area and flow turbulence. Near the trailing edge where the blade external heat transfer coefficients are very high and the passages are required to be very thin due to the narrow trailing edges of the airfoils, pin fins are often used for high convective cooling while enhancing structural integrity. For the combustor walls, impingement cooling between the double-wall is commonly used along with pin fins between the two walls for increased surface area and turbulence while also improving structural rigidity. The external (hot side) wall of the combustor’s double-wall generally contains a high-density array of film-cooling holes.

Advanced internal cooling strategies for vanes and blades are often limited by the design of the die for the ceramic cores or the ability to cast small features. The tooling often restricts the designs of internal cooling passages. While there has been significant research on developing more effective ceramic cores for the casting, as discussed above additive manufacturing could open new opportunities. Another potential internal cooling strategy is to place microchannels in the skin of turbine airfoils to bring the cooling closer to the airfoil surface.³⁵ With advances in additive manufacturing, it is feasible to completely redesign cooling strategies to include lattice structures and even more complicated passages.

After cooling the internal surfaces of the airfoils and combustor liners, the coolant flow is exhausted through film-cooling holes. In the case of the airfoils, the coolant is also exhausted through slots in the trailing edge. On the external (hot gas) side of the airfoil and combustor walls, it is preferable to reduce convective heat transfer from the hot gas path. State-of-the-art manufacturing methods for film-cooling holes use either laser-drilling or electro-discharge manufacturing, which are processes completed after casting the turbine airfoil. In combustor walls, which are often double-wall designs, film-cooling holes are similarly either laser-drilled or electro-discharge machined.

Additive manufacturing methods can lead to complex film-cooling hole shapes that are better integrated with the internal coolant supply channels, which is particularly important for the entrances to the film-cooling holes. More complex film-cooling hole shapes may improve the quality of the film protection on combustor and turbine hardware. New manufacturing methods can also lead to removing the limitation of requiring a line of site as well as improve the tolerancing.

Integral to innovative cooling designs is the development of high-temperature materials used to make turbine airfoils. As discussed above, in the section “Research Area 2: Structural Materials and Coatings,” there is significant ongoing research to develop CMCs, which show significant promise in terms of increasing operational temperatures by as much as 100°C/180°F beyond those of existing single-crystal alloys.

Benefit

This research topic could accelerate ongoing research in innovative cooling strategies that enable higher turbine inlet temperatures, resulting in increased thermal efficiency while meeting life-cycle cost requirements. This research addresses a gap in the development of advanced cooling innovations in light of new high-temperature materials and additive manufacturing methods. Furthermore, research in this area will further reduce the cost and risk of large-scale adoption of additive manufacturing techniques and hot section cooling strategies in gas turbines. An example of what has happened in the past in terms of increased turbine inlet temperatures has demonstrated the impact of innovations in cooling strategies, particularly film-cooling. A disruptive increase in thermodynamic cycle performance is expected with the large-scale adoption of innovative cooling strategies enabled by additive manufacturing techniques and high-temperature materials.

This research topic could advance relevant innovative cooling technologies from TRL 1 to perhaps as high as TRL 6 by 2030. The high TRL is expected as a result of integrating additive manufacturing and advanced materials.

___________________

³⁵ This is also referred to as double-wall cooling.

The research topic has medium to high technical risk, depending on the technology, because of potential difficulties in scaling up cooling innovations for application in operational gas turbines. The full extent of this risk will depend somewhat on the development of additive manufacturing capabilities to reliably make the innovative cooling features at a cost and durability that are beneficial to the industry. The successful accomplishment of this research topic is therefore closely linked to the success of additive manufacturing as well as the time and cost to develop new cooling strategies.

This research topic applies to power generation, aviation, and oil and gas. Reducing fuel burn for all three applications is an important goal, particularly for power generation and for aviation. Improving the efficiency of oil and gas turbines when they are operating at partial loads has also been established as a priority.

Research Topic 4.2
Full Conjugate Heat Transfer Models

Summary

Conductive heat transfer is typically the dominant form of heat transfer in solids, while convective heat transfer typically dominates in liquids. A full conjugate heat transfer model analyzes heat transfer involving both solids and liquids in a particular system.

Validated full conjugate heat transfer techniques enable advances in optimizing combustor and turbine cooling configurations. Full conjugate techniques capture the complex, 3D thermal energy exchange between cooling film flows, main gas path flows, internal flows, and the solid components more accurately than the more commonly used loosely coupled conjugate analytical processes that consist of separate and sequentially executed lower fidelity submodels. The application of full conjugate techniques will increase gas turbines’ thermodynamic efficiency, while meeting their life requirements with less uncertainty. These full conjugate heat transfer techniques will have more influence during the engineering design process if they are validated with heat transfer data acquired from coordinated experiments on canonical geometries and flow conditions, with the support of industry, academia, and government stakeholders.

Background

As discussed above, state-of-the-art cooling strategies for combustor and turbine modules include but are not limited to closely packed arrays for film-cooling holes that generate low-temperature films that (1) insulate the underlying metal and protective thermal barrier coating from the high-temperature combustion products in the main gas path and (2) augment the rate of convective heat transfer between the cooling air flow in these holes and the surrounding metal. The rapid mixing of the protective film flows with the surrounding fluid is a complex, 3D process that is highly dependent on the geometrical characteristics of the cooling hole array, the internal cooling flow momentum and quality¸ and the external flow momentum and turbulent fluctuation levels. Furthermore, wall-bounded film flows that rapidly mix with fuel-rich hot gases promote secondary chemical reactions in the film and diminish the intended benefit of film cooling. The complex physics associated with the above are currently not understood.

Computational thermal models for predicting hot section metal temperatures typically use a loosely coupled conjugate heat transfer approach. For a turbine blade, submodels that capture the heat transfer processes among the internal fluid flow, the external fluid flow, and the metal and thermal barrier coating system are separately and sequentially executed until the temperatures at the interfaces of these models converge to the same value. The aerothermal submodels for the external airfoil surface use a combination of empirically driven low-fidelity and advanced high-fidelity tools to capture the effects of film mixing and insulation, heat transfer augmentation,

secondary chemical reactions, and energy transfer between the film and the solid blade surfaces. These submodels are typically validated with results from controlled experiments, during which the aforementioned effects are often measured separately. For cooling strategies that include closely packed arrays of cooling holes, however, the cooling films interact, the solid conduction pathways become more 3D, and the energy exchange between the external fluid flow and the blade solid becomes more complex. These complex aerothermal interactions are better captured with full conjugate heat transfer modeling techniques.

Benefit

Validated, full conjugate heat transfer modeling techniques for combustors and turbines with complex cooling configurations could be used to improve the accuracy of metal temperature predictions by reducing the modeling error associated with the simplified aerothermal submodels that are typically used in the loosely coupled conjugate analytical process. Enhanced predictive accuracy would increase the accuracy of hot section component life forecasts, improve hot section component durability, reduce combustor and turbine cooling air flows, and enable higher turbine inlet temperatures for greater thermodynamic efficiency while satisfying mission life requirements.

This research topic has a medium technical risk. Its success depends on (1) the generation of comprehensive data sets obtained from full conjugate heat transfer experiments on canonical combustor and turbine cooling configurations and flow conditions that are accessible by the technical experts in the gas turbine industry, academia, and government and (2) the validation of full conjugate heat transfer models with these publicly available data sets. This activity would require careful coordination among these key stakeholders.

This research topic applies to power generation, aviation, and oil and gas applications for gas turbines operating at full and partial load.

Research Topic 4.3
Fundamental Physics and Modeling in Particle-Laden Flows

Summary

Gas turbines in many geographic regions operate in increasingly challenging environments, where the concentration of particles such as sand or atmospheric particulates can significantly degrade gas turbine performance and often lead to shutdowns, especially for aircraft and the oil and gas industry. The basic physics associated with these environments is not well understood and requires integrated study using high-fidelity simulations and experimental validation for relevant environmental conditions ranging from simple to complex phenomena associated with particle ingestion.

Environmental particles can erode compressor blades. Within the hot sections of a gas turbine (i.e., the combustor and turbine modules), if a particle’s residence time is long enough, a rapid rise in the particle temperatures occurs. This can cause particles to adhere to component surfaces, thereby setting off a chain reaction of severe events. When a particle adheres to a surface, the metal temperature generally increases by either reducing the coolant flow due to blocked internal passages or increases the thermal resistance between the wall and coolant air. Higher metal temperatures, in turn, lead to higher temperatures of the particles adhered to the wall, thereby increasing the likelihood that more particles will adhere to the surface.³⁶ Upon cooling, the presence of the melted particles results in spallation of the protective coatings.

___________________

³⁶ W.S. Walsh, K.A. Thole, and C. Joe, 2006, “Effects of Sand Ingestion on the Blockage of Film-Cooling Holes,” pp. 81-90 in Proceedings of the ASME Turbo Expo 2006: Power for Land, Sea, and Air. Volume 3: Heat Transfer, Parts A and B, ASME, doi.org/10.1115/GT2006-90067.

This research problem is specific to gas turbines because of the particular external environments in which they operate as well as the high temperatures present in the turbine. Modeling the complexity of this problem requires an integrated approach using high-fidelity numerical simulations along with experimental validation. Defined test cases are needed that range from simple, fundamental benchtop simulations to more turbine-relevant complex cases to assess how to develop a better understanding of the various mechanisms affecting turbine operations.

Background

As global flight patterns increasingly traverse developing nations and as power generation and oil and gas turbines continue to be installed in a wide range of environments, the threat of small-particle ingestion into gas turbines grows. For power generation, contaminants that can reduce gas turbine performance include rust from upstream components and unfiltered particulates from the surrounding environment. For aircraft propulsion, contaminates of interest include volcanic ash, fine sand particulate suspended in the atmosphere or ingested during takeoff and landing, and industrial pollutants such as those generated by coal-burning power plants. Unlike gas turbines for aircraft propulsion, gas turbines for both power generation and oil and gas applications can use filters to remove many larger particles (>10 μm), but smaller particles remain in the main gas path flow.

Poor air quality affects the performance of each gas turbine module differently. In the compressor module, erosion is the concern: environmental particles drawn in by the fan can subsequently impact the compressor blades. Both the fan and compressor sections work to pulverize the particles. Once reaching the high-pressure compressor section, from which discharge air is bled to cool hot section components, the particles are small enough to be carried with the secondary cooling flows, where temperatures are much hotter, causing particle deposition. The particle deposition can block internal passages and cooling holes. In the main gas path, particles may be deposited on external airfoil surfaces, which increases their roughness. Rough turbine airfoils cause increased aerodynamic losses and can lead to early boundary layer transition on the airfoil resulting in high external heat transfer from the hot gases passing along the airfoils.

Where and how the particles deposit within a hot section component strongly depends on their size, composition, temperature, the internal cooling geometry, and the method of introduction. The mechanisms of particle transport and deposition within gas turbines are not well understood because all of the relevant conditions are nearly impossible to simulate in a controlled experimental environment. In the turbine module, the friction drag from the high-speed coolant can keep particles in an aerosol state where the particles track the flow. However, given the particle mass and momentum, the particles do not necessarily follow the streamlines through the turns or various cooling features. Instead, the particles impact surfaces where there are several forces, which are not well understood, that will dictate whether the particle will adhere to the surface.³⁷

Benefit

This research topic could accelerate ongoing research by identifying the principal mechanisms that drive the degradation of turbine durability from particle-laden flows. Once identified, these mechanisms would then be captured in experimental and numerical turbine simulations. The ultimate benefit of this research area is to provide a physics-based understanding of particle transport and deposition in high-pressure turbines that can be used to drive conceptual, particle-tolerant turbine cooling designs and to improve the quality of turbine component lifing forecasts in particle-laden flows. The latter will enable expanded operational limits. Currently, there are

___________________

³⁷ Ibid.

substantial shortcomings in understanding the driving mechanisms of particle deposition on turbine components. Basic, fundamental test cases are nonexistent, resulting in an inability to execute integrated, methodical experiments that enable the validation of low- and high-fidelity particle transport and deposition models relevant to turbine operations in particle-laden flows.

Physics-based models could be developed to drive various advanced, particle-tolerant turbine designs by 2030. Achieving a high-level of certainty in turbine lifing predictions would require significant breakthroughs in particle transport and deposition research and particle-tolerant turbine design concepts.

This research topic has high technical risk because the models may need to be tailored to the specifics of each case because of the complex interactions. The test cases to fully replicate gas turbine conditions at high pressures and temperatures are difficult at best.

This research topic applies to aviation, power generation, and oil and gas applications, by providing a better understanding of how turbine operations are affected by particle-laden flows, which improves turbine cooling designs as well as lifing models needed. For the aviation applications, particle-laden flows can disrupt operations by requiring aircraft to detour around regions such as volcanic plumes or over developing countries with an especially high concentration of particles, increase engine wear, and possibly lead to a loss of propulsion in flight. For power generation and oil and gas applications, particle-laden flows increase the frequency of maintenance and reduce the overall efficiency of the gas turbines.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.8. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrows (with an arrowhead at each end) show where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the thermal management research area are not shown.

RESEARCH AREA 5: HIGH-FIDELITY INTEGRATED SIMULATIONS AND VALIDATION EXPERIMENTS

Overview

Computational fluid dynamics (CFD) has been an important tool in aerospace engineering over the past four decades, and it has lowered development costs by reducing the number of physical tests required in the design process. The predominant CFD tool has been based on Reynolds averaged Navier Stokes (RANS) equations, which contain calibrated phenomenological models to represent the effect of turbulence fluctuations on the averaged flow quantities. The accuracy of RANS is limited by phenomenological modeling assumptions. There are several flow features in the flow path of a gas turbine that RANS models have difficulty predicting, including (1) flow separation and turbulent mixing and (2) quantities such as thermoacoustic oscillations and fluctuations of pressure and temperature that require accurate prediction of unsteady turbulence fluctuations. Large eddy simulations are high-fidelity computations that attempt to capture most of the energetic unsteady 3D flow features in flows such as those in the interior of a gas turbine. Subgrid-scale models are used to account for the effects of unresolved small-scale turbulent flow motions. This is in contrast to the RANS approach, for which the effect of all turbulence scales on the mean flow are modeled.

This research area includes three research topics:

• Numerical Simulation of Subsystems and System Integration
• Coordinated Experimental Research
• Computer Science and the Utility of Simulation Data

Research Topic 5.1
Numerical Simulation of Subsystems and System Integration

Summary

Integrated numerical simulations can capture interactions among gas turbine modules. Interactions of interest include dynamic couplings, flow distortion, unanticipated heating or loading,³⁸ and thermoacoustic instabilities that manifest only when the system is integrated. Greater insight into system coupling yields more accurate aerothermal and structural boundary conditions and, by extension, more realistic module predictions than single-module models with simplified boundary conditions applied at the interfaces between modules. High-fidelity numerical simulation tools would be benchmarked, calibrated, and then validated with experimental data acquired on canonical single-module and multimodule configurations (see the section “Research Topic 5.2: Coordinated Experimental Research,” below) in order to maximize the effectiveness of these tools during the design process. Applying validated computational models could yield improved learning outcomes from subsystem rig and full engine tests, permit faster engineering design optimization, and reduce engineering development costs.

___________________

³⁸ Stage loading is a measure of the load on a turbomachinery stage (compressor, fan, or turbine). It is related to pressure ratio across the stage.

Background

The first integrated multifidelity simulation³⁹ of an annular sector of a realistic gas turbine engine (PW6000) was demonstrated a decade ago.⁴⁰ In that simulation. the combustor and its upstream diffuser were computed using large eddy simulation, and the rotating machinery was simulated using RANS. The integration effort lacked robustness, as boundary conditions at the module interfaces had to be improvised. In the ensuing decade, computational power has increased by more than three orders of magnitude, and significant strides have been made in the development of accurate and efficient numerical methods that are especially suitable for prediction of the multiphysics turbulent flows⁴¹ encountered in gas turbines. This combination of advances in hardware and software have opened new opportunities for detailed engineering analysis in the design process, resulting in reduced design cycle time, avoidance of costly time and potential engineering rework, and more optimally designed gas turbine components. High-fidelity simulation capabilities have recently been used to study combustion instabilities in the GE 7HA heavy-duty gas turbine.^42,43 These integrated calculations used a single large eddy simulation code, but simulated only the central portion of the gas turbine, which includes the compressor discharge chamber (including the combustor prediffuser), the combustor, and the turbine’s first stage stator.

Benefit

This research topic could accelerate ongoing research in this area by leveraging the significant advances made over the past decade in high-fidelity numerical simulation capabilities for analysis and design of the next-generation gas turbines, and provide a cost-effective means of assessing integration effects early in the design process. It would accelerate the implementation and testing of the advanced numerical technology in the gas turbine sector.

The proposed research aims to demonstrate the applicability and predictive capability of a state-of-the-art high-fidelity integrated numerical technology for end-to-end simulation of realistic gas turbines. It would be the first integrated large eddy simulation of its kind to include both turbomachinery and combustor components. Additional physics-based modeling research will likely be needed, depending on the results of the validation experiments and comparison with the experimental data (see the section “Research Topic 5.2: Coordinated Experimental Research,” below).

This research topic is expected to advance relevant technology from TRL 2 to TRL 6 by 2030, depending on the level of engagement by design engineers and the feedback they provide.

This research topic has moderate technical risk because of the issues that may have to be resolved with subgrid-scale models (e.g., turbulent combustion, liquid fuel atomization, and wall models for flow and heat transfer) that may surface as the simulations are applied to realistic gas turbines. Just as important is the risk that new simulation tools will not be adopted if users continue to rely on legacy tools that have not leveraged recent advances in simulation technology. Of paramount importance is the involvement of gas turbine designers during the development of tools and models, without which the final outcomes will be at risk.

Physics-based simulations can provide near-instantaneous detailed data for engineering analysis. They also enable faster and more cost-effective exploration of the design space. If used judiciously, these simulations can reduce the number of expensive physical tests and make each test that remains much more valuable. Coordinated validation experiments are necessary to enhance the credibility of the simulation results and to reduce the risk of drawing misleading conclusions from numerical experiments.

___________________

³⁹ Different numerical models with different levels of accuracy and cost are used in different modules.

⁴⁰ G. Medic, G. Kalitzin, D. You, E.v.d. Weide, J.J. Alonso, and H. Pitsch, 2007, “Integrated RANS/LES Computations of an Entire Gas Turbine Jet Engine,” Paper AIAA 2007-1117 at 45th AIAA Aerospace Sciences Meeting and Exhibit, http://bit.ly/2WLZQzj.

⁴¹ Multiple physical phenomena such as gas–liquid interactions and chemical reactions are active.

⁴² J. Hines, “OLCF Helps GE Deliver Next-Generation Gas Turbines,” Oak Ridge National Laboratory, http://bit.ly/2NJmrZ5, accessed November 5, 2019.

⁴³ J. Hines, 2016, “Better Combustion for Power Generation,” Oak Ridge National Laboratory, May 31, http://bit.ly/36AEgSQ.

As an engineering tool, high-fidelity numerical simulations apply to power generation, aviation, and oil and gas applications. These simulations can be used in the analysis and understanding of physical phenomena such as unsteady thermal effects, unsteady loading on rotors, and thermoacoustic instabilities.

Research Topic 5.2
Coordinated Experimental Research

Summary

High-fidelity numerical simulations will require phenomenological models for unresolved physics for the foreseeable future. Research of interest includes modeling of turbulent combustion, wall heat transfer, wall–particle–turbulence interactions, surface roughness, and atomization of liquid fuel jets. Experiments are necessary to validate both component and integrated (e.g., the combustor and turbine) large-scale simulations.

Background

In high-fidelity simulations, the large-scale features of the flow are computed on a space–time computational grid, and the interaction of the unresolved small-scale features with the resolved scales are modeled via the so called subgrid-scale models. Subgrid-scale models are necessary for physical phenomena such as small-scale turbulence near a wall, liquid fuel atomization, and chemical reactions. Although there is a large body of experimental data available in canonical configurations (e.g., heat transfer in a turbulent boundary layer and chemical reactions in counter-flowing jets), significant gaps exist, and additional experiments are needed to guide and validate integrated simulations. For example, the following heat transfer measurements would be highly beneficial, especially if they are coordinated with companion numerical simulations and model developers:

• Heat transfer to the combustor liner (to characterize the effectiveness of effusion cooling near a swirling, reacting flow).
• Measurements of engendered thermal stresses in the combustor liner measurements.
• Heat transfer measurements on turbine blades in the presence of active cooling, surface coatings, or incident vitiated flows (see the preceding section, “Research Area 4: Thermal Management,” for more information).

Similarly, there is considerable experimental data available for thermoacoustic responses in realistic combustors, but acoustic boundary conditions for the facilities are poorly characterized and yet essential for high-fidelity numerical simulations. In turbomachinery, measurements to characterize end-wall effects (e.g., blade tips and hubs) are needed, especially for prediction of hub stall and tip loading in compressors.⁴⁴

Spatially resolved measurements of emissions would be challenging but very beneficial. This would involve experiments that would measure CO oxidation axially along a combustor (to help design combustor lengths) or particulate “number densities” (for soot) in both the near and far field of the reaction zones.

Benefit

This research topic could accelerate ongoing research in this area by developing validated unresolved physics models, which would enable the assessment of subgrid-scale models and could reduce uncertainties in the results of numerical simulations. Validation experiments would add credibility to both simulations of individual

___________________

⁴⁴ N. Gourdain, F. Sicot, F. Duchaine, and L. Gicquel, 2014, Large eddy simulation of flows in industrial compressors: A path from 2015 to 2035, Philosophical Transactions of the Royal Society A 372(2022), doi:10.1098/rsta.2013.0323.

gas turbine modules and large-scale simulations of the gas turbine as a whole. To fully realize the benefits of this research, coordination between laboratory experiments and numerical simulations is essential. Such coordination would ensure the measurements of relevant parameters in realistic conditions, accelerate model development, and enhance the reliability of the simulations.

This research topic could advance relevant technology from TRL 2 to TRL 6 by 2030. It has medium technical risk because of the difficulty in avoiding inapplicable results, which can arise if the experimental research is done in isolation rather than in coordination with numerical simulations.

Detailed experimental data of flow, temperature, heat transfer, and combustion products are of high value in all sectors of gas turbine space: industrial, aviation, and oil and gas industries. In addition to their utility in model development and validation of high-fidelity simulations, experimental data obtained in relevant conditions provide benchmarking platforms for mixed fidelity engineering models in all gas turbine applications.

Research Topic 5.3
Computer Science and the Utility of Simulation Data

Summary

Mapping high-fidelity numerical algorithms to emerging computer architectures is key to bringing to bear the latest advances in high-performance computing to the gas turbine industry. Workflow advances in software (including grid generation, rapid input/output data, and in situ diagnostics) are needed for widespread adoption of the high-fidelity simulation tools by design experts.

Background

The NASA Vision 2030 study⁴⁵ makes recommendations for developing advanced CFD capabilities for aerospace applications by 2030. It identifies challenges for development of advanced CFD software, including the following:

• Managing complex geometries, scalable mesh generation, and adaptation.
• Efficient deployment of next-generation hardware and innovative algorithms that could, for example, enable high-fidelity simulations at a sufficiently reduced cost so that they could be routinely used for high-throughput design studies.
• Efficient data mining capabilities for learning from data sets generated by large-scale simulations. This includes development of physics-constrained machine learning algorithms for (subgrid-scale) modeling of unresolved thermal and fluid phenomena.

All three of the above challenges define research for high-fidelity numerical simulations that are described by the other two research topics included in this research area. Physical geometry and flow path in a gas turbine is extremely complex. Efficient grid generation algorithms are needed to produce high-quality grids with rapid turnaround. The gap in the ability to produce data and the ability to move data (for post-processing in workstations) is widening exponentially. The net result is that analyzing simulation data are becoming exceedingly difficult, and large simulation data are analyzed less because they are cumbersome to interrogate. Leveraging present advances in computational power will require mapping simulation algorithms and codes onto emerging high-speed computer

___________________

⁴⁵ J. Slotnick, A. Khodadoust, J. Alonso, D. Darmofal, W. Gropp, E. Lurie, and D. Mavriplis, 2014, CFD Vision 2030 Study: A Path to Revolutionary Computational Aerosciences, NASA/CR-2014-218178, https://go.nasa.gov/2rab8l5.

architectures. Targeted research is needed to address these particular bottlenecks and challenges to enhance the usability and effectiveness of high-fidelity simulations.

Benefit

This research topic would accelerate ongoing research in this area by increasing the speed of computational algorithms and large-scale data analysis methods, which are critical for effective use of numerical simulations in engineering analysis and in reducing gas turbine design cycle time. It would also facilitate the increased use of high-fidelity numerical simulations by design engineers and potentially reduce the number of costly physical tests. Designers could ask “what-if” type questions in the virtual environment and understand cause and effect relationships better than by other available means. Efficient data analytic tools (e.g., dynamic mode decomposition⁴⁶) would enhance the ability to interrogate simulation data for engineering analysis. At present, there is relatively limited ongoing research in this area.

This research topic could advance relevant technology from TRL 4 to TRL 7 by 2030. Lack of adoption of the simulation tools by the designers poses the most significant risk. Furthermore, if the tools are developed without guidance from actual designers of gas turbines, it is unlikely that the resulting simulation technology would have its potential impact.

Efficient mesh generation technology, data mining techniques for interrogation of large data sets, and leveraging advanced computer architectures would be beneficial to all gas turbine applications.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.9. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrows (with an arrowhead at each end) show where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the research area on high-fidelity integrated simulations and validation experiments are not shown.

RESEARCH AREA 6: UNCONVENTIONAL THERMODYNAMIC CYCLES

Overview

The conventional approach for improving gas turbine efficiency relies on increasing cycle pressure ratio, increasing turbine inlet temperature, and improving the efficiency of individual turbine modules (compressor, combustor, and turbine). The development of unconventional thermodynamic cycles would constitute an alternative approach. Research into unconventional thermodynamic cycles to improve the gas turbine performance such as thermal efficiency can have two distinct paths:

• As an enabler of advanced design parameters (e.g., by increasing efficiency by enabling higher turbine inlet temperature and cycle pressure ratios).
• As a performance enabler in and of itself (e.g., by increasing efficiency independent of turbine inlet temperature and cycle pressure ratio).

___________________

⁴⁶ P. Schmid, 2010, Dynamic mode decomposition of numerical and experimental data, Journal of Fluid Mechanics 656:5-28, http://bit.ly/2NJmrZ5.

Possible research and development (R&D) subjects include intercooling (with or without recuperation⁴⁷), reheat or sequential combustion, and complex thermodynamic cycles to reduce the amount of air that exits the compressor and is used for hot gas path cooling. This would include, for example, closed-loop cooling of turbine stage-one stator nozzle vanes with steam or a combination of steam cooling with reheat.

Many thermodynamic cycles could potentially improve gas turbine performance via their unique differences from the standard gas turbine Brayton cycle. Cycles of potential interest include the following:

• Pressure gain combustion cycle with simultaneous pressure and temperature rise in the combustor. In contrast, the Brayton cycle used by conventional gas turbines features constant-pressure combustion. Variants of pressure gain combustion are:
  • Pulse(d) detonation combustion
  • Rotating detonation combustion
  • Shockless explosion combustion
• Semi-closed oxy-combustion cycle. This cycle combusts fuel using pure oxygen, which makes it much easier to remove CO₂ from the exhaust.
• Turbocompound cycle. This cycle combines an internal combustion reciprocating (i.e., piston-cylinder) engine and a gas turbine in a hybrid cycle arrangement.

___________________

⁴⁷ Recuperation refers to preheating of compressed air with hot exhaust gas prior to entering the combustor. The objective is to reduce fuel burn in the combustor and increase cycle efficiency.

• Hybrid gas turbine and fuel cell combined cycle. This approach would incorporate fuel cells in a combined cycle power plant.
• Supercritical CO₂cycles. These cycles would use pure liquid CO₂ at supercritical pressures and temperatures as the working fluid in a closed system based on a Brayton or Rankine cycle. Such a system could be used either as a stand-alone gas turbine or as the bottoming cycle in a combined cycle power plant in combination with a conventional gas turbine as the topping cycle.

The cycles listed above would be most easily implemented on large, ground-based turbines (i.e., turbines for power generation applications), although some might eventually be practical in oil and gas or aviation applications.⁴⁸

Because of the maturity of conventional gas turbine technology, which goes back for more than a century, it will be difficult to mature systems using unconventional thermodynamic cycles to the point that they can be implemented in such a way that the impact on gas turbine cost; size; reliability, availability, and maintainability (RAM); and other aspects of life-cycle cost is acceptable from a cost-performance trade-off perspective.

This research area includes three research topics:

• Gas Turbines with Pressure Gain Combustion: Technology
• Gas Turbine Cycles for Carbon-Free Fuels
• Gas Turbine Cycles with Inherent Carbon Capture Ability

___________________

⁴⁸ In fact, pressure gain combustion and turbocompound cycles were originally conceived for aircraft propulsion applications.

Research Topic 6.1
Gas Turbines with Pressure Gain Combustion: Technology

Summary

The most efficient variant of the ideal gas turbine Brayton cycle involves constant volume heat addition. Achieving this in a steady-flow device such as a gas turbine presents significant difficulties even in laboratory conditions. Approximation of the ideal process in actual hardware, under the generic name of pressure gain combustion (PGC), mainly via detonation, has been investigated since 1950s in a mostly sporadic fashion.

While the underlying thermodynamics is unassailable, it is extremely difficult to implement a PGC process in a steady-state, steady-flow device such as a gas turbine. The best-known technique to achieve this is the explosive combustion taking place in the cylinders of a reciprocating internal combustion engine when the piston is at or near the top dead center. In fact, the first successful gas turbine in commercial operation, designed and manufactured by Hans Holzwarth in 1908 in Germany, had such a combustion chamber with a two-stage turbine.

A Holzwarth-like implementation in a modern gas turbine is problematic at large air flows and outputs due to the limitation imposed by the size of the explosive combustion chambers and the intermittency of gas flow into the turbine stage (reduced efficiency). Currently, the most promising technology to achieve this in a gas turbine with axial flow through the compressor and turbine is PGC. The specific PGC technology currently under investigation for aviation and power generation applications is detonation combustion. (An earlier version, pulsed detonation combustion, has been dropped in favor of the rotating detonation combustion due to the better amenability of the latter to quasi-steady flow applications.) Shockless explosion combustion is another approach to constant-volume combustion that merits additional investigation.

The thermodynamic principles underlying the transformative potential of PGC to elevate cycle thermal efficiencies by several percentage points are indisputable. There is already a widespread effort on the fundamental combustion research.

Background

Ideal cycles such as the Brayton cycle include heat addition and heat rejection processes. Ideal heat addition can take one of the following forms: (1) constant pressure (Brayton cycle), or (2) constant volume (Atkinson cycle⁴⁹). Implementation of the ideal cycle heat addition in actual “flange-to-flange” gas turbine can be done in one of the two ways: (1) combustion (open cycle), or (2) heat exchange (closed cycle). The combustion (or heat exchange) process in the actual gas turbine hardware can neither be constant pressure nor be constant volume due to flow friction, heat loss, and myriad other loss mechanisms.

Temperature-entropy representations of Brayton and Atkinson cycles are shown in Figure 3.10. In both cases, point 1 is the inlet to the compressor, point 2 is the exit from the compressor and the inlet to the “heat adder” (e.g., the combustor of a gas turbine in practice), point 3 or 3A is the exit from heat adder and inlet to the turbine, and point 4A is the exit from the turbine (or nozzle, for an aircraft engine). The key physical mechanism for improving efficiency with a constant volume heat addition cycle is the ability of the heat adder to act in part as a compressor, thereby increasing the temperature and pressure of the working fluid simultaneously. Thus, for the same overall cycle pressure ratio, the mechanical compressor in a constant-volume heat adder cycle draws off less power than the compressor in a constant-pressure heat adder cycle.

___________________

⁴⁹ Referred to as the Humphrey cycle in some references, especially in aeronautical treatises.

The most direct means of improving the efficiency of a large power generation turbine is to raise the turbine inlet temperature. The current state of the art can enable turbine inlet temperatures up to about 1,700°C/3,100°F. If PGC (e.g., via detonation combustion) can be successfully implemented in a gas turbine, it could improve cycle efficiency up to 2 percentage points and reduce fuel burn by about 3 percent. This is equivalent to increasing turbine inlet temperature by about 200°C/350°F, which would be very challenging to achieve with low emissions and adequate parts life. Thus, this technology has the potential for a transformative improvement in cycle performance of gas turbines for power generation applications.

Benefit

This research topic could accelerate the advancement of gas turbine technology for PGC from TRL 3 to TRL 6 by 2030. The primary hurdle to overcome in PGC development is the fact that developing hardware to implement combustion in a quasi-steady flow process with an adverse (i.e., increasing in the direction of flow) pressure gradient is a challenging engineering task.

Tailor-made PGC cycle configurations can prove themselves to be more amenable to practical implementation in a successful prototype (rather than trying to “squeeze in” an additional, complex piece of equipment into the existing, already tightly spaced architecture).

The research topic has high technical risk because, ultimately, regardless of the resources spent on it, a concept readily amenable to practical implementation may turn out to be unachievable. This is balanced by low (relatively speaking) risk from a resource allocation perspective because thermodynamic cycle research does not involve complex and expensive hardware or infrastructure.

This research topic applies to power generation, aviation, and oil and gas applications.

Research Topic 6.2
Gas Turbine Cycles for Carbon-Free Fuels

Summary

Combustion of hydrogen in modern, dry, low NO_x gas turbine combustors is problematic due to issues associated with NO_x emissions, combustion stability, and safety, among others.⁵⁰ A new gas turbine cycle for carbon-free fuels such as hydrogen could provide designers with more options for solving hydrogen combustion issues.

Background

Potential carbon-free or carbon-neutral fuels include synthetic hydrocarbons produced via chemical reactions from CO₂ (captured from power plant stack or ambient air) and hydrogen generated via electrolysis of water using renewable energy sources (e.g., solar or wind). Coal gasification with carbon capture and sequestration is another option (carbon-neutral but not carbon-free). Investments to enable practical application of these fuel production schemes and others at large scale are ongoing. The main hurdles for using these fuels in gas turbines are achieving low NO_x emissions and integrating the power generation cycle with the fuel production system for high thermal efficiency, reliability, availability, and maintainability.

___________________

⁵⁰ Details on issues associated with the use of hydrogen as a fuel appear in Chapter 1, in the section “Background Information for the Performance Improvement Criteria.”

Benefit

This research topic could accelerate ongoing research in this area by expanding the scope of ongoing research to address hydrogen issues, which is currently focused on improvements that can be implemented with gas turbine designs that are based on the Brayton cycle. In other words, an unconventional cycle could provide the framework for burning hydrogen in a stable manner with low NO_x emissions without hampering gas turbine thermal efficiency, reliability, availability, and maintainability.

This research topic could advance relevant technology from TRL 1 to TRL 3 by 2030.

The research topic has high technical risk because, ultimately, regardless of the resources spent on it, a concept readily amenable to practical implementation may turn out to be unachievable. The main hurdle to overcome in that respect is that material availability can preclude the achievement of a feasible solution. This is balanced by low (relatively speaking) risk from a resource allocation perspective because thermodynamic cycle research does not involve complex and expensive hardware or infrastructure.

In any event, efforts spent on successful cycle design can accelerate the transition to zero-carbon hydrogen economy.

Prima facie, this research topic applies to all types of gas turbine applications. Nevertheless, it may prove difficult to design a cycle that satisfies demanding requirements of aircraft propulsion units (e.g., size, weight, safety, and reliability). More likely applications include land-based electric power generation and marine vessel propulsion.

Research Topic 6.3
Gas Turbine Cycles with Inherent Carbon Capture Ability

Summary

Gas turbines burning natural gas generate 60 percent less CO₂ per each megawatt-hour of electricity (on average) than coal-fired power plants. Carbon capture can further reduce the environmental impact of gas turbines. Unconventional thermodynamic cycles that use oxy-combustion (i.e., the combustion of fuel using pure oxygen) can provide high thermal efficiency with easy removal of CO₂ from the flue gas stream.

Background

Because oxy-combustion uses pure oxygen instead of atmospheric air (which is mostly nitrogen), with natural gas as a fuel (mainly methane, CH₄), gas turbine exhaust consists entirely of water vapor and CO₂. Thus, capturing CO₂ for industrial use or sequestration is simply a matter of cooling the gas and condensing water vapor out of the exhaust. The biggest impediment is the power-intensive production of pure oxygen in an air separation unit.

The Allam cycle is a patented example of an oxy-combustion cycle that is currently being investigated by individual investors and OEMs. The Allam cycle differs from other oxy-combustion cycles in that CO₂ constitutes 95 percent of the fluid flow in the combustor (by mass), with the rest made up by oxygen and fuel. The resulting combustion product is 90 percent CO₂ and the parasitic power consumption of the air separation unit is minimized by the lower oxygen requirement. A 50-megawatt thermal (MWth) demonstration plant is being built in Texas at a projected cost of $140 million, funded partly by major players in the power industry.⁵¹ The commissioning and testing of the combustor as a stand-alone unit in a building adjacent to the turbine building was undertaken

___________________

⁵¹ For details, refer to the website of the primary technology developer, NET Power, http://www.netpower.com.

in late 2018. In 2019, the entire plant is expected to start the test runs. A 300 MW electrical commercial plant is said to be in the works.

Benefit

This research topic could accelerate ongoing activities in this area by supplementing private research that is already under way on the Allam cycle with an organized approach drawing on larger resources (e.g., manpower, facilities, and funds) to find alternative oxy-combustion cycles that would be easier to incorporate in a gas turbine. This research topic would significantly help to advance oxyfuel-combustion gas turbine technology from TRL 3 to TRL 6 by 2030.

The research topic has high technical risk because, ultimately, regardless of the resources spent on it, a concept readily amenable to practical implementation may turn out to be unachievable. This is balanced by low (relatively speaking) risk from a resource allocation perspective, because thermodynamic cycle research does not involve complex and expensive hardware or infrastructure.

Even if the goal of a commercially ready product is not achieved by 2030, the lessons learned along the way would more than justify the research and development expenditure.

This research topic applies to stationary gas turbine applications—more specifically, land-based gas turbines for electric power generation. (It is practically impossible to store or use captured carbon—very difficult to achieve in the first place—on a moving platform with extremely limited space.)

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.11. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrow (with an arrowhead at each end) shows where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the unconventional thermodynamic cycles research area are not shown.

RESEARCH AREA 7: SYSTEM INTEGRATION

Overview

The first serial-production gas turbine as it is known today is the Jumo-004, which entered service in 1943 in the Me-262 German interceptor aircraft. This basic architecture (i.e., a gas turbine core consisting of a compressor module, combustor module, and turbine module on or around a common shaft in the direction of the gas flow) has not changed since and is not expected to change substantially in the foreseeable future.

Within the conventional architecture of a gas turbine, systems integration opportunities are limited. Nevertheless, in the ensuing eight decades, myriad variations to the conventional architecture have been in albeit limited commercial use. They are listed below:

• Cooling air cooler, which allows rotor or hot gas path cooling air to be extracted from the compressor and then cooled in an external heat exchanger;
• Enhanced air-cooling system, which is similar to a cooling air cooler with a booster compressor;
• Combustor bypass, which is used for CO emission control at partial load;
• Steam cooling of hot gas path components in a gas turbine using steam from the bottoming cycle of a combined-cycle plant;
• Intercooling⁵² (LMS100 aeroderivative gas turbine);
• Reheat or sequential combustion⁵³ (GT24/26 gas turbines); and
• Recuperation (Solar’s Mercury 50).

Within the three modules of a conventional gas turbine architecture, subsystem integration could be improved through the incorporation of component-level changes. Examples include the following:

• Blade integrated disks (blisks);
• Magnetic bearings;
• Variable stator vane in turbine stage 1;
• Trapped vortex, ultra-compact combustion;
• Auto-thermal, on-board syngas generation;
• Integrated combustor and stator vane;
• Counter-rotating open rotor (aviation only); and
• Ultra-high bypass turbofans (aviation only).

The scope of this research area could encompass the potential and feasibility of various combinations of known options for improving system and subsystem integration (enumerated above) and as yet unknown options. This research area includes three research topics:

• Gas Turbines with Pressure Gain Combustion: System Layout
• Closed Cycle Gas Turbines
• Hybrid Gas Turbine Systems

Research Topic 7.1
Gas Turbines with Pressure Gain Combustion: System Layout

___________________

⁵² Intercooling refers to adding a cooler between successive compressor stages to reduce mechanical compression work.

⁵³ Reheat refers to a two-step heat addition (combustion) process with a turbine expansion in between. The goal is to increase effective heat addition temperature of the cycle and thus its efficiency.

Summary

PGC offers significant efficiency gain in simple or combined cycle configuration. However, the unsteady and high-speed nature of the detonation combustion process (the most promising PGC variant) precludes straightforward replacement of conventional combustion systems in a flange-to-flange gas turbine. Significant research and development is requisite for an optimal architecture to cost-effectively realize the maximum benefit from PGC. Without substantial investment into system layout/configuration, advances made in PGC technology will not be able to make their way into a feasible commercial product.

Unlike aviation turbines, gas turbines for power generation are not hampered by space and size limitations. Therefore, researchers and designers have more degrees of freedom available to them to arrange key system components to alleviate the difficulties stemming from the unsteady nature of the PGC process.

Background

Additional information on PGC is provided in the section “Research Area 6: Unconventional Thermodynamic Cycles,” above.

Benefit

This research topic would support system design optimization simultaneously and in coordination with basic combustion system research and development. This would reduce the time required to develop a working prototype and enable an exchange of ideas between the two research activities, thereby improving component and system design. Postponing system design work until after the completion of combustion development will unnecessarily delay the introduction of a working prototype by many years. Concurrent research in system and component layout and hardware design and combustion fundamentals is the best path forward.

This research topic could advance relevant technology from TRL 3 to TRL 6 by 2030 by concurrent research and development on the fundamental thermal/combustion science and practical aspects of field implementation with continuing information exchange between the two disciplines.

The research topic has high technical risk given that even after 60 years of research and development on PGC, a commercially viable product is still not yet in sight. Moving from theory to practice has proven to be very difficult.

This research topic applies to power generation because of the large size and weight of a PGC system.

Research Topic 7.2
Closed Cycle Gas Turbines

Summary

Closed cycle gas turbines are readily amenable to efficient and clean power generation from carbon-free energy sources such as solar and nuclear. Especially with working fluids such as supercritical CO₂ and unconventional thermodynamic cycles (e.g., the Allam cycle), there is a significant potential for carbon-free power generation with or without carbon capture.

Closed cycle gas turbines are not amenable to the simple architecture of conventional (open cycle) flange-to-flange gas turbines with only three components. The resulting complexity has been the primary hurdle to widespread commercial acceptance even though the basic concept has been around for decades (including limited commercial deployment). Focused research and development in system layout/hardware design optimization can change that.

Background

Closed cycle gas turbines have been around since the 1930s. In 1939, the Swiss company Escher-Wyss built a 2 MWe test installation in their factory in Zurich. The power cycle, with air as the working fluid in a closed loop, was named after its inventors, J. Ackeret and C. Keller, as the “AK cycle.” Eventually, between 1950 and 1981, several experimental and commercial closed cycle gas turbine power plants were built in Europe, Japan, and the United States.

The coupling of a high-temperature, gas-cooled nuclear reactor with a closed cycle gas turbine power conversion unit using helium as the working fluid was first proposed by C. Keller in 1945. However, in the 1940s these technologies were in their infancy—clearly, Keller was way ahead of his time. As a result, of all the closed cycle gas turbine power plants built so far, only one, ML-1, a 400 kWe unit designed and built for the U.S. Army, had a nuclear reactor as the heat source. All others were fossil-fueled.

In 1974, a 50 MW closed cycle helium gas turbine power plant entered service in Oberhausen, Germany. This was an intercooled-recuperated closed cycle machine with a cycle pressure ratio of 27:1, a turbine inlet temperature of 750°C/1,400°F (85 kg/s helium mass flow rate), and a thermal efficiency of 23 percent. Alas, as a result of myriad design shortcomings, the turbomachinery could not reach its design performance.

Even with decades of sporadic interest in its design and operation, closed cycle gas turbines have not advanced to the state of a mature, commercial product. However, due to its unique nature (i.e., an external heat source), a closed cycle gas turbine is a highly flexible power generation platform amenable to renewable as well as fossil fuel-fired applications (e.g., nuclear reactor, solar tower, or coal-fired heater). This justifies a renewed focus on research and development into this technology.

Benefit

This research topic could accelerate ongoing research in this area by filling a void in power generation technology. Closed cycle gas turbine system research and development is not the subject of a focused initiative. It is mostly relegated to being an element of research into concentrated solar power, advanced modular nuclear reactors, or gas turbines using supercritical CO₂. If a breakthrough is made, this research topic could advance closed cycle gas turbine technology from TRL 3 to TRL 7 by 2030.

This research topic has medium technical risk because of the difficulty of achieving a cost-effective design with high reliability, operability, and maintainability.

This research topic applies to power generation and, perhaps, oil and gas, depending on the selection of the heat source.

Research Topic 7.3
Hybrid Gas Turbine Systems

Summary

The performance of a simple cycle gas turbine can be significantly enhanced via integration with another technology (as opposed to simple “attachment” as in the case of Brayton–Rankine combined cycle facility). The best-known examples are a fuel cell–gas turbine hybrid system and turbocompounding. In hybrid systems, the goal is to achieve compact and cost-effective integration of two constituent subsystems without degrading system reliability or performance.

Research into hybrid gas turbine systems in the past has been sporadic, proceeding without a long-term dedicated focus. When cheap fossil fuel–based centralized power generation technologies were dominant, this was a natural outcome. Presently, sustainable (carbon-free), distributed generation presents itself as the future of electric

power generation. In a similar context, hybrid prime movers (i.e., batteries combined with a gasoline engine) have already found their place in the car industry. Other transportation applications are within sight as well.

Background

By far, the largest loss mechanism in a gas turbine Brayton cycle is associated with the irreversible nature of combustion. This inefficiency could be avoided, at least in part, by replacing the combustion-based Brayton cycle heat addition with another process such as a fuel cell—specifically, a solid oxide fuel cell. In a fuel cell–gas turbine hybrid system, the fuel cell replaces the combustor and acts as the “topping cycle.” In other words, the fuel cell generates power through the direct conversion of the chemical energy of the fuel to (1) electrical energy and (2) hot gas, which is then expanded in the gas turbine. In this scheme, the gas turbine acts as the bottoming cycle. Efficiencies of up to nearly 70 percent are possible in such a system.

Research and development of key solid oxide fuel cell technologies applicable to electric power generation systems are coordinated through the Solid State Energy Conversion Alliance. The primary focus of this research is utility-scale power generation with a coal feedstock that generates cost-effective electricity with near-zero levels of harmful emissions, facilitates capture of more than 97 percent of the carbon in the fuel, and has an efficiency of 63 percent or more for subbituminous coal feedstock and minimal water consumption.

Immediately following World War II, turbocompound aircraft engines (which combined a conventional piston engine with a gas turbine⁵⁴) were seriously considered as the next-generation technology for high power and efficiency. Indeed, British Napier-Nomad engine still holds the brake efficiency record for an aircraft engine. Nevertheless, the extreme weight and complexity of the turbocompound engine made it no match for much simpler and lighter gas turbines. For stationary shaft or electric power generation, however, turbocompounding still holds significant efficiency advantage at small ratings (e.g., 100 MW class).

Benefit

This research topic could accelerate ongoing research in this area by developing hybrid gas turbine systems, which can be advantageous in certain applications such as small-scale distributed generation or cogeneration.

This research topic could advance relevant technology from TRL 3 to TRL 7 by 2030 by reducing the system cost without hampering thermal efficiency or RAM.

Despite the technical challenges involved, this research topic has medium technical risk primarily because of the large body of work done earlier. Economic and commercial considerations such as established manufacturing base and cheap fuel precluded to make the final push into the proverbial next step. A focused effort with matching resources, assisted by changing priorities (i.e., carbon-free power generation in sync with a large renewable portfolio), can accomplish this.

This research topic applies primarily to power generation applications. Size and weight would likely make aviation applications impractical.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.12. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrow (with an arrowhead at each end) shows where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the system integration research area are not shown.

___________________

⁵⁴ Turbocompounding is different from the widely used turbocharging in that both the engine and the turbine contribute to shaft power generation. In turbocharging, there is no net shaft output generated by the turbocharger.

RESEARCH AREA 8: CONDITION-BASED OPERATIONS AND MAINTENANCE

Overview

The most basic approach to the maintenance of mechanical systems is limited to repairing them when they fail. Preventive maintenance reduces system breakdowns by conducting maintenance before system failure. The most basic approach to preventive maintenance uses a schedule with predetermined intervals between specified maintenance actions. Preventive maintenance of this type is not affected by system operating status or level of performance. Over time, the desire to maximize the reliability and availability of some systems has greatly increased the cost of preventive maintenance. In order to reduce these costs, condition-based maintenance practices were developed. Condition-based maintenance determines when preventive maintenance is needed based on information collected during system operation rather than relying on a predetermined maintenance schedule. Condition-based maintenance consists of three steps: (1) data acquisition, (2) data processing, and (3) maintenance decision making. For gas turbines, the adoption of long-term service agreements provided additional motivation to develop and improve conditions-based maintenance. Long-term service agreements enabled gas turbine owners to predetermine the costs of ownership. This new business model required OEMs to verify that gas turbines operated within the contractual limits. This required sensors, data acquisition systems, and data storage and analysis over the product life cycle. With gas turbines being closely monitored, both OEMs and owners started to exploit opportunities to reduce risk, operating and maintenance costs, and design margins through increasingly sophisticated analysis of sensor data. Over the years, the combination of sensor data with prognostics, digital twins, and AI has greatly advanced the benefit of condition-based maintenance.

Condition-based operations and maintenance (CBOM) of gas turbines, for the purpose of this report, is the next step beyond condition-based maintenance. CBOM is defined as the capability to optimize the operation, maintenance, repair, and overhaul of gas turbines throughout their life cycles.

CBOM technologies developed under this research area would enable gas turbine owners to increase operating hours and reduce operating costs by extending maintenance and overhaul intervals, reducing unplanned maintenance, reducing the time it takes to complete maintenance and overhauls, and allowing operators to schedule maintenance and overhauls at optimal times. Reducing unplanned maintenance is of particular value. A turbine may be shut down for much longer than it takes to conduct the maintenance as the necessary personnel, parts, and equipment are brought to the site.

Advanced CBOM technologies will also enable gas turbine operators and manufacturers to have operational laboratories, which will enable gas turbine owners or OEMs to have a deeper understanding of the components, parts, and systems that have large levels of uncertainty. For instance, extending the repair interval of a new additively manufactured component on one experimental gas turbine would generate data relevant to other products and thereby benefit the entire product family. Because of its interconnectivity, this research area can also significantly leverage the research and technology related to digital twins and their supporting infrastructure, as discussed later in this chapter.

This research area includes three research topics:

• Sensors
• Inspection and Repair Technologies
• Advanced Controls

Research Topic 8.1
Sensors

Summary

Sensors for gas turbines face challenging requirements in terms of size, cost, temperature, pressure, reliability, and lifetime. Sensors capable of operating within the harsh internal environment of a gas turbine are required to support real-time monitoring of components. Sensors are frequently used to measure temperatures of metal components or a gas as well as pressures within a gas turbine. The signals from these sensors have traditionally been transmitted via hard wiring. Within a gas turbine, high-pressure, high-temperature gases flow at high speed through components with complex geometries and advanced materials. Obtaining accurate information about the state of the gases and materials across the gas turbine provides operators with the information necessary to improve turbine operations and reduce the cost and impact of maintenance.

Background

Each individual wired sensor requires a wire (or pair of wires) that may run long distances from the sensor to a signal collection device. If the sensors are placed on rotating blades, slip rings are also needed. Fully instrumenting a large gas turbine with wired sensors for system validation is time consuming (typically 3 to 6 months) and expensive (more than $1 million), and even then, there is a high likelihood that some sensors and wiring will fail prematurely. Wireless sensors that are easily installed, relatively inexpensive, robust, reliable, and use energy harvesting for power would have great benefit compared to conventional wired sensors. The sensors and the signal transmitting system would need to be capable of withstanding high temperatures, harsh environments (corrosive and oxidizing), and vibrations for the life of the turbine component.

Temperature data are very important, and a number of different techniques are currently used to measure the temperatures of gas turbine components. This includes wire thermocouples, sprayed thermocouples, thin film thermocouples, pyrometers, thermal paint, thermal liquid crystals, and infrared thermography. However, each of these techniques has limitations and would benefit from further development.

Achieving the goals of this research topic will require dedicated research on specific sensors and sensor capabilities. Important contributions could also be made by investigating options for improving the integration of sensors with turbine components using emerging technologies such as additive manufacturing.

Benefit

This research topic could accelerate ongoing research in this area by providing higher fidelity data, which would enable a reduction in gas turbine design and operating margins. Advanced sensors with improved capabilities, lower cost, wireless connectivity, longer life, and increased accuracy are needed to monitor physical conditions such as temperature, pressure, and vibration within the gas turbine. Sensors with better reliability would reduce the required number of sensors by reducing the need for redundancy. Additional information of interest includes the condition of components in terms of, for example, coating spallation, cracking, and cooling flow restrictions. It is particularly difficult to develop sensors that can function reliably in the harsh operating conditions in the combustor and turbine modules and on rotating components. The data provided by sensors can be used to support digital twins, new product development and validation, and operational monitoring in support of CBOM. In some cases, improved sensors could also enable rapid introduction of new technologies.

This research topic could advance relevant technology, particularly with respect to wireless sensors, from TRL 3 to TRL 8 by 2030.

This research topic has medium technical risk because the gas turbine operating environment is harsh and quite often the space available to place sensors is relatively small. Powering the sensors and transmitting information from the sensors is also a challenge because of high temperatures, corrosive environments, and rotating components.

This research topic applies to power generation, aviation, and oil and gas applications because the sensor requirements and challenges for all three gas turbine applications are similar.

Research Topic 8.2
Inspection and Repair Technologies

Summary

Gas turbine maintenance costs and availability are two of the most important concerns for operators. Typically, inspection and repair requirements are outlined in the OEM manual provided to gas turbine owners. Borescope inspection programs traditionally are used to monitor in situ condition of key gas turbine components such as the combustor and turbine blades and vanes. Downtime can be substantially reduced by extending the time between planned maintenance (especially for major overhauls) and expediting return to service following maintenance. Key tasks include the following:

• Improving the quality of in situ inspection data, thereby enabling higher confidence assessments of gas turbine conditions.
• Assessing critical component features and capabilities.
• Tailoring in situ cleaning and maintenance.
• Improving time and material planning for maintenance.

This research topic would develop in situ inspection and repair technologies that draw from emerging technologies such as robotics, telepresence, teleoperation, AI, and automation.⁵⁵

___________________

⁵⁵ J. Eggart, C.E. Thompson, J. Sasser, and M. Merine, 2017, Heavy-Duty Gas Turbine Operating and Maintenance Considerations, GER-3620M, GE Power, Atlanta, Ga., https://invent.ge/2Nj12qQ.

Background

Gas turbines with extended hours of operation will experience degradation in performance efficiency that is typically not restored without engine downtime, disassembly, and component-level maintenance. Components that are major contributors to required maintenance are gas path airfoils in the compressor, as well as hot gas path components in the combustor and high-pressure turbine modules. The state of the art today for in situ inspection can be described by two overarching concepts. First, inspection is manually conducted by individual experts, assisted by limited measurement technologies, who perform navigation tasks by hand, perform visual inspection of assets, and use their judgement to identify and classify defects. The skill sets required to complete inspection by these means are built up over time and experience. Navigation of the spaces required to complete inspections is very difficult in some cases, requiring great situational awareness and concentration. Interpretation of the video images of the inspection site is a second skill set, which is typically unrelated to the first, and operators are required to travel to inspection sites in order to get the visual inspection data. Second, fixed inspection cameras are becoming more common for large power turbines for inspecting airfoils in a consistent way to leverage automation and AI for interpretation of inspection results. Either a camera is positioned on a rotor blade and used to image every stator blade as the rotor is slowly rotated, or a camera is positioned in a static position and used to image every rotor blade as the rotor is slowly rotated. These techniques make it much easier to qualify maintenance staff to serve as inspectors. These techniques also make it easier to assess inspection data using automation and AI systems. However, each of these inspections typically (1) requires dedicated fixtures and tools, and (2) are relatively inflexible in application. In addition, these techniques only work where a rotor component and a stator component are in close proximity and suitably positioned relative to each other.

Novel combinations of the following emerging technologies would enable substantial improvements for in situ inspection and repair:

• Teleoperation and telepresence;
• Miniaturized robotic mechanisms;
• Monocular simultaneous localization and mapping;
• Autonomous navigation;
• Virtual- and augmented-reality data presentations;
• AI-assisted interpretation of inspection data; and
• New human/machine interfaces.

Benefit

This research topic could accelerate ongoing research related to the following:

• Miniaturization of sensor technology and end-effector technologies for inspection and repair to be introduced through borescope ports or other engine passages.
• Enabling inspection and repair experts to work remotely from gas turbine site, thereby reducing the skill set required of local technicians.
• Using sensors in confined spaces, which enables spatial metadata tagging to simplify and automate navigation.

These technology advancements will enable enhanced capability for in situ inspection and repair of key components for defects, changes in dimensions, missing material, coating condition, debris, corrosion, damage assessment, and rotor clearance. An additional benefit is that advanced inspection and repair technologies can impact the existing fleet as soon as they meet technical readiness, thus resulting in a faster return on investment. The information generated from these inspections would help validate sensor data and analytics methodologies for monitoring gas turbine metrics and efficiency.

This research topic could advance relevant technology from TRL 3 to TRL 6 or higher for many inspections being conducted autonomously by robotic systems equipped with AI by 2030.

The research topic has medium technical risk because of the following:

• Durability of miniaturized tools. Miniaturized robotics for inspection and repair is required to maximize the ability to navigate the remote or critical locations in the gas turbine. These tools will need to be robust for cost and inspection purposes, but also for reliability relative to extremely high confidence that the tool that entered the engine will exit the engine as expected and in entirety.
• Data management. New inspections and repair technologies will enable the collection of vast amounts of data. It is not known today what data will be useful in the future, so it is important to store and aggregate the data to help train and enable new inspection AI-based on evolving requirements.

There is also one important nontechnical risk. Regulatory changes may be required to redistribute responsibility and authority for inspections and repair using telepresence or teleoperation.

This research topic applies to gas turbines for power generation, aviation, and oil and gas applications. For aviation, this research will increase time on wing, optimizing timing for component repair. For power generation and oil and gas, it will maximize online operation and minimize long-term maintenance costs.

Research Topic 8.3
Advanced Controls

Summary

This research topic would develop “smarter” gas turbines that can be controlled more effectively by exploiting information from relevant domains such as the external environment (e.g., weather forecasts), commercial conditions (e.g., fuel cost and the price of electricity), and customer preferences (e.g., risk aversion). This research topic would draw heavily on global advances in the state of the art of AI in general and machine intelligence in particular. This research topic would benefit gas turbines by improving the performance through control and reducing the operational costs of the gas turbine through the application of digital twins. It has the capability to impact the industry in the short term, as this research topic has a relatively low development cost and because the research results can be applied to existing gas turbines much more readily that advances in related fields such as heat transfer, combustion, and materials.

Background

As described in Chapter 1, in the section “Background Information for the Performance Improvement Criteria,” electric grids around the world are undergoing significant changes, in large part because of greater incorporation of renewable energy sources and the development of highly distributed ground power systems.

Benefit

This research topic could accelerate ongoing research in this area by developing better control algorithms that incorporate AI to enable gas turbines to operate at their optimum point, which will result in lower emissions, improved part-load efficiencies, and extended product life. It will also enable lower maintenance and repair costs by providing improved guidance in terms of maintenance logistics.

This research topic could advance relevant technology from TRL 1 to TRL 8 by 2030. Such a progression is believed to be possible because of rapid advances in AI globally, because control methods have already been in development in several other fields, and because their application to real products is what makes sense for these digital solutions.

The research topic has low technical risk because it focuses on developing mathematical algorithms, machine learning, and data analytics applications. Moreover, the research area can build from ongoing research and development in the automotive and robotics industries, whose control algorithms can be extended. There is also one important nontechnical risk. Gas turbine life-cycle data have important value to gas turbine operators and OEMs, and much of the data is viewed as proprietary. Establishing data sharing agreements among researchers, operators, and OEMs are therefore essential for successful completion of this research topic.

This research topic applies primarily to power generation given that it is driven by requirements associated with the changing electrical grid. The results of this research topic, however, could also benefit gas turbines for aviation and oil and gas applications to the extent that improved control technologies developed would improve their ability to meet their system requirements.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.13. The green arrow (with a single arrowhead) shows where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrow (with an arrowhead at each end) shows where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the CBOM research area are not shown.

RESEARCH AREA 9: DIGITAL TWINS AND THEIR SUPPORTING INFRASTRUCTURE

Overview

A digital twin is a virtual representation (or digital model) of a physical gas turbine component (e.g., blade or vane), module (e.g., compressor, combustor, or turbine), or system (i.e., the gas turbine). High-fidelity digital twins will accurately depict the real-time condition of the physical part (physical twin) to predict performance issues early, facilitate optimum system operational and maintenance management, and reduce the development and manufacturing lead time for new components through experience gained in the field.

The digital twin comprises both physics-based and probabilistic models along with data generated from all stages of the physical system’s life. The digital twin is continually evolving to reflect the changes that occur in the physical part during manufacture, operation, service, and repair. This is accomplished through the collection and analysis of new information from sensors during each phase of the system life cycle in addition to input from

models. Advanced data analytics are required to handle and organize the enormous amounts of information that feed into the digital twin. The fidelity of the model may be further enhanced through the use of machine learning.

The digital infrastructure that supports digital twins is known as the “digital thread.” The digital thread encompasses validated models of turbine operation, structured and unstructured data, real-time data analytics, and standards for interoperability between disparate data systems. Key functions of a digital thread are collecting, interpreting, and transmitting data from many different sources in a variety of formats to develop and maintain the digital twin, including an accurate history of components. Digital threads will have the ability to locate, validate, and provide the data required by the physics-based lifing models that constitute the core of a digital twin. The digital thread links events that define the history of a physical part. It starts when a component, module, or system is conceived and finishes when the part is removed from service.

Digital threads will tag data with appropriate metadata to enable understanding, identification, and automated manipulation of data for archiving and retrieval. There is currently a wealth of information that could be used to support digital twins, but the infrastructure to bring all the information together in a useful manner is often lacking.

Ideally, the digital thread would be capable of securely retrieving and storing very large quantities of data in human- and machine-readable formats. Sources of data include the following:

• OEMs;
• Operators (e.g., utilities, airlines, and pipeline companies);
• Third-party repair and overhaul facilities;
• Computer-aided design models;
• Internal and external manufacturing suppliers for components, including sensors;
• Alloy suppliers; and
• Casting, forging, and machining vendors.

Data of particular interest encompass the following:

• Design models;
• Operating conditions;
• Operational inspection results;
• Output of engineering analyses;
• Maintenance, repair, and overhaul;
• Manufacturing data (e.g., final dimensions);
• Material pedigrees and processing parameters;
• Traceability of data to subsystem- and system-level requirements; and
• Model-based definitions of material performance to facilitate design of components based on their specific locations within a gas turbine (and the specific operating conditions at those locations).

Collecting and collating high-quality data has a direct impact on the usefulness and reliability of the digital twin. Incomplete or inaccurate data may significantly degrade the importance of the digital twin. In some cases, however, important data may be lacking. For example, some information relating to the pedigree of a part may not be available because a vendor may view it as competition sensitive and choose not to release it.

This research area includes one research topic, which follows.

Research Topic 9.1
Digital Twins and the Digital Thread

Summary

This research topic focuses on the development of both digital twins and the digital thread that supports digital twins. To optimize the performance of gas turbines and maximize component life, it is important that the influence of changing operating parameters such as ramp rates, partial load, and over firing⁵⁶ are understood, and that tools are available to simulate and predict the outcome of such changes on an appropriate time scale. In some cases, this will require the processes to operate in real time.

Background

Gas turbine components operate under very harsh environments. The temperatures in the turbine section often exceed the melting temperature of the alloys, and their successful operation is made possible only through the use of thermal barrier coatings and effective cooling configurations. Loss of a protective coating or blockage of a cooling passage can lead to premature failure of a physical component. Digital twins fed with real-time data from sensors can actively predict the behavior of the physical asset. This information can be used to make actionable decisions to control the operation of the gas turbine.

Although concepts for digital twins and the digital thread have existed for many years, they are not widely implemented for gas turbines. The aviation industry is beginning to embrace digital twins for maintenance, repair, and overhaul,⁵⁷ and OEMs are seeking to employ digital twins for power generation turbines.⁵⁸ Recent advances in digitalization have created an environment that will enable the realization of this technology. The ability of a digital twin to accurately represent and predict the behavior of its physical twin is directly related to the availability and fidelity of data that characterize the state and operational history of the physical part. A comprehensive model requires high-fidelity data from all stages of life, from design, through manufacture, operation, and retirement. This requires the capability to capture and process large quantities of structured and unstructured data, often in real time, using data analytics. Data security is crucial, as is interoperability between disparate data systems.

Sensors are an integral part of the digital twin and digital thread, and they are essential for providing real-time data representing the condition of the physical asset and/or the environment in which it is operating. Research is needed to develop sensors that are (1) capable of operating under the extreme conditions within a gas turbine, (2) tailored to fit within the geometric constraints of the turbine, and (3) able to monitor critical parameters. The development and requirements for relevant sensors are addressed in the section “Research Area 8: Condition-Based Operations and Maintenance,” above.

Benefit

This research topic could accelerate ongoing research in this area by developing models to predict the behavior of physical parts as well as the software and tools necessary to provide a robust and effective mechanism to gather, sort, and analyze large volumes of data from many different sources with various formats.

Some aspects of digital thread infrastructure (e.g., mechanical lifing models) have already achieved TRL 6, and in certain niche applications (e.g., the incorporation of CMCs in aviation gas turbines by GE Aviation), the infrastructure is also well developed (TRL 6). However, the ability to integrate necessary elements of an effective digital twin and digital thread are not well developed for most applications and are generally at or below TRL 3 today. This research topic could advance relevant technology to TRL 6 by 2030, and a complete infrastructure could be demonstrated as a prototype within this time frame.

___________________

⁵⁶ Over firing is temporarily running the gas turbine above its design firing temperature. It is the fastest method of boosting the power output of a gas turbine and gas turbine–combined cycle power system. Operators limit the use of over firing because excessive firing temperatures can reduce the life of parts in the hot gas path and increase maintenance costs.

⁵⁷ N. Etong, 2019, How MRO is unlocking huge opportunities for digital twins in aviation, Avionics International, March 26, http://bit.ly/2CjAjEm.

⁵⁸ GE Power Digital Solutions, 2016, GE Digital Twin: Analytic Engine for the Digital Power Plant, https://invent.ge/33lvhmB.

The research topic has medium technical risk because the individual parts must be developed and validated before being incorporated into an overall digital infrastructure.

This research topic applies to gas turbines for power generation, aviation, and oil and gas applications because they will all benefit from having advanced digital twins supported by a robust and reliable digital thread. This will ensure that all the pertinent data are fully utilized to support high-fidelity digital twins.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.14. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrows (with an arrowhead at each end) show where two research areas are mutually supportive to a substantial degree. Research areas that do not have a strong interrelationship with the research area on digital twins and their supporting infrastructure are not shown.

RESEARCH AREA 10: GAS TURBINES IN PIPELINE APPLICATIONS

Overview

The first U.S. long-distance natural gas pipeline (120 miles) was completed in 1891 from Indiana to Chicago. A boom in pipeline construction took off in the 1920s.

Initially, gas pipelines were free flowing. That is, the pipelines did not have any compression equipment. As distances increased, however, compressor stations were needed to overcome friction losses of the gases within the pipelines. The first compressor stations used reciprocating compressors driven by internal combustion engines. It was not until the early 1950s that gas turbines were introduced in pipeline compressor stations. These turbines

were paired with centrifugal compressors. Since then, gas turbines have been predominant in pipeline compression stations, with electric motor–driven compressors as the most common alternative. Most reciprocating compressors in the oil and gas industry are now found in production facilities where the volume flows are relatively small and the desired pressure ratios are high.

The United States has a network of natural gas pipelines that include about 3 million miles of pipelines and more than 1,400 compressor stations. Most of the compressors are driven by gas turbines. As discussed in Chapter 2, improvements in pipeline compressor turbines would have more impact on the oil and gas industry than gas turbines used for any other purpose. Still, in many cases large industrial power generation plants are at the receiving end of a pipeline and therefore must be able to burn the fuels transported in the pipeline. Additional background information on pipeline turbines appears in Chapter 1. This research area features two distinct research topics:

• Efficiency of Pipeline Gas Turbines Under Partial Load
• Safe Operation of Gas Turbines in Pipeline Applications with Hydrogen Fuels

Research Topic 10.1
Efficiency of Pipeline Gas Turbines Under Partial Load

Summary

The focus of this research topic has been largely neglected mainly because compressor station operators do not need to pay for the natural gas in the pipeline that is diverted to the operation of the gas turbines in the compressor station. Future emission requirements as well as efficiency improvements to minimize the carbon footprint are likely to change this situation.

Gas turbines in pipeline applications as well as many upstream applications regularly operate at variable partial load conditions for extended periods of times. Future pipeline operations will require more flexibility from the gas compressor stations to balance load swings of renewable energy sources and varying consumer demands. Several approaches have the potential to support the success of this research topic, as follows:

• Aerodynamic solutions
• Design solutions
• System solutions

Background

Aerodynamic Solutions

The operating temperatures and pressures at various points within gas turbine modules and components change when the gas turbine operates at different loads and different ambient conditions. Improving efficiency when operating under partial load is also constrained by trade-offs among the ability to meet emission standards, and how quickly the turbine can accommodate load changes.

While variable geometry in the axial compressor of a gas turbine is an established and proven mechanism to optimize gas turbine performance during start-up, shutdown, and transient operations, variable aerodynamic solutions in the turbine section are not well researched. This could include bleed avoidance for lean premix gas turbines for operation at partial load, advanced closed loop guide vane controls, and maintaining high firing temperature at partial load. Application-specific aerodynamic solutions can be easily modified to optimize existing conditions without paying the penalty of adjustable geometry.

Appropriate research subjects include variable geometry that allows adjusting the aerodynamic performances or removes the need for bleed air at partial load conditions. While variable geometry in the compressor section is well established, variable geometry in the hot section poses operational difficulties. Possible other solutions could involve the use of methods that do not alter the flow path geometry mechanically but by aerodynamic means (e.g., jets). Additive manufacturing provides methods to customize gas turbines for specific applications. Other improvements include more sophisticated controls, including the full utilization of digital twin models, feedback methods within the closed loop gas turbine control system, and other AI methods.

Methods could also allow the design of more efficient, optimized solutions because the operating range of the individual component (i.e., airfoil) changes less, if the geometry is variable.

Methods are already used for gas turbines with two shafts that allow high firing temperatures to be maintained at partial load.⁵⁹ These methods typically require bleeding compressed air, thus creating inefficiencies. A method that allows maintaining a high firing temperature under partial load without bleeding would be advantageous.⁶⁰

Design Solutions

Improved designs for relatively small gas turbines could increase efficiency and improve quick start capability. If gas turbines are most efficient at full load, a larger number of small gas turbines will be more adaptable to changing operating conditions. Smaller gas turbines, however, are often not as efficient as larger turbines, and using multiple small turbines instead of a single large turbine would increase the number of start and stop cycles for the individual turbines.

System Solutions

Pipelines allow for significant storage effects, and an advanced control system could modify the operation of compressor stations along an entire pipeline to maximize the fuel efficiency of the pipeline gas turbines as a whole.

A systems approach could lead to innovative developments such as the following:

• Hybrid drives that would allow compressors to be driven either by a gas turbine or an electric motor.
• Inlet cooling and heating to optimize the gas turbine inlet state for the anticipated load.
• Exhaust heat recovery to create power for additional compressors (e.g., turbines driven by steam or other fluids such as supercritical CO₂,⁶¹ or organic fluids compatible with the organic Rankine cycle⁶²), including hybrid drives. This would allow exhaust heat to generate electricity or heat a working fluid. In either case, recovered exhaust heat could be used either to power compressors, to augment the power of a gas turbine compressor train, to reduce gas turbine inlet temperature, or for other unrelated tasks such as generating electricity for the grid, supporting oil and gas processes, or providing chilled air for facility cooling.

The above system solutions could be used in combination with excess power from renewable energy sources to support energy storage concepts, including the production and storage of hydrogen.

Another system solution could include systems that predict gas usage at the supply points of pipelines and consequently optimize the operation of gas turbine compressors in the pipeline system.

___________________

⁵⁹ Mechanical drive turbines usually have two independent shafts that rotate at different speeds. One shaft connects the air compressor and high-pressure turbine of the gas turbine. The second shaft connects the low-pressure turbine with the mechanical load (e.g., a pump or compressor).

⁶⁰ The motivation here is different from providing combustors for low emissions at partial load. Here, the goal is to maintain high firing temperatures that enable higher efficiency.

⁶¹ Supercritical CO₂ refers to carbon dioxide when in a fluid state.

⁶² The operating fluid of an engine based on the organic Rankine cycle is an organic fluid with a boiling temperature that is lower than the boiling point of water.

As discussed below, the U.S. pipeline system has immense storage capability. For example, a typical stretch of 1 mile of 42 in. diameter pipeline under about 50 bar pressure stores about 750 MW-hr of energy. Utilizing the existing pipelines in a changing electric power generation environment can overcome one of the biggest challenges with intermittent renewable energy sources, which is energy storage. The United States’ vast network of pipelines also makes it possible to distribute energy to areas where it is most needed. Research in the field of gas turbine performance optimization is key to operate the existing assets most efficiently at all load points.

Benefit

A gas turbine that can operate at partial load with high efficiency and low levels of harmful emissions would (1) allow gas turbines to be more competitive with electric motor drives; (2) reduce CO₂ emissions; (3) reduce fuel consumption and life-cycle cost; and (4) facilitate off-peak gas flows to natural gas power plants, which will become increasingly important as renewable energy sources become more prevalent. In addition, these gas turbines would provide a safer and more reliable alternative to compressor driven by electric motors, especially in regions that are remote from power transmission lines or the electrical grid is not reliable.

The measures described above can be developed in different timelines and in some cases retrofitted in existing facilities as they become available. For example, the systems solutions can be introduced without having to wait for the aerodynamic or design solutions.

This research topic could accelerate ongoing research in this area by specifically targeting partial load operations.

This research topic could advance relevant technology from TRL 3 to TRL 7 by 2030. Relevant technologies could be developed and implemented in new gas turbines and, ideally, existing installations as well.

The research topic has medium technical risk because some of the enabling technologies needed to improve the partial load efficiency are at relatively high TRL. Others will have to be developed and validated in prototype testing.

This research topic applies to oil and gas and, potentially, to power generation applications.

Research Topic 10.2
Safe Operation of Gas Turbines in Pipeline Applications with Hydrogen Fuels

Summary

The introduction of significant levels of hydrogen into the existing natural gas pipeline network would need to overcome significant challenges. In general, the difficulty of overcoming these challenges increases as the percentage of hydrogen increases. The challenges are described in detail in Chapter 1.⁶³

This research topic would address challenges that impact gas turbine design and operations, such as the need for new combustors, sensors, and energy conversion systems to enable gas turbines to operate safely and efficiently with hydrogen fuels. This research topic would not address other challenges that impact the natural gas pipeline system as a whole, such as the potential for hydrogen embrittlement of pipeline materials or the need to increase flow volumes with the introduction of hydrogen in order to transport the same amount of energy as a given volume of natural gas. Furthermore, the compression work per energy unit goes up substantially with the introduction of hydrogen. As a consequence, the gas turbine power and turbine speed will both have to be increased.

___________________

⁶³ See the discussion of fuel flexibility in Chapter 1, in the section “Background Information for the Performance Improvement Criteria.”

Gas turbines used for pipeline compression stations must meet specific standards.⁶⁴ Complying with these standards will require close collaboration among technology development organizations, gas turbine OEMs, and pipeline operators to enable the development of gas turbines that can safely operate with hydrogen fuel without sacrificing performance.

Background

There are two main reasons for introducing a mix of hydrogen and natural gas in pipelines, as follows:

• The increase in renewable power generation capacity is leading to substantial overproduction of electricity at certain times of the day. Producing hydrogen is one way to productively use that excess capacity.
• Many initiatives are driving the decarbonization of fuels to decrease carbon emissions.

While the generation of hydrogen with excess electricity has just begun, it is anticipated that the amount of hydrogen that can be generated during peak production times in the future could lead to hydrogen concentrations in natural gas pipelines up to 10 percent by 2020, 20 percent by 2025, and 100 percent by 2030.⁶⁵

This research topic will accelerate ongoing research for using high-hydrogen fuels in pipeline gas turbines. As the percentage of hydrogen increases, it becomes more difficult to meet current standards and expectations for pipeline gas turbine safety and performance. Specifically, gas turbine applications in the pipeline industry are operated over a wide range of loads with more varied operating conditions than large power generation turbines. Cost-effective combustion technologies are needed for operation over this range of conditions without increasing the levels of harmful emissions or reducing gas turbine durability.

Benefit

This research topic could accelerate ongoing research in this area by designing a gas turbine combustion system along with other components and systems that would allow safe operation of gas turbines with a fuel mix that includes a range of hydrogen.

By 2030, this research topic could advance relevant technology from TRL 3 to TRL 8 for fuel mixtures with up to 50 percent hydrogen and to TRL 6 with fuel mixtures substantially in excess of 50 percent hydrogen.

The research topic has medium technical risk up to concentrations of 50 percent hydrogen in the fuel, as existing combustions systems can be modified to handle the new fuel mixtures. With more than 50 percent hydrogen in the fuel, the technical risks become high because the current dry, low NO_x emissions combustion systems will not function properly. A complete redesign of combustors and, to a lesser extent, other gas turbine modules is therefore likely. Added complexity comes from the likely requirement to use fuel with a wide range of hydrogen concentrations in the same gas turbine.

This research topic applies to oil and gas applications (particularly pipeline compressor stations driven by gas turbines) and, to a lesser extent, power generation.

Interrelationships with Other Research Areas

Key interrelationships between this research area and the other research areas are shown in Figure 3.15. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. Research areas that do not have a strong interrelationship with the research area on gas turbines in pipeline applications are not shown.

___________________

⁶⁴ See, for example, standards promulgated by American Petroleum Institute, specifically API616 for gas turbines, API617 for natural gas compressors, and ASME PTC22 and PTC10 for gas turbine and natural gas compressor testing respectively.

⁶⁵ In this report, hydrogen concentrations are reported in terms of mole percentage (which describes the number of hydrogen molecules present compared to the total number of gas molecules) rather than, for example, mass percentage.

RESEARCH AREAS: INTERRELATIONSHIPS AND RECOMMENDATION

The key interrelationships among the research areas are shown in Figure 3.16. The green arrows (with a single arrowhead) show where the accomplishment of one research area will substantially support the accomplishment of another research area. The red arrows (with an arrowhead at each end) show where two research areas are mutually supportive to a substantial degree. As shown, the different research areas are closely linked to one another, and efforts to support gas turbine research in different areas would be most effective if research plans are well coordinated.⁶⁶

___________________

⁶⁶ The interrelationships involving each individual research areas are easier to discern in the figures at the end of the discussions of each research area earlier in this chapter.