3
Power System Design

This chapter describes how matching power source technologies (sources) with particular electronics applications (sinks) can affect the energy efficiency of systems. It then discusses the pivotal role of hybrid power/energy systems and provides insight into how to optimize the energy efficiency of an integrated source-sink system.

It is important to match energy requirements for each power-using device with the characteristics of the power sources. Figure 3-1 depicts the idealized characteristics of a battery, where voltage and efficiency throughout the discharge cycle and the charge capacity are not affected by the rate of discharge. Real batteries and other real power sources do not have such idealized characteristics. Voltage outputs and efficiencies of power supply systems are strong functions of each device’s design details and the attributes of the duty cycle that must be serviced by the device. To optimize a system, both the dynamic characteristics of the power supply as well as attributes of the duty cycle must be understood.

Examples of such design considerations are highlighted in the following sections. These examples portray the need for detailed understandings of the dynamic characteristics of the power supply system and of the duty cycle. Without such details and a good analytical model to evaluate options, it is not possible to create an optimized system.

The various sources of power have different characteristics, as depicted for fuel cells and batteries in Figure 3-2. This figure shows the variation in efficiency with power output for a typical rechargeable battery and a direct methanol fuel cell. As is shown in the diagram the efficiency of the direct methanol fuel cell is somewhat above 30 percent at its rated power. As power output from the fuel cell is decreased, the efficiency drops continuously until it approaches zero at very small power output levels. This drop in efficiency is caused by the power requirements of balance-of-plant (BOP) components needed to support the fuel cell processes. This drop in efficiency makes the fuel cell a poor choice for systems where much of the energy is used at power levels lower than the rated level of the fuel cell. The battery, on the other hand, has a relatively high efficiency across the variation in power output from rated power to zero output. This high efficiency is related to the fact that energy is not needed to maintain the battery operation. Thus, batteries become the most energy-efficient source where the power outputs must swing over a large range of values.

DYNAMIC POWER

The traditional approach for estimating power supply capacity is to measure power demand in major operational states (standby, idle, peak), estimate the fraction of time to be spent in each of the states (the duty cycle), and sum over the resulting weighted averages. While this approach provides an estimate of system capacity, it ignores the dynamic behavior of the power demand. The power system can be more effectively designed if the dynamic profile of utilization is taken into account.

For example, a pager has very low power drain when in monitoring mode. Even when interfacing with a user to retrieve messages or to send a message the power requirement is easily met by a small battery. However, when the pager must transmit, the power required is substantially higher. Rather than construct a power supply to support the peak demand, pagers employ an electrochemical capacitor, which can be trickle-charged to provide large amounts of energy when the short message is transmitted. This hybrid power solution is smaller and weighs less than a system designed to meet the capacity of the transmission spikes.

Figure 3-3 depicts nonideal behavior, wherein voltage decreases with the time of discharge and the charge capacity decreases with discharge rate. In some battery chemistries, some capacity is recovered between periods of discharge.

It is common in small, portable electronic devices to reduce battery weight by drawing from the high demand portion of the battery output curve, causing a drop in



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 27
Meeting the Energy Needs of Future Warriors 3 Power System Design This chapter describes how matching power source technologies (sources) with particular electronics applications (sinks) can affect the energy efficiency of systems. It then discusses the pivotal role of hybrid power/energy systems and provides insight into how to optimize the energy efficiency of an integrated source-sink system. It is important to match energy requirements for each power-using device with the characteristics of the power sources. Figure 3-1 depicts the idealized characteristics of a battery, where voltage and efficiency throughout the discharge cycle and the charge capacity are not affected by the rate of discharge. Real batteries and other real power sources do not have such idealized characteristics. Voltage outputs and efficiencies of power supply systems are strong functions of each device’s design details and the attributes of the duty cycle that must be serviced by the device. To optimize a system, both the dynamic characteristics of the power supply as well as attributes of the duty cycle must be understood. Examples of such design considerations are highlighted in the following sections. These examples portray the need for detailed understandings of the dynamic characteristics of the power supply system and of the duty cycle. Without such details and a good analytical model to evaluate options, it is not possible to create an optimized system. The various sources of power have different characteristics, as depicted for fuel cells and batteries in Figure 3-2. This figure shows the variation in efficiency with power output for a typical rechargeable battery and a direct methanol fuel cell. As is shown in the diagram the efficiency of the direct methanol fuel cell is somewhat above 30 percent at its rated power. As power output from the fuel cell is decreased, the efficiency drops continuously until it approaches zero at very small power output levels. This drop in efficiency is caused by the power requirements of balance-of-plant (BOP) components needed to support the fuel cell processes. This drop in efficiency makes the fuel cell a poor choice for systems where much of the energy is used at power levels lower than the rated level of the fuel cell. The battery, on the other hand, has a relatively high efficiency across the variation in power output from rated power to zero output. This high efficiency is related to the fact that energy is not needed to maintain the battery operation. Thus, batteries become the most energy-efficient source where the power outputs must swing over a large range of values. DYNAMIC POWER The traditional approach for estimating power supply capacity is to measure power demand in major operational states (standby, idle, peak), estimate the fraction of time to be spent in each of the states (the duty cycle), and sum over the resulting weighted averages. While this approach provides an estimate of system capacity, it ignores the dynamic behavior of the power demand. The power system can be more effectively designed if the dynamic profile of utilization is taken into account. For example, a pager has very low power drain when in monitoring mode. Even when interfacing with a user to retrieve messages or to send a message the power requirement is easily met by a small battery. However, when the pager must transmit, the power required is substantially higher. Rather than construct a power supply to support the peak demand, pagers employ an electrochemical capacitor, which can be trickle-charged to provide large amounts of energy when the short message is transmitted. This hybrid power solution is smaller and weighs less than a system designed to meet the capacity of the transmission spikes. Figure 3-3 depicts nonideal behavior, wherein voltage decreases with the time of discharge and the charge capacity decreases with discharge rate. In some battery chemistries, some capacity is recovered between periods of discharge. It is common in small, portable electronic devices to reduce battery weight by drawing from the high demand portion of the battery output curve, causing a drop in

OCR for page 27
Meeting the Energy Needs of Future Warriors FIGURE 3-1 Characteristics of an ideal battery: (a) constant voltage and (b) constant capacity. FIGURE 3-2 Power source efficiency variation with load. Typical efficiency for DMFC and battery. SOURCE: Adapted from data from Ball Aerospace 20-W fuel cell.

OCR for page 27
Meeting the Energy Needs of Future Warriors FIGURE 3-3 Typical voltage discharge profiles. Nonideal battery properties: (a) voltage change; (b) loss of capacity; and (c) recovery. SOURCE: Adapted from Linden, 1995. capacity. Figure 3-3 shows that for a real battery, the rate of discharge affects the battery’s apparent capacity. Figure 3-3a shows that voltage drops more rapidly for the higher discharge rate represented by curve 2 than for the lower discharge rate represented by curve 1. Figure 3-3b compares the percent of initial capacity and how it is affected by discharge rate. If C represents the capacity available at a standard discharge rate, then discharge rates from 0.1C to 10C are portrayed on the abscissa. As can be noted, initial capacity drops from 100 percent at 0.50C to approximately 70 percent at 10C. Figure 3-3c shows that if the discharge is intermittent, some recovery in cell output voltage occurs between periods of discharge. Figure 3-4 presents another analysis of a battery system used on various duty cycles. For continuous discharge, the battery specific energy drops from approximately 138 Wh/kg at a specific power of 75 W/kg (point B) to about 90 Wh/kg at a specific power of 300 W/kg (point A). When intermittent operation is added, it is the peak power rather than the average power that determines capacity. In Figure 3-4, point B and point C are loads with the same average power, 75 W/kg, but the battery delivers a much higher capacity for point B than for point C. The reason for this is that the load of point B is a constant 75 W/kg, but the load of point C is an intermittent discharge with a peak power of 300 W/kg and a 25 percent duty cycle. The capacity delivered for the intermittent discharge of point C is much closer to the capacity delivered for a continuous discharge at the peak power value of 300 W/kg (point A) than it is to the capacity for a continuous discharge at the average power value (point B). Using the average power, the capacity would be estimated at 140 Wh/kg, nearly 30 percent greater than the actual capacity. There is a slight dependence on the duty cycle, as shown by the family of curves representing intermittent loads with peaks of 100, 200, and 300 W and duty cycles of 25, 50, and 75 percent. However, despite this dependence, for intermit-

OCR for page 27
Meeting the Energy Needs of Future Warriors FIGURE 3-4 Doyle’s Li ion model results for capacity versus average power, showing difference between continuous and intermittent loads of the same average value. SOURCE: Martin et al., 2003. tent discharges, the capacity at a continuous discharge of the peak power is a better estimate than the capacity at a continuous discharge of the average power. The typical power demand of mobile systems usually includes several periods of peak demand interspersed with potentially long periods of very low demand. Figure 3-5 depicts such a measured power demand of a speech recognition system on a mobile platform. Each peak represents processor and disk activity while recognizing a sentence. HYBRID CONCEPTS A hybrid power source usually combines a high-energy/ low-power component with a low-energy/high-power component. Examples of hybrid power sources are battery + battery (e.g., Li ion + Zn/air), battery + capacitor (e.g., Li ion + electrochemical capacitor) and fuel cell + capacitor combinations. The rationale for hybrid power sources is to leverage the high-energy component with the high-power component, extending mission life and enhancing power capability while minimizing system weight or volume. Atwater et al. (2000) have demonstrated hybrid power sources with Li ion + Zn/air, Zn/air + electrochemical capacitor (EC), fuel cell + EC, and fuel cell + lead-acid combinations. In the Li ion + Zn/air hybrid, it was shown that the combined mission life (based on a communications equipment load profile) of the hybrid is almost six times longer than that of the individual components. In terms of specific energy, the hybrid had 198 Wh/kg, compared with 126 Wh/kg in the Li ion battery and 177 Wh/kg in the Zn/air battery. Optimization of a hybrid power source is very complex, and the optimized combination of power sources might enhance only for a certain range of load regimes. Thus, it is highly desirable to develop a model that can predict and analyze the performance of various combinations of power

OCR for page 27
Meeting the Energy Needs of Future Warriors FIGURE 3-5 Power profile of a user interaction with a mobile computer. Each of the spikes represents the translation of a sentence from English to Serbo-Croatian, at which time the processor and disk are concurrently operating at full performance. NOTE: TCP, transmission control protocol. SOURCE: Reilly et al., 2000. sources before one actually fabricates a hybrid. Further work should be carried out to develop models for various combinations such as battery + battery, fuel cell + battery, and fuel cell + capacitor hybrids. In addition, effort should be devoted to understanding critical factors such as self-discharge and temperature effects. It is well known that capacitors generally have a higher self-discharge rate than batteries. Thus, for a battery + capacitor hybrid, the capacitor will drain energy from the battery, reducing mission life. Similarly, Zn/air cells do not perform well at low temperatures. Thus, for a hybrid power source involving Zn/air as one of the components, the low-temperature effect might necessitate a bigger Zn/air cell to achieve the same performance as at room temperature. HYBRID ANALYSIS FOR THE SOLDIER SYSTEM The component electronics of the soldier system will operate on duty cycles that vary significantly in power needs, so there are varying needs for high specific power and high specific energy. Appendix C of Energy Efficient Technologies (NRC, 1997) presented data on the effective use of a battery + capacitor hybrid system when high but short-duration power demands are required by a system. These characteristics are usually not found in the same power source. An exception is the battery + capacitor hybrid, which yields system weight gains and improved performance when peak power demand is of short duration (10 milliseconds or less). When the peak power demand is of longer duration, a hybrid might combine a fuel cell and a battery. To highlight the benefits of such an approach, the committee analyzed the characteristics of a possible fuel cell + battery hybrid system using a direct methanol fuel cell (DMFC). While a system with any fuel could have been used for this analysis, a DMFC system was chosen because detailed full-load and part-load operating characteristics for such a system were readily available. A hydrogen fuel cell with comparable system-specific power, specific energy, and efficiency would also display benefits if used in such a hybrid combination. The direct methanol fuel cell is a device with a high specific energy. This is a direct result of the fact that it is a fuel conversion device, and the energy stored in a typical fuel is at least an order of magnitude higher than the specific energy available in batteries of even the most advanced chemistry. On the other hand, the typical battery has a capability for higher specific power level than the typical fuel cell. For these reasons, hybrid systems with combinations of power sources are very attractive to satisfy overall power needs, especially for long missions.

OCR for page 27
Meeting the Energy Needs of Future Warriors Applicability of Hybrid Technologies In the 1997 NRC report, several fueled hybrid energy systems were compared with current battery technology on the basis of source mass as a function of mission time in kilowatt-hours (NRC, 1997). In general, these calculations were done for roughly a 50-W system, and the most obvious conclusion to be drawn from the data is that there is a break point where it is more mass-efficient to use batteries than fueled systems. This is due to the dead weight of the converter, which is always present. Anytime the mass of batteries for a given mission is less than the dead weight of the converter, it is more advantageous to use a battery. As the kilowatt-hours for a given mission time increase, the dead weight becomes negligible compared with the fuel weight, and when that happens, if the converter is efficient, the fueled system is much less massive than a comparable battery. As the specific power of converters improves, the point at which converters are a more appropriate choice for a mission than batteries will move to shorter mission durations. This simple analysis is not all inclusive, and other factors must be considered when comparing single-type energy systems with fueled hybrids. The most energy-dense single-type source is batteries based on lithium technologies where specific energy is approaching 200 Wh/kg. Hybrids offer enormous advantages from a simple energetics point of view for longer mission times. Conversion at a modest 20 percent of the lower heating value (LHV) of the fuel leads to specific energy factors 2 to 5 times better than those of the best primary batteries. From a military standpoint, there are two important categories of hybrids: those that are air-breathing and those that are not air-breathing. In the latter category fall (1) the battery + battery hybrid (an extremely high specific energy and low specific power battery such as the lithium/(CF)x technology, coupled with a more conventional lithium battery having low specific energy but high specific power) and (2) the battery + electrochemical capacitor hybrid, successfully used in consumer electronics. The other category of hybrid (the air breathing) includes fueled and air-breathing combinations, perhaps coupling a rechargeable battery with a metal/air battery (such as zinc/air, aluminum/air, lithium/air), a motor generator (internal or external combustion driven), or a fuel cell (PEM/H2, DMFC, SOFC). Table 3-1 is a compilation of the advantages and disadvantages of each of the technology types for attributes of interest to military applications. Figure 3-6 analyzes the potential of such a hybrid system. The components chosen for an optimum system will depend heavily on the duty cycle demands of the soldier system combined and on the power system component characteristics. Commercial organizations that deal with these power management problems have simulation programs that use detailed system attributes for the potential power sources along with the details of the power sink demands and duty cycle of the system to be designed. These analytical programs are proprietary parts of their design group toolbox. Such high-fidelity simulations are key to their ability to introduce new products rapidly to the market. To highlight the value of such analyses, Figure 3-6 depicts the system weight requirement for three candidate systems for meeting the 72-hr mission requirements for a prescribed duty cycle. Since no actual duty cycles are available for the soldier system as yet, the duty cycle chosen was for a 20-W average demand cycle with periodic peak demand of 50 W effective 10 percent of the time. The system chosen for analysis was an advanced direct methanol fuel cell combined with an advanced rechargeable battery. The battery was sized to meet the peak power demand of the duty cycle (50 W). The fuel cell was chosen to meet the mean power level required for the duty cycle. Both devices were among the candidates that could be available on an intermediate time horizon for applications to soldier systems. Figure 3-6 shows that for this arbitrary 72-hr mission, the fuel cell + battery hybrid clearly outperforms either the battery or the fuel cell individually. The fuel cell weighed approximately 1.8 times as much as the hybrid system. The weight of the battery-only system was almost 2.5 times the weight of the hybrid system. The weights for the batteries were determined by using the target mission energy requirement. The limiting variable was that the required specific power had to be easily handled by the battery. For the fuel cell-only system, the dry system weight was assumed to be 2.5 times the 1.75 kg weight of a 20-W system, so that specific power was assumed to be similar for the 20-W and 50-W systems. Additional assumptions in the analysis were these: batteries were available in 0.5 kg sizes; fuel canisters for the fuel cell weighed 0.6 kg and supplied 800 Wh of energy for the full-load condition on fuel cell; for the 40 percent load condition in the fuel cell-only case, efficiency degrades, so that fuel canisters yield 612.5 Wh (Figure 3-2). Although the analysis above was for an arbitrary duty cycle, the 2.5 ratio of peak to mean power and the 10 percent time at peak power seem reasonable for the demand of typical soldier applications. The weight benefit accrues as a function of these two ratios and becomes larger as the peak to mean power ratio increases and the duration of peak power shrinks as a percentage of total time. Thus, benefits would disappear as peak to mean power approaches 1 or as the percentage of time spent at peak power approaches 100. Looking at Figure 3-6, it is apparent that for short missions the power source of choice based on system weight would always be a battery. As battery specific energy values improve, the length of mission for which they outperform other system alternatives will increase. Soldier systems operate in a unique environment, characterized by the extremes of dismounted warfare. They must be “ruggedized” to withstand physical punishment and not pose an extra hazard for the soldier under enemy fire; they must operate in various climes, including conditions of sand and dust; they must have simple controls, enabling such

OCR for page 27
Meeting the Energy Needs of Future Warriors TABLE 3-1 Comparison of Single Battery versus Hybrids for Attributes of Importance in Military Applications Factor Single Type: Battery Hybrid: Non-Air-Breathing (Battery + battery hybrids) Hybrid: Air-Breathing (Fueled System + Metal/Air Battery) Effects of environment Minimal. Typically operates from −30°C to +70°C. (However, low temperatures significantly reduce specific energy of battery unless it is warmed with use.) Disposal problems. Orientation-independent. Can be submerged. Typically operates from −30°C to +70°C. (However, low temperatures significantly reduce specific energy of battery unless it is warmed with use.) Disposal problems. Orientation-independent. Can be submerged. Needs preheating to operate at low temperatures. Some units are orientation-dependent. Minimal disposal problems. Performance altitude-dependent (but has been shown to work where humans work at >15,000 ft). Sensitive to dust and pollutants in air. Special precautions required for liquid immersion. Logistics Some restrictions on transport of lithium technologies. Many suppliers offshore. Readily available for civil applications. Some restrictions on transport of lithium technologies. Many suppliers offshore. Readily available for civil applications. Logistics dependent on fuel. Systems operating on logistics fuel are immature and at low technology readiness levels. Logistics not in place for other fuels such as H2, methanol, natural gas, aluminum, zinc, and carbon. Logistics infrastructure Logistics infrastructure in place to deal with lithium technologies. Logistics infrastructure in place to deal with lithium technologies. Logistics infrastructure would have to be developed to implement. If widely accepted, energy-efficient systems will reduce logistics burden. Versatility/utility Extremely versatile. Many sizes possible. Can be adapted to power almost anything. Enormous range of sizes and shapes. Has limited specific energy with little room for improvement. Must always have two units. Able to provide higher power to limit of power-dense unit. Duty cycle determines relative sizes of two units. Operates as a battery trickle charger. Operates primarily in a battery charger mode, but can provide power directly to load up to rated limit. Ultrarapid recharging of primary energy store. Must shut off all inlets and outlets if immersed. Some versions highly sensitive to dust and pollutants in air. Special procedure for low-temperature operation. Acoustic, thermal, and chemical signature problems. Three to eight times more energy dense for long missions. Safety Safe at low specific energy and low discharge rates. At high specific energy and high discharge rates, units may explode or rupture, dispersing toxic chemicals. Safe at low specific energy and low discharge rates. At high specific energy and high discharge rates, units may explode or rupture, dispersing toxic chemicals. Fuel-dependent safety. Reactants are separate. Some units are hot, presenting fire hazard if fuel spills. Not inherently explosive. Reliability Highly reliable. High specific energy/power units have safeguards to prevent explosive rupture events. Highly reliable. High specific energy/power units have safeguards to prevent explosive rupture events. Motor generator sets in civil and military applications have excellent ratings in larger sizes. Insufficient data to estimate reliability for small sizes and for various fuel cell systems. Manufacturability Large civil infrastructure currently manufactures batteries in an enormous range of sizes and shapes. Large civil infrastructure currently manufactures batteries in an enormous range of sizes and shapes. At present the electronic and software infrastructure to operate optimized hybrid systems does not reside in the military sector. Small motors and generators have established manufacturing infrastructure to produce large quantities. Little market for small fuel cells; hence manufacturing infrastructure is limited. Market demands will establish infrastructure. Availability Readily available in commercial sector. Military needs not always met (as for all energy technologies that are to be used for military applications). Special tooling and facilities may be needed; these come at a premium price. Having materials infrastructure helps reduce some costs of an integrated power system. But ultimately, packager and systems integrator must provide government with a military-class system based on those materials. Readily available in commercial sector. Military needs not always met (as for all energy technologies that are to be used for military applications). Special tooling and facilities may be needed; these come at a premium price. Having materials infrastructure helps reduce some costs of an integrated power system. But ultimately packager and systems integrator must provide government with a military class system based on those materials. Small motors and generators up to 1 to 2 hp are widely available at low cost. Stirling is emerging, but established market and manufacturing infrastructure exists for cooling applications. Fuel cells in this size range are one-off, with no market incentive to develop mass market manufacturing capability.

OCR for page 27
Meeting the Energy Needs of Future Warriors FIGURE 3-6 Soldier power demand for 20-W average, 50-W peak 10 percent of the time. Performance comparison for batteries alone, fuel cell alone, and hybrid battery + fuel cell. Data are based on Ball Aerospace 20-W DMFC and Li ion rechargeable battery and assume fuel cell system dry weight changes directly with peak power required and fuel packaging is the same for both large and small fuel cell. things as a stealth mode to evade detection; and they must be waterproof. For this reason, a key consideration in developing a hybrid system is the amount of time submerged with no access to air. Hybrids must be smart in that they can automatically close any air ports to protect against intrusion. These requirements would necessitate the inclusion of a battery or other non-air-breathing source to supply power when submerged operation is required. Battery + Battery Hybrid The committee was provided data on an experimental demonstration of the characteristics of a Zn/air + Li ion battery combination proposed for use in the Objective Force Warrior-Alternative Technology Demonstration (OFW-ATD), as shown in Figure 3-7 (Graham and Feldman, 2003). This comparison is flawed in that it does not consider the weight attributes of such a combined system. In general, battery + battery hybrids show an advantage over a single battery system only if the energy battery is incapable of meeting the power peaks required by the mission. This can be evaluated by calculating the specific power required of the energy battery to produce a given peak power, then dividing it by the weight of the energy battery, sized according to the total energy required for the mission. In an example provided by the Army for OFW, the estimated peak power required was 40 W (Graham and Feldman, 2003). The energy source was a Zn/air battery having a specific energy of 300 Wh/kg. A 24-hour mission would require 14.82-W average power for 24 hours, or 356 Wh, corresponding to a Zn/air battery weighing 1.095 kg. The peak specific power at which the Zn/air battery would be required to operate would be 40 W/1.095 kg, or 36.5 W/kg, which is well within the capabilities of a Zn/air battery. Thus, there would be no need for a high-specific-power battery for this particular mission. If the peak specific power required is much higher, then a high-specific-power battery may help. In this example, a more useful comparison would be the discharge time for a Zn/air battery of the same weight as the

OCR for page 27
Meeting the Energy Needs of Future Warriors hybrid system. Here there would be no significant difference in discharge times, though the voltage dip during high power pulses would be smaller for the hybrid. Table 3-1 illustrates the differences between single and hybrid sources for several performance categories. In addition, effort should be devoted to understanding critical factors such as self-discharge and temperature effects. It is well known that capacitors generally have higher self-discharge rates than batteries. Thus, for a battery + capacitor hybrid, the capacitor will drain energy from the battery, shortening mission life. Similarly, Zn/air cells do not perform well at low temperatures. Thus, for a hybrid power source involving Zn/air as one of the components, low temperatures might necessitate a larger Zn/air cell in order to achieve the same performance as at room temperature. SYSTEM CONFIGURATION CHOICES The choice of system elements must consider the specific characteristics of both the energy source and energy sink elements of the system. The wide range of energy sinks includes standard computer and display hardware using powers of milliwatts to watts; laser target designators demanding 100 W or more; soldier cooling hardware demanding tens of watts continuously; and exoskeletal devices that demand very large amounts of power and energy. The proposed sources of energy include primary batteries (with high energy densities and modest internal ohmic resistance); rechargeable batteries (with lower energy density and lower ohmic resistance); fuel cells (with low energy output but high energy density related to their use of high-energy fuels); and engine-driven generators (with high outputs and high energy densities but problematic noise and heat signatures). Hybrid systems have the ability to improve system power usage by limiting the voltage drops that are imposed on the power supplies for higher current demand duty cycles. This ability reduces the heat generated by ohmic resistance in the power supply. Additionally, such hybrid combinations can avoid the transitions into low effectiveness operation of one of the power supply components. See Figure 3-7 for a comparison of data for the Zn/air battery curve as one example. These power sources might be used at different times or in combination to best satisfy soldier needs. To determine the most appropriate combination for satisfying a soldier’s FIGURE 3-7 Performance of hybrid as compared with performance of single components in power load cyclic profile of 9 min, 12 W, and 1 min, 40 W. Hybrid’s gain is approximately 4 hours. SOURCE: Graham and Feldman, 2003.

OCR for page 27
Meeting the Energy Needs of Future Warriors mission needs, one must consider source and sink characteristics along with mission requirements and duty cycles. Any two sources may be combined into a hybrid to satisfy soldier system needs. As described previously, the combination of rechargeable batteries and fuel cells might be used to meet periodic high current demand and high energy needs in combination. The combination of an air-breathing generator and rechargeable batteries might be better for a mission that requires a soldier to be immersed in water. Matching Source with Sinks The Army has defined several mission scenarios, and each must be validated so that appropriate systems can be selected confidently. Before any system is selected, it is important that the combination of energy source, energy sink, and soldier mission requirements and duty cycles be considered jointly. Comparing power source performance metrics under identical load conditions and operational scenarios allows for the best assessment of energy alternatives. This requires knowledge of the power demand and time data for every piece of equipment for every soldier for a statistically significant number of missions. The OFW-ATD Program includes a modeling effort that predicts power demand by positing mission scenarios, estimating duty cycles, and using power sink specification data for all components of the Land Warrior soldier system. In addition, the OFW Program plans to monitor the power demand of specific components during actual or simulated missions. The information gathered will be used to validate the models and provide realistic boundary conditions for total energy, average power, peak power, and duty cycles for various missions. The validated models should lead to more effective planning and designs. For example, the optimal suite of energy storage and energy conversion devices, fuel quantities, etc., could be determined for each mission. MODELING REQUIREMENTS Commercial developers of power systems simulate the power use of their systems so they can rapidly optimize hybrid system choices. These modeling efforts require detailed descriptions of the power supply characteristics of the candidate components and are considered to be key proprietary parts of their in-house design processes. The dynamic response characteristics of components—that is of the available battery types, capacitors, and fuel cells—as a function of current flow rate and ambient conditions must be known to obtain accurate results. The Army will need a similar modeling approach in order to make appropriate system design choices. The type of duty cycle encountered in real field applications is key to acquiring an optimized design and must be determined through experimentation. The Army should invest in such a modeling capability, which would be essential for effective power management at the system power input level. This capability could be combined with other power management capabilities focused on system power output and power demands. Researchers at the University of South Carolina have developed modeling software known as the Virtual Test Bed (VTB), with the goal of optimizing the usage of charge storage devices for specialized applications. They can input the parameters for general battery systems and then study how the battery will perform under specified loads. Software can also be used to model hybrid power sources, where a battery is used for low power and an electrochemical capacitor is used for pulsed power. With this approach, the South Carolina team successfully improved power utilization for a device that utilizes a hybrid system (Dougal et al., 2002). The results are largely nonintuitive, and extensive modeling was needed to identify the optimum power source. It is clear to the committee that high-fidelity modeling will be needed to optimize the Army soldier system. For models to be useful, the power and energy usage of a mission must be specified. However, until the OFW system has been put in place and usage data gathered, the power and energy inputs to these models are not available. The mission requirements are especially critical when pulsed power is needed, in which case a battery + capacitor system might be useful; also, the duty cycle of the pulsed power must be known. For instance, a minute-long pulse can be delivered efficiently with a battery, but a capacitor + battery hybrid power source might be better for a device that senses at low power and then transmits data using a millisecond pulse. Because engineers of military equipment are often looking to adapt new technology, they resist modeling efforts until they have their system design completed. Although the modeling effort can be time consuming, it should not be delayed. One challenge with the modeling approach is to develop code with general rules that can be rapidly adapted by systems engineers to help guide their choices toward components that may save power in the overall system. Also, the models need to be able to take into account the behavior of real systems—that is, systems that fade in performance over time or have a range of performance values. In summary, modeling has the potential to save time and money in the development of efficient portable electronic systems if accurate system inputs can be supplied. The modeling can complement experimental data as it narrows down the parameters of optimization. Any power solution ultimately needs to be verified with experimental data, but modeling can expedite selection of the power source. Ideally, the military should develop and acquire new equipment based on recommendations and considerations gained from power sources modeling, so that the lifetime of the equipment can be maximized.