As discussed in earlier chapters, one of the most important steps in ensuring the electric power delivery system’s resilience to terrorism is to ensure that it is as resilient as possible against more routine disturbances, that it can be rapidly restored if and when a disruption occurs, and that while the disruption is in progress, the impact on critical services is as modest as possible. The committee has concluded that, with a few notable exceptions, there is relatively little R&D that can be targeted just at terrorism, but that much that is intended to improve operations also will help against terrorism. Many of the most promising technologies under development for improving the power system may not harden it against terrorist attack, but they often will improve grid resilience and response and recovery. This chapter assesses research needs for reducing the risk from terrorist attacks in the context of overall power delivery system needs. It also notes alternative strategies by which the electric power system could be guided to greater robustness.
As discussed in Chapter 2, recent decades have witnessed chronic underinvestment in sustaining and upgrading the U.S. transmission and distribution system. The same has been true for research investments. Funding for R&D is also addressed in this chapter.
This chapter addresses R&D needs to meet the three goals discussed in previous chapters:
• Reducing vulnerability to terrorist attacks (Chapter 6); and
Because the electric power system is one of the most complex systems every built, R&D programs to improve it are understandably complex as well. No one or two items will solve the problem of protecting against terrorism, mitigating impacts, and supporting recovery, although certain priorities can be identified.
Physical attacks on the bulk power system1 and on critical components of the distribution system can cause widespread, potentially long-term outages. Thwarting such attacks involves developing physical security and sensing technology that enhances the robustness of the system to physical attacks on various components of the power system and provides adequate early warning.
Improved means for countering cyber attacks also are needed and can be furthered by research to ensure secure communications, protect the energy management systems (EMSs) that control the bulk power network, and enhance the development of distribution management systems (DMSs) for controlling the distribution system. A wide range of intelligent electronic devices, relays, and controls at substations (primarily at the distribution system levels) are potentially vulnerable because they can be accessed remotely via several different types of communication networks.
Reducing vulnerability and enhancing resilience involve modifying the electric power system to better manage the loss of key components. R&D can provide a variety of options for enhanced monitoring, reduced system stress, improved reliability, incorporation of advanced technology, specific components, efficient demand-side management, and the use of distributed energy resources.
1It is again noted that the term “bulk power system” generally applies to large central generation stations and those portions of the transmission system operated at voltages of 100 kV or higher.
Any physical or electrical disturbance affects the performance of the electric power system. Therefore, advanced emergency control techniques that would adjust disrupted power flow to an acceptable operating state would make the system more resilient to malicious attacks. Particularly important is the development of improved tools and strategies that allow a more nuanced real-time treatment of which loads are and are not served during restoration.
Reducing the impact of an attack (and its consequences) involves developing and using advanced network technologies and control features at both the bulk power system level and the distribution system level. Distributed energy resources could also play a significant role in minimizing power disruptions to customers, powering critical services and facilities, and facilitating restoration. Several concepts in this area involve the expanded use of combined heat and power technology, distributed generation, and micro-grids. Such technologies already are in use but not fully deployed. However, considerable research focused on hardware, control systems, control policy, and the impacts of alternative regulatory arrangements is needed to enable resolution of technical and regulatory impediments to integrate such resources into the overall system.
The extended loss of electric supply due to a malicious attack could have a significant impact on several interdependent civilian infrastructure systems,2 including water treatment and pumping facilities, sewage treatment plants, transportation, communication systems, gas pipelines, and traffic control systems. Although studies have qualitatively evaluated the impact of the loss of power supply on specific systems, they have not, for the most part, considered all interdependent systems collectively. Moreover, in most regions, efforts have not been made to investigate and model the impacts of a long-term curtailment of the electricity supply. A critical aspect of system interdependencies is that official policies will be needed to coordinate these systems, establish hierarchies in terms of responsibilities and control following an attack, enunciate a clear public message, and continuously update information in a coordinated fashion.
The need for a well-coordinated, automatic or semiautomatic plan for restoring the electric system after a coordinated malicious attack has been a topic of intense discussion in the electric power industry. North American Electric Reliability Council (NERC) guidelines require every region to have such a plan. Automating recovery to reduce the possibility of human error, however, is an enormous task requiring significant investment in research toward developing techniques to coordinate various options and develop decision-making tools.
This section discusses a wide range of specific technologies for which R&D is promising. They are grouped into eight technology areas according to how they will benefit the power system.
Increasing the power flow capacity of transmission lines can increase security because it provides greater ability to bypass a damaged line in delivering power from generating stations to load centers.
The transfer capability of some transmission circuits can be increased by raising the operating voltage and reconfiguring conductors into a more compact arrangement on existing rights-of-way.
New, recently developed conductors having composite cores or using aluminum alloys have higher current-carrying capability than conductors in general use. Under high rates of power flow, they have less mechanical sag at high temperatures because of lower thermal expansion as compared to typical conductors with steel cores. Reducing the sag of a loaded line allows greater loading of lines, although increased thermal capacity, if not used properly, can place more stress on the power system.
High-temperature Superconducting Cables
High-temperature superconducting cables can potentially carry three to five times as much current as conventional cables of the same size, but considerably more research is required before these cables can be made technically successful and ready for widespread use. Although initial assessments indicate that such cables are very complex and expensive special-purpose devices with limited applications, they nonetheless deserve consideration.
New composite materials that are inherently insulating and corrosion-resistant could potentially replace metals in the support structures for substations and transmission lines and could also allow for reconfiguring existing rights-of-way to increase power flow. Many complex issues still have to
be addressed regarding their selection and application and to reduce costs.
Greater control of energy flows reduces the risk of cascading failures and may speed restoration of power after a major outage. Medium-voltage (4-13 kV) and high-voltage (>69 kV) high-power electronic-based controllers can provide flexibility and speed in controlling the flow of power over transmission and distribution lines. New energy storage units can help level loads and improve system stability. Some specific examples of these equipment technologies are given below.
Flexible AC Transmission System (FACTS) Devices
High-voltage power electronic-based controllers are currently being demonstrated. FACTS controllers can increase the power transfer capability of transmission and distribution lines and improve overall system reliability by reacting almost instantaneously to disturbances. The unified power flow controller and the convertible static compensator are key examples of FACTS technology. They control both real and reactive power flows among transmission corridors and maintain the stability of transmission voltage. More research, design, and development is needed to reduce the cost and enhance the performance of FACTS technologies. The next steps should include the development of the fourth generation of FACTS controllers using advanced power electronics devices.
Advanced Power Electronic Devices
The next major step in the development of power electronic devices would be to replace the silicon-based thyris-tors used in current devices with thyristors based on wideband-gap semiconductor materials, such as silicon carbide, gallium nitride, or very-thin-film diamond materials. These materials have the potential to reduce the cost of power electronic controls.
FACTS Integrated with Storage
Fast-response devices for energy storage could be used with FACTS controllers to provide ride-through capability for transient and brief outages. One promising technology, superconducting magnetic energy storage (SMES), responds to disturbances in less than one AC cycle and provides ride-through capability for multi-second outages. Research is needed to adapt the high-temperature superconducting materials described above for cables for use in the high-field SMES environment, potentially lowering the cost so that these units can be used to support the electric power transmission system.
Voltage-sourced converters can be used to connect independent asynchronous AC transmission systems. Other thyristor-based controllers can supply reactive power (i.e., volt-ampere-reactives) for voltage support and reactive power management in transmission systems. Connection of systems that now cannot be connected might lead to increased power flow.
Intelligent Universal Transformers
The intelligent universal transformer concept involves a state-of-the-art power electronic system and is not a transformer device in the traditional sense. It would be designed to replace conventional transformers with a power electronic system that steps voltage as traditional transformers do, but can also manage and control consumer demand and power flows, and compensate for reactive power.
Substantial improvements in the cost and performance of sensors and communications media and equipment offer the prospect of increasing the capacity of existing power system facilities by monitoring and compensating for the operating conditions of numerous devices simultaneously. Examples include the following.
Integrated Communication Architecture
Overlaying a communication architecture on the existing power delivery system could be a foundation for enhancing the functionality of the power system and, therefore, its resilience. This requires an open standards-based systems architecture for an infrastructure for data communications and distributed computing. Several technical elements of this infrastructure include, but are not limited to, data networking, communications over a wide variety of physical media, and embedded computing technologies. Challenges remain in fully deploying such an architecture while meeting cyber security challenges.
Wide Area Measurement System
The Wide Area Measurement System (WAMS), based on a combination of satellite communications employing time-stamping with fiber or wireless, will provide the real-time information needed for integrated control of large, highly interconnected transmission systems. By constantly monitoring the health of a network across a wide geographical area, WAMS can detect abnormal system conditions as they arise.
Dynamic Thermal Circuit Rating Technology
Dynamic thermal circuit rating (DTCR) technology can be used to increase the thermal loading on individual transmission lines. Present limits are both static and often conservative, based on assumed weather conditions. DTCR uses real-time information about weather, load, temperature, line tension, and/or line sag to estimate actual thermal limits, allowing higher thermal capacity of lines. Certain DTCR devices are commercially available, and others are currently being demonstrated on a few transmission systems.
Video Sag Monitoring
Direct monitoring of line sag can be used to extend the effectiveness of DTCR even further. A video “sag” meter has been prototyped that uses a digital camera mounted on a transmission tower to monitor the vertical position of the line. Sag monitoring is listed separately here and not included under the broader title of remote video monitoring of critical components because it enables dynamic operation.
Topology estimators can be used to accurately determine the real-time transmission grid configuration status of an interconnection. Accurate information on topology is necessary for accurate state estimation and the subsequent security-constrained dispatch that is the key computation for solving congestion problems.
Improved Simulation and Modeling
Faster-than-real-time simulation and improved modeling would enable very rapid computation of the power condition’s status, and in turn:
• Faster-than-real-time, look-ahead simulations of operating conditions;
• What-if analyses from both the operations and the planning points of view;
• Integration of risk analysis into system models and quantification of effects on system security; and
• Through the use of advanced simulation, pattern recognition, and diagnostic models, determination of the location and nature of suspicious events.
Monitoring of Constraints
Sensor output, communication, and computation can be used in combination to monitor the effect of transmission constraints on wholesale power market activities. Operating in a limited fashion for the Eastern Interconnection, this capability could be enhanced to include probabilities of line outages on which a probabilistic reliability index could be based.
Database Protocols Development
A common information model is needed for transmission and distribution operations and maintenance databases. It would support interoperability by greatly reducing the number of needed software translators in situations involving a range of applications.
Growth in the demand for electric power in dense urban areas will continue to challenge the capacity of the traditional medium-voltage underground network grids installed in most large cities to provide reliable power. To meet projected increases in demand while still providing safe, reliable, and affordable power, utilities will have to reconfigure networks and minimize secondary (low-voltage) cable. Technology options include the following:
Submersible (Underwater) Fast Switches
Fast switches enable connection of customers to alternate power sources during system reconfiguration, and a capability for reconfiguration at medium voltage minimizes the impact of a catastrophic event at a single power station or circuit. In underground networks where flooding is possible, there is a need for switches that can operate underwater while still energized so as to mitigate outages.
Low-voltage Switches and Smart Fuses for Isolation
Low-voltage devices such as automated breakers (switches) and smart fuses that respond to appropriate rise times allow for reconfiguration and isolation of faulty sections of the low-voltage grid network.
The technologies discussed in this chapter will contribute to enhanced security of the electric power system even though that is often not their primary goal. Technologies specifically intended to improve security will, in most cases, provide significant benefits in the face of major equipment failures resulting from natural disasters as well as terrorist attacks.
Probabilistic Vulnerability Assessment
A key priority among efforts to improve overall system security is to assess power system vulnerabilities to terrorism and identify the most effective countermeasures. Probabilis-
tic vulnerability assessment is a framework for objectively identifying the most significant threats to the electricity supply chain and assessing the relative cost-effectiveness of various potential solutions. The probabilistic methods developed in this effort will also provide the basis for improved assessment of risks encountered during normal power system operations.
Emergency Control and Restoration
Following a major terrorist attack or natural calamity, a system is needed to focus the initial response on prevention of cascading. Wide-area control and the use of fast-acting autonomous agents may create self-sufficient “islands” that can maintain power within a large blacked-out area.
Complex Interactive Networks
Continuation of R&D on complex interactive networks would enable analysis of information about the status of the power delivery system and the secure communications system after an attack, as well as coordination of their use for adaptive islanding. Once a stable configuration of power delivery system islands is established, algorithms for self-healing would gradually return the power delivery system to its normal state as more resources became available.
Most sensing and control agents in a power system today simply respond to changing local conditions according to preprogrammed instructions. Enhanced intelligent network agents (INAs) would have decision-making capability, based on internal analysis of network-wide conditions. Once implemented, INA technology would facilitate adaptive islanding and the smart power delivery system, which is among the technologies described below.
Smart Power Delivery System
The smart power delivery system would contain transmission-class fault anticipators tied to a network of distributed data processors communicating with regional operations centers, allowing simulations to be run to determine optimal corrective responses to any disruption. When attacks occur, a network of sensors would instantly detect a voltage fluctuation and communicate this information to intelligent relays and other equipment located at substations. These relays would automatically execute corrective actions, isolating the failed lines and re-routing power via power electronic-based controllers to other parts of the power delivery system. Many consumers would be unaware that a disruption had occurred. Additionally, advanced system analysis will allow utilities to determine reliability metrics based on probabilistic techniques, which would lead to improved asset utilization.
Integrated Asset Management
More sophisticated maintenance procedures will be vital to hardening the power system and ensuring the reliability of increasingly complex transmission networks. Software is needed that would interpret the raw data coming from real-time monitors into the critical information needed by system operators.
Integration of Distributed Energy Resources
There is a need to develop interconnection standards and requirements related to integrating distributed energy resources with power delivery systems. The effect of distributed resources on system performance, especially at high penetration rates, also needs to be determined.
Real-time analysis of system stability and security will be needed to properly detect a multi-pronged terrorist attack or a sequence of other natural or man-made disasters. Online analytical tools are needed that will take this information, such as the data available from WAMS, and determine automatically what actions should be taken to prevent incipient disturbances from spreading. Meeting real-time system control requirements will require completing such analysis in a fraction of a second. Power system visualization would improve operator situational awareness, allowing a faster response to rapidly deteriorating situations.
Solid-state and Superconducting Fault Current Limiters
Unless carefully planned, the location of generation on a given power system can pose a risk of short-circuit currents that are dangerous to utility field personnel and may cause considerable damage to the power system. Fault current limiters would use either power electronics or superconductivity to limit short-circuit currents. These solid-state devices not only would act as a circuit breaker, but also would act in milliseconds to limit fault current levels.
Solid-state Power Electronic Circuit Breakers
Solid-state breakers will allow the system of the future to respond more quickly to disruptions and terrorist attacks.
As noted in Chapters 3 and 8, the large power transformers in generating station switchyards and major substations are vulnerable to terrorist attack and could take months or years to replace. Options for bypassing damaged substations to bring power from remote generating stations to load centers are very limited because the grid is already stressed
during peak demand. The result of a coordinated attack on key substations could be rolling blackouts over a wide area until the substations are repaired.
Under such conditions, the availability of compact, easily transported recovery transformers would be invaluable. Recovery transformers would be usable for a variety of applications to replace the large power transformers optimized for a particular substation. They would be smaller for easier transport and relatively inexpensive. They would also be less efficient and therefore more costly to operate, and so would be used only until a regular replacement is available.
Recovery transformers need further development and testing. Then a reasonable supply of them would have to be manufactured and stored in strategic U.S. locations for use to recover as quickly as possible from any widespread disaster affecting a large part of the electric transmission infrastructure (see Chapter 8). The increased standardization of substation transformers being embraced by utilities will facilitate use of these recovery transformers.
Chapter 3 detailed a set of very-near-term developments that relate to physical security, including advanced design and engineering steps to harden substation sites and to make key components less vulnerable, improved sophisticated electronic surveillance technology that integrates sensor and monitoring, and security systems for high-voltage submarine cables.
An array of R&D opportunities exist related to consumer products for enhancing the public’s resilience to terrorism, particularly in residential and urban settings, but these are not considered within the scope of this report and so are not addressed here.
Making the nation’s power system truly secure from disasters will require true consumer connectivity that includes the optimization of end-use devices. Means for achieving this include those outlined below.
Demand-side management (DSM), which is defined as the further deployment and utilization of energy-efficient electric end-use devices and greater use of consumer load control, will also be critical to complement the supply options inherent in a secure power system. DSM includes the ability to dispatch both loads and distributed energy resources. A variety of new communications and customer-interface technologies will be needed to enable load control to complement the options available for response to security concerns.
Advanced Distribution Automation
Advanced distribution automation (ADA) is defined as distribution monitoring and control, distribution system management, and consumer interaction (e.g., load management, “smart” metering, and real-time pricing). ADA will enable real-time optimization, such as operating distributed energy resources when other facilities have been compromised. Two developments are needed to make ADA a reality: (1) an open communication architecture and (2) a redeveloped power system from an electrical architecture standpoint. ADA will use various advanced technologies, including communications systems, distributed computing, embedded system computing, sensor and monitoring technologies, and power-electronics-based components.
Self-healing Control Methodology for Distribution Systems
For the distribution system to be secure, it is essential to enable distribution system monitoring through a web of sensors integrated with an overall control methodology to respond to terrorist attacks and reduce the duration and impact of failures through a self-healing methodology.
A series of web-enabled, inexpensive sensors that can be linked to global positioning satellites would allow higher levels of control of control.
High-speed, online sensors are needed for detecting distortions in the 60-cycle power line carrier. Waveform distortions need to be correlated with early indicators of system component failure. Pattern recognition software is needed that will analyze the power line waveform and detect pre-failure indicators in real time.
Although the technologies described below are not directly related to addressing threats from terrorism, they would collectively reduce the stress on the electric system infrastructure and thereby contribute to its resilience in the face of attack.
Much of the artificial illumination in place today is considerably less efficient than theoretically, or even practically, possible. Increased use of high-efficiency lighting systems
that combine efficacious light sources with luminaries that effectively direct light where it is desired, coupled with controls to adjust light levels as needed, will collectively improve overall lighting efficiency.
Efficient Space Conditioning (Building Heating and Cooling)
Considerable progress has been made in the last few decades toward improving the efficiency of space conditioning equipment. Much of the progress is due to state building codes and federal standards that dictate the minimum efficiency of new air conditioning systems. More opportunities exist to further enhance the efficiency of heating and cooling systems and thus reduce demand for electric power.
Efficient Domestic Water Heating
Electric water heaters lose heat through tank walls and piping. Research is needed on newer systems that produce hot water on demand, thereby eliminating the storage tank and its associated losses of heat. In addition, R&D is needed on (1) heat pump water heaters that can utilize heat from the surrounding air to heat water while providing cooling and dehumidification of the surrounding room air space, and (2) systems that recover waste heat from air conditioning systems.
Distributed generation (DG), micro-grids, and other distributed energy resources technologies can augment the large central power generators of the present-day electric power delivery system. Energy conversion efficiencies for DG technologies are still substantially below those for conventional generation technologies. However, it is often possible to use the waste heat in industrial processes, an approach known as combined heat and power (CHP), boosting overall efficiency to high levels (e.g., 75 percent). Key DG technologies requiring R&D are intelligent control systems, high-efficiency internal combustion engines, microturbines, fuel cells, and Stirling engines. Also needed is R&D on CHP for residential applications, photovoltaic devices and low-cost “balance of system” electronics, solar-thermal systems, and building-integrated and concentration solar systems.
Electric Energy Storage
Electric energy storage refers specifically to a capability for storing already-generated electrical energy and controlling its release for use at another time. Most electrical energy storage systems have demonstrated efficiencies of between 60 and 70 percent, a level that must be improved significantly to make applications such as load leveling feasible. Key energy storage technologies requiring R&D are lead acid batteries, nickel-cadmium batteries, nickel-metal hydride batteries, lithium-ion batteries, vanadium redox flow batteries, sodium-sulfur batteries, flywheel energy storage, ultracapacitors, miniature compressed air energy storage, and superconducting magnetic energy storage.
The technologies discussed above are correlated in Table 9.1, with the goals to which they may contribute: thwarting attacks, reducing vulnerability, and reducing the impact of prolonged outages. Although relatively few technologies are listed directly for thwarting attacks, reducing vulnerability to and the impacts of attacks also reduces terrorists’ incentives for attacking the power system. Therefore to some extent, all the technologies listed in Table 9.1 will contribute to thwarting attacks.
The committee was assisted in the selection of these technologies by the advice of many experts in industry, academia, and research institutions whose views were solicited in a widely circulated questionnaire. This exercise and the results are described in Box 9.1. The full list of promising R&D projects considered in the questionnaire is shown in Appendix H.
The committee believes that the following should have the highest priority in the mid- to long-term time frame:
1. Development, demonstration, and deployment of high-voltage recovery transformers;
2. Development and demonstration of the advanced computational system intended to support more rapid estimation of system state and broader system analysis;
3. Development of a visualization system for transmission control centers to support informed operator decision making and reduce vulnerability to human errors;
4. Development of dynamic systems technology and demand response demonstrations to allow interactive control of consumers and consumer loads;
5. Development of multilayer control strategies that include capabilities to island and self-heal the power system; and
6. Development of improved energy storage that can be deployed as dispersed systems.
The market is very good at commercializing well-developed basic technology ideas. However, many of the ideas discussed in this chapter are not yet at the stage that they can be readily turned into operating hardware or systems. The earlier the stage of development, and the longer the interval from idea to commercial application, the lower the prob-
|Thwart Attack||Reduce Vulnerability||Reduce Impact|
|Technologies that allow significant increases in power flow||
• Reconfiguring conductors
• High-amperage conductors
• High-temperature superconducting cables
• Composite structures
|Equipment that allows greater control of energy flows||
• Flexible AC transmission system (FACTS) devices
• Advanced power electronic devices
• FACTS integrated with storage
• Voltage-sourced converters
• Intelligent universal transformers
|Advanced monitoring and communications equipment||
• Integrated communication architecture
• Wide-area measurement system
• Dynamic thermal circuit rating technology
• Video sag monitoring
• Topology estimators
• Improved simulation and modeling
• Monitoring of constraints
• Database protocols development
|Technologies that enable increased asset utilization||
• Submersible (underwater) fast switches
• Low-voltage switches and smart fuses for isolation
|Technologies that are particularly intended to enhance security||
• Probabilistic vulnerability assessment
• Emergency control and restoration
• Complex interactive network
• Smart power delivery system
• Integrated asset management
• Integration of distributed energy resources
• Real-time analysis
• Solid-state and superconducting fault current limiters
• Solid-state power electronic circuit breakers
• Recovery transformers
• Physical security technologies
|Technologies that enable greater connectivity and control||
• Demand-side management
• Advanced distribution automation
• Self-healing control methodology for distribution systems
• Low-cost sensors
• Pre-failure indicators
|Technologies to reduce demand on the power system||
• Efficient lighting
• Efficient space conditioning (building heating/cooling
• Efficient domestic water heating
|Distributed energy technologies||
• Distributed generation
• Electric energy storage
ability that conventional market forces will result in research and development being done. Society funds longer-term fundamental research as a way to provide options for the future. However, the question of how much society should invest in research to develop basic ideas to protect the electric power system from terrorist threats is difficult to answer for three reasons:
1. Because the probability of terrorist attacks on the power system, the magnitude of such attacks, and the likelihood of success are all unknowable, it is impossible to calculate accurately what the benefits of R&D might be.
2. It cannot be known beforehand what new technologies and options research will make available.
3. As indicated above in this chapter, most investments in power delivery system research would serve broad needs, not just the need to protect the system from terrorist attacks. Even those investments that are most antiterrorism-specific have other beneficial aspects (e.g., recovery transformers could be moved quickly to a stricken area after a large earthquake or hurricane).
In view of these considerations, the best that can be done is to develop some order-of-magnitude arguments concerning research investments. The committee was unable to find any rigorous estimates of the national impact of prolonged blackouts resulting from terrorist attacks. In Chapter 1, the committee concluded that a sophisticated terrorist attack could cost hundreds of billions of dollars, mostly from the loss of economic activity while power is unavailable.
Over the next decade, a well-designed research program could result in knowledge and technology that could significantly reduce the cost of a large, long-term blackout caused by terrorist attack. This is particularly true if that research also included some of the strategies discussed in Chapter 8 that would make critical social services less vulnerable in the face of disruption of electrical supply. The committee has not been able to develop meaningful quantitative estimates of the probability of attack. However, a simple parametric assessment can help to bound the potential value of R&D undertaken to reduce the power delivery system’s vulnerability to terrorist acts, as shown in Figure 9.1. For example, suppose that over the coming decade, there is a 1 in 100 chance that a large coordinated terrorist attack on the electric power delivery system could impose societal costs of the order of $100 billion. A 1 in 100 chance of a loss of $100 billion can then be represented as an expected loss of $1 billion (gray horizontal line) in Figure 9.1. If a research investment over that same decade could reduce losses from such an attack to $10-billion and the cost of deployment of the new technology and systems could be supported as meeting the conventional needs of the system, then the value of the research in this case could be roughly $100 million (see gray curved line at vertical axis).
Of course, new technical knowledge alone is not sufficient. Knowledge must also be put to work in the form of deployed systems. Those investments are typically much larger than the investments required to do the research. However, given the conclusion reached above in this chapter— that to a first order, much of the research needed to better prepare to deal with terrorism is very similar to the research needed to make general improvements and upgrades to the power delivery system—much of the cost of implementation might well be justified by other societal needs.
In 2004, the Electric Power Research Institute (EPRI) did an extensive analysis of the costs of making all of the improvements needed to deploy the advanced technologies detailed in this chapter. (EPRI, 2004). EPRI estimated that the power sector was spending about $18 billion per year (in 2004 $) on capital investments in the transmission and distribution system and that an additional expenditure of $165 billion over 20 years, or $ 8.3 billion per year, would be needed to fully deploy the technologies relevant to enhancing the resilience and functionality of the power delivery system. To develop these technologies so that they are available for deployment, the committee believes that an additional R&D investment of approximately 10 percent of those additional expenditures, or $800 million per year, over current funding levels would be needed. This amount is in addition to investments in R&D targeted at power generation or environmental sciences.
In gathering input for chapter 9, the Committee on Enhancing the Robustness and Resilience of Electrical Transmission and Distribution in the United States to Terrorist Attack prepared and circulated a questionnaire to industry and academic experts in transmission and distribution R&D needs, including several members of the committee. The questionnaire first asked respondents to allocate a research budget1 across the research areas shown in Table 9.1 and then across the technologies listed for each area, “considering all the needs and objectives of the U.S. electric power transmission and distribution system.” Respondents were asked to think about “(1) the importance of the area to the future operation of the U.S. electric power system, and (2) how easy it would be to make progress in each area (i.e., the marginal returns per R&D dollar invested).” After completing the first part of the questionnaire, respondents were asked to go through the same tasks again, this time considering “only the need to improve the security and reliability of the U.S. electric power transmission and distribution system.” Respondents were asked to rate the technologies listed in Table 9.1as to their potential importance in enhancing the resilience of the nation’s power delivery infrastructure.
Based on responses to the questionnaire, the following technologies were viewed as high-priority R&D goals by most experts:
• High-voltage recovery transformers;
• Systems to improve operator awareness and system visualization;
• Advanced demand response based on dynamic systems;
• Multi-layer control strategies;
• Distributed control and recovery;
• Distributed generation and micro-grids;
• Low-cost undergrounding techniques;
• Physically robust/resilient poles, conductors, etc.;
• Solid-state transformers;
• Smart meters;
• Distribution power electronic devices;
• Advanced relaying and protection;
• Advanced failure detection and location; and
• Improved distributed storage.
Most of the R&D priorities identified by questionnaire respondents showed little differentiation between those needed for improving today’s system without a specific focus on the risk of terrorism and those identified with such a focus. However, when the focus was countermeasures to the risk of terrorism, the following emerged as clearly more important:
• High-voltage recovery transformers,
• Systems to improve operator awareness and system visualization,
• Advanced demand response based on dynamic systems,
Judicious investments in research and development of pertinent technologies can help to enhance the quality of human life and better serve society’s needs, as well as reducing the costs of increasing the capacity of the transmission and distribution systems to handle increasing loads. A balanced, cost-effective approach to investment in R&D and to the subsequent use of technology can make a sizable difference in mitigating risks.
R&D on electricity transmission and distribution in the United States is conducted by a variety of organizations. The U.S. DOE has a significant effort aimed at a select group of technologies, primarily concerning electric power transmission technologies, and especially focusing on superconductivity for cables and short-circuit current limiters. EPRI has a substantial effort, funded both by U.S. utilities and by institutions from as many as 30 other countries. Other national efforts, supported by DOE, EPRI, utilities, and several equipment suppliers, are carried out through organizations like the Power System Engineering Research Center (PSERC) and the Consortium for Electric Reliability
• Multilayer control strategies (including capabilities to island and to self-heal), and
• Improved distributed storage.
In general, while respondents acknowledged that improving end-use energy efficiency would reduce stress on the electric power infrastructure, they nearly uniformly felt that R&D related to reliability, demand response, control, hardening the system, and recovery had priority over reducing the stress on the system by decreasing demand through enhancing efficiency.
In addition to the needs described above for development of technologies, systems, and software, respondents identified several other “nonhardware” research topics.
Public Perception of Risk
Although there is considerable literature about the general public’s perception of risk, very little research has been done on reactions to blackouts, whether caused by natural disasters, equipment failure, or terrorism. A team of researchers could be prepared to be deployed within hours following such an event. One goal of the work would be to develop protocols for responding effectively to major disruptions of the power supply, so that the public would be kept informed and made aware of constructive steps to take.
Lessons Learned from Blackouts
A research team organized by the Department of Homeland Security (DHS) and deployed following blackouts could learn about efforts made by utilities, government officials, business leaders, and others to respond in resilient ways.
The organizational structure employed and the effort made to communicate with the public following a significant terrorist attack on the power system need to be addressed. In particular, managing the public response to distress could contribute substantially to mitigating the loss of life and the discomfort experienced by the public following a terrorist attack on the power system. DHS could develop guidelines for communications under these conditions.
DHS could consider research into the unique problems that could result from terrorist attacks on the power system in areas where centralized markets exists. Disruption of markets can be as difficult to deal with as problems with the physical electric system and could lead to chaos if the potential consequences and countermeasures are not thought out in advance. Such work should develop guidelines for market operators to use in the event of market disruption.
As the respondents reallocated priorities for R&D related to security, they tended to decrease funding for all other items.
1No precise budget was specified, but respondents were told “if your allocation would depend on how much money you have available and for how long, assume you have $400 million per year for at least the next decade.”
Technology Solutions (CERTS). The manufacturers of the electrical apparatus and equipment used in power systems also conduct research related to development of new equipment. Most of these efforts are modest and are conducted outside the United States. Smaller firms increasingly are developing technologies that are digitally based and intended for potential deployment on power systems.
In addition, individual utilities sponsor some R&D projects, but these internal R&D budgets and R&D staffs are only a fraction of what they were in the mid-1990s. Two states have substantial R&D programs. The California Energy Commission (CEC) has a major transmission research effort underway. The New York State Energy Research and Development Authority (NYSERDA) has complementary work underway as well.
Nationally, a temporary R&D tax credit enacted as part of the Economic Recovery Tax Act of 1981 has been extended several times, although the R&D tax credit that expired on December 31, 2005, was not renewed until December 2006, resulting in a 1-year gap. In recent years, support for R&D investment has been constricted by a number of factors, including reduced federal funding and the cost pressures on private industry. As a result, it has become increasingly important that there be renewed support for research funding.
It is also essential that support for necessary improvements to the U.S. electric power delivery system be continuous in this critical time. The tax credit provision strengthens the innovation, productivity, and competitiveness of the U.S. economy and is vital to U.S. leadership in technological innovation and global competitiveness in the 21st century.
Many of the technologies described above are not yet sufficiently developed to be attractive private sector research investments toward deployable products and systems, even with tax credits. Others are still too expensive or do not have the level of functionality required for wide adoption, even though they may provide substantial benefits to society as a whole.
The current level of R&D funding in both the public and private sectors of the electric industry is at an all-time low. Neither the utility industry nor the electrical apparatus industry is spending as much as could be justified by the expected benefits of improved technology, particularly for longer-term research. The committee believes that a much larger annual R&D investment is required in order for today’s transmission and distribution technologies to evolve and for the necessary new technologies to become realities.
In trying to be responsive to their stakeholders, utilities typically tend to limit R&D to areas of immediate application and payback. Aside from these short-term developments, utilities have little incentive to invest in R&D for the longer term. Furthermore, for regulated investor-owned utilities, there is the additional pressure caused by Wall Street to sustain and increase dividends. In addition, during the restructuring of the last decade a substantial number of utilities agreed to rate caps, which, in the face of ongoing cost increases, put pressure on what were perceived as discretionary budget items such as R&D. Government is likely to be the only source of funding for basic and long-term R&D. Therefore, this research is unlikely to be undertaken unless the government significantly increases funding for electric transmission and distribution R&D.
There have been various attempts in regulatory proceedings to encourage or establish increased levels of R&D investments. The results from such efforts have been mixed. In some cases, funds have been used for economic development activities or local demonstrations of already commercially available technology, activities that contribute little to stimulating the innovations in science and technology that are needed. Usually, developments by any one state are not sufficient to influence the market for technology. Collaborative programs have had more success in this regard; however, states have difficulty in funding any research outside their state.
In addition, the enthusiasm among state regulators to encourage higher levels of R&D for the utilities they regulate is tempered by the difficulty of providing strong business cases for R&D—the results of which are inherently unpredictable. Moreover, investments in R&D often require patience before longer-term paybacks are realized. Yet another difficulty in encouraging R&D concerns the phenomenon of “free-rider” utilities, so-called because they take advantage of R&D done by others—often while participating in collaborative arrangements. Such free-riders inhibit some entities from joining collaborative efforts.
In addition to problems with state mandates and underinvestment in the industry, research priorities differ by utility and by region. The extent to which utilities have staff capable of managing research activities also varies, as does the strength of their connections with local universities, national and commercial laboratories, and national research organizations.
Low levels of support for R&D have led to dramatic shrinkage in university programs in power systems. For a while the field was seen by many electrical engineering (EE) departments as uninteresting. Today, with all the new developments underway, that is no longer true. However, when having to choose between hiring an assistant professor in power engineering who might manage to secure research support of a few hundred thousand dollars per year, and an assistant professor in a field such as micro-electronics who might succeed in securing research support in excess of a million dollars per year, EE department heads have been understandably reluctant to replace retiring power engineers or add new junior faculty in this area. The result has been a growing shortage of people with strong technical capabilities in this field.
Societal benefits from adequate R&D investment in the electric power delivery system could extend far beyond the benefits from enhancing the resilience of the power system. These include the economic benefits from enhancing the depth of research in the United States overall and the enhancements in overall productivity. A modern power delivery system is critical to supporting the nation’s future and will not evolve without increased R&D.
To achieve the level of R&D expenditure discussed above, R&D budgets would have to be increased substantially both in industry and by the federal government. To date, no agreement has been reached by the diverse players in the power industry, political decision makers, or society as a whole on a strategy to secure funding at a level to adequately address research needs in the electricity industry. This committee likewise found total agreement hard to attain, with all but a few members of the committee agreeing that federal legislation and regulations should be pursued that can achieve the following goals for the electric power sector:3
3The committee did not achieve consensus on the need for substantial additional federal funding because of the following issues: a) as a mature industry, electric power companies and suppliers should be able to fund their own research; b) rapidly expanding grids in other countries should provide ample incentives for new developments; and (c) much of the underlying R&D is done by other industries (e.g,, communications and information
• A coherent national plan for increasing both public and private sector R&D funding to address electricity needs;
• An increase in the current level of U.S. R&D (public and private) to $10 billion per year. While somewhat speculative, this amount is approximately three times the current level of R&D, but only about 3 percent of total U.S. R&D, only about 0.1 percent of U.S. annual GDP, and less than 5 percent of annual utility revenue;
• An approximate doubling of federal electricity R&D budgets, increases that should not be burdened with further earmarks;
• A federally legislated requirement that the electricity industry’s share of this increase for R&D should come from consumers;
• Specification by such a mandate that 3 percent of the amount charged on a consumer’s electricity bill be directed to R&D. Existing programs and R&D budgets that meet the criteria outlined below should be awarded the funds raised by the 3 percent levy. The program should be designed to require each and every industry or market participant4 to invest 3 percent of the value-added portion of their revenues annually in R&D as defined below. Value-added should be defined as follows:
—For the generation portion, it should be the total cost of generation.
—For the transmission ownership portion, it should be the transmission wires charge.
—For the transmission operations portion, it should be the cost of operations.
—For the distribution portion, it should be the distribution wires charge.
—For the retail service provider, it should be the marginal cost of services provided to the consumer.
• Structuring of the program to ensure that the amount invested in R&D is fully recoverable from consumers according to a method that involves every U.S. provider and consumer in as fair and equitable a manner as possible. Consumers generally are the intended beneficiaries of the outcomes of the needed R&D and ultimately must pay the bill;
• Management of the investments in R&D by the industry participant (1) to conduct R&D directly itself or to contract such work to a for-profit research provider or (2) to fund R&D performed by nonprofit research institutions, national public-private collaborations, or state and federal government entities, such as national laboratories;
• Regular open review of each individual industry participant’s R&D portfolio by a consortium of its stakeholders to obtain input on research direction and priorities;
• Exclusion of activities from the proposed R&D program according to the definition by the Internal Revenue Service, which is as follows: “Scientific research does not include activities of a type ordinarily carried on as an incident to commercial or industrial operations, as, for example, the ordinary testing or inspection of materials or products or the designing or construction of equipment, buildings, etc.” (Treasury, 1986);
• Oversight of the program by an appropriate combination of accountability authorities—such as state energy regulatory commissions, the Federal Energy Regulatory Commission, or the Internal Revenue Service—charged with ensuring that research dollars are being applied to their intended targets. To facilitate tracking, appropriate accounting systems will have to be implemented;
• A 10-year sunset and review embedded into the program design.
The committee recognizes the potential for a variety of pitfalls in a program with the general objectives outlined above. If they are not carefully crafted, such programs also can be subject to abuse. Accordingly, the committee recommends that an executive branch agency be charged with developing a proposal that addresses the issues in implementing such a program.
In large measure, today’s electric power system can be viewed as comprising more than 130 cohesive electrical zones. These zones have evolved based on utilities’ efforts to meet the growth in electrical load by locating generating facilities reasonably close to customer load centers and arranging a network of electric transmission and distribution systems (wires, breakers, transformers, and so on) to meet customer needs. These zones were tied together over time (interconnected) to enhance reliability and to enable the most cost-effective and efficient use of generation. Many zones are considered “control areas” and are controlled in an independent way that includes coordination with other control areas in a region. Today’s control areas could be described as being partially independent while being integrated with neighboring control areas.
The configuration of today’s power system is based largely on central station power plants located in control areas. The power delivery system that integrates these power
technology) which the electric industry should be able to adapt and apply without more federal spending. Most committee members conclude that the needed R&D will not take place on a useful schedule without more federal involvement.
4This includes vertically integrated utilities, power generators, transmission owners, transmission operators, distribution utilities, and retail providers (where they are active).
production facilities with consumers is constrained, as evidenced by the growing number of failed wholesale transactions. In addition, the power delivery system is mechanically controlled with only limited integration of communications, automation, or computational ability. Figure 9.2 depicts the potential evolution of today’s power system along two dominant dimensions—one the degree of centralization, the other the degree of system integration—fully integrated communications, sensors, and computational ability vs. greater autonomy depending on how automation occurs.
In this paradigm, the issue of whether tomorrow’s power system will become more decentralized or more centralized is the greatest driver. The path taken will affect decisions about which technologies to pursue most vigorously, but the committee does not recommend one approach over the other.
The basic philosophy in the decentralized approach is to first increase the independence, flexibility, and intelligence of local systems for optimization of energy use and energy management at the local level, and then to integrate local systems as necessary or justified for delivering power supply and services that consumers desire. Four configurations are associated with a decentralized approach:
• Device-level power systems
• Building-integrated power systems
• Distributed power systems
• Fully integrated power systems
The decentralized approach starts with the notion that consumers increasingly expect energy-consuming devices and appliances to operate optimally. Optimal operation not only potentially enables a highly mobile digital society, but also, once the optimal performance of devices is defined, provides elements of performance which enable, in turn, a building-integrated system. Building-integrated systems can also accommodate increasing consumer demands for independence, convenience, appearance, environmentally friendly service, and cost control.
Building-integrated systems can, in turn, be integrated into distributed systems, which can then be interconnected and integrated with technologies that ultimately enable a fully integrated national-scale “perfect” power system (Figure 9.3). Note that such systems could be restricted in terms of their rating size and might not have the advantage of economies of scale that current interconnected centralized systems have. Each configuration in this approach reflects a distributed level of both instrumentation and control and would require a complementary set of milestones on the path to comprehensive national power system perfection.
The four different configurations refelect development of the system in two important dimensions:
• Level of intelligence and energy capacity in distributed devices and systems. Increased investment in local intelligence and infrastructure also accelerates progress through entrepreneurial leadership opportunities not initially available at higher levels of system integration.
FIGURE 9.3 Development path for the perfect power system. SOURCE: Galvin Electricity Initiative (2006).
• Level of integration of the entire power delivery infrastructure. Higher levels of integration require ever more significant transformation of the infrastructure for communications and control, as well as of the overall power delivery infrastructure.
Each of these configurations can essentially be considered a possible structure for a future power system in its own right, but each stage logically evolves to the next stage based on the efficiencies, and the quality or service value improvements, to be attained. In effect, these potential system configuration stages build on each other starting from a device-level power system connected to other device-level power systems that then can evolve into a building-integrated power system, a distributed power system, and eventually a fully integrated power system as diagrammed in Figure 9.4. Figure 9.4 also highlights technologies that would have to be further developed for this concept to evolve.
The optimum configuration may vary for different environments. For instance, the availability of inexpensive and clean central generation (e.g., advanced coal, advanced nuclear, advanced hydro, and large wind systems) may accelerate the migration to the fully integrated stage, whereas other service systems developing from new portable, localized, or distributed infrastructures may achieve their final optimum in the distributed structure.
In a stochastic simulation of a completely decentralized system, Zerriffi (2004) showed that such systems could achieve dramatic improvements in power delivery reliability in the face of system disruptions (see also Farrell et al., 2004). Although no civilian system has approached this level of decentralization, some military systems have begun to evolve toward it.
Distributed systems have also become attractive to those concerned with energy efficiency and reducing CO2 emissions, because it is typically possible to operate them as high-efficiency combined heat and power systems. The net energy use efficiency of such systems can be twice that of central stations in which “waste” heat must be disposed of via cooling towers. Recent analysis by King (2006) suggests that even with current technology and rate structures, micro-grids could be cost attractive in some applications. However, there are significant regulatory barriers that must be addressed if such systems are to become widespread (King, 2006; Morgan and Zerriffi, 2002).
FIGURE 9.4 Evolution of possible configurations (from center outward) and relevant nodes of innovation (in outer ring) enabling the power system. SOURCE: Galvin Electricity Initiative (2006).
The centralized approach assumes that the creation of an intelligent electricity power delivery infrastructure will evolve from the existing power system through bottom-up transformation created by individual companies adding advanced capabilities piece by piece onto the existing grid.
The basis of this transformation is that over the last few decades, advances in diverse technologies—solid-state electronics, microprocessors, sensors, communications, and information technology (IT)—have transformed society and commerce, permanently increasing society’s capabilities and expectations. These advances also present new opportunities for operating and using the electric power network, opportunities not envisioned when the power delivery system was first formed. For the power system itself, there is the possibility of creating a nimbler, more flexible network that marries electric power with cutting-edge communication and computing capabilities—an intelligent system that can
predict power problems before they get out of hand and heal itself when damage is unavoidable.
Another aspect of an intelligent system is the ability to fully utilize existing assets through greater system control and flexibility, along with new concepts of designing for high reliability. Opportunities for improving the overall efficiency of the power system equipment use and operation, while still maintaining reliability, are possible in areas such as a dense urban environment, where existing assets are located in close proximity but are often not fully employed.
For electricity customers, a smart power system means not only enhanced power reliability and security but also new services that can add value by giving customers options for control of use, and thus the cost of electricity. For example, customers may be able to monitor their building or industrial-process energy use in real time, choose from a menu of service packages to best fit their energy needs and use patterns, and even sell excess electricity from distributed generation back to their power provider. The promise of a smart power delivery system clearly carries advantages for utilities and consumers.
The change to an intelligent digital system will come from the gradual confluence of innovative projects undertaken by individual companies, rather than through a sudden transformation. Although the new smart devices and technologies developed for these projects will be of value individually, the greater benefit to the power network will be realized only when they all work together. Ensuring that the individual sensing, communications, and computing equipment installed over the coming years can be integrated with other systems and, eventually, come together to form a single system requires an overall power network architecture—that is, common methods and tools for planning and designing the smart systems, and a complete suite of standards. For this purpose, current information technology has some shortcomings. Architecture and standards for power systems have to include consideration of how the legacy systems can be preserved and integrated.
At present, more than 150 different communications protocols are used in the U.S. electric utility industry. Interoperability in today’s environment is thus impossible. The industry and the federal government have begun to recognize this deficiency and have initiated several efforts to formulate an architecture that could underpin a smart power system.
These various approaches all rely, in one way or another, on one or more innovative technologies. Many of these technologies have not been fully researched, developed, or demonstrated.
Currently available technology can and should be used more extensively to protect the power delivery system against terrorists, disgruntled employees, or severe natural disasters. There are, however, serious limits (both economic and technical) to how much protection current technology can provide. Advanced technology can raise these limits significantly. The committee’s assessment of the status of research and development for the electric power delivery system led it to draw the following general findings.
Finding 9.1 Even in the absence of terrorist attacks, current and projected future inadequacies in the electric power delivery system are likely to result in deteriorating reliability, excessive instances of degraded power quality, and the inability to provide enhanced services to consumers.5 Inadequate investments in this infrastructure and growing demand for electric power have led to an increasingly stressed system.
Finding 9.2 Underinvestment in R&D for the electric power delivery system has been even more pronounced than underinvestment in the infrastructure. New technologies and techniques are not being developed that could overcome stresses and reduce the cost of delivering electric power to meet the new and growing needs to which the system must respond.
Finding 9.3 There is considerable overlap between the R&D needed to reduce vulnerability to terrorist attack and the R&D that can address the challenges already faced by the power delivery system. An R&D strategy for the power delivery system focused exclusively on terrorism is likely to be less cost-effective and less successful than an integrated strategy to address all the needs and challenges confronting the system, including those posed by terrorism.
Finding 9.4 EPRI, DOE, and a number of utilities and corporations have all engaged in R&D road mapping exercises for the electric power delivery system. The most critical needs are already well identified, and a much larger and more comprehensive R&D program could be created rapidly. The elements of this program are listed in Table 9.1. A more extensive list is shown in Appendix H. DOE would have primary responsibility for most of this program.
DHS should cooperate with DOE to support the following parts of an enhanced R&D program for electric power transmission and distribution to harden the system against terrorism, mitigate the impacts of terrorist acts, and enhance recovery.
Recommendation 9.1 Complete the development and demonstration of high-voltage recovery transformers, and
develop plans for the manufacture, storage, and installation of these recovery transformers.
Recommendation 9.2 Continue the development and demonstration of the advanced computational system currently funded by the Department of Homeland Security and underway at the Electric Power Research Institute. This system is intended to assist in supporting more rapid estimation of the state of the system and broader system analysis.
Recommendation 9.3 Develop a visualization system for transmission control centers which will support informed operator decision making and reduce vulnerability to human errors. R&D to this end is underway at the Electric Power Research Institute, Department of Energy, Consortium for Electric Reliability Technology Solutions, and Power System Engineering Research Center, but improved integration of these efforts is required.
Recommendation 9.4 Develop dynamic systems technology in conjunction with response demonstrations now being outlined as part of an energy efficiency initiative being formed by EPRI, the Edison Electric Institute, and DOE. These systems would allow interactive control of consumer loads.
Recommendation 9.5 Develop multilayer control strategies that include capabilities to island and self-heal the power delivery system. This program should involve close cooperation with the electric power industry, building on work in the Wide Area Management System, the Wide Area Control System, and the Eastern Interconnection Phasor Project.
Recommendation 9.6 Develop improved energy storage that can be deployed as dispersed systems. The committee thinks that improved lithium-ion batteries have the greatest potential. The development of such batteries, which might become commercially viable through use in plug-in hybrid electric vehicles, should be accelerated.
The committee believes that electric power R&D budgets should be increased substantially, although there was no consensus as to the appropriate source of the funding. Resolution might come as a result of considering research policy options: What are the impacts if the funding comes from the government, or from private industry, or from some combination thereof? One possibility is a federally mandated program constructed such that each industry participant invests some fraction (say 3 percent) of the value-added portion of its revenues annually in R&D, that the expense is fully recoverable, and that the cost is allocated to every U.S. provider and consumer as fairly and equitably as possible. DHS should work with DOE and the Office of Management and Budget to substantially increase the level of federal basic technology research investment in power delivery.
EPRI (Electric Power Research Institute). 2004. “Power Delivery System of the Future: A Preliminary Study of Costs and Benefits.” Palo Alto, Calif.: EPRI.
Farrell, A.E., H. Zerriff, and H. Dowlatabad. 2004. “Energy Infrastructure and Security.” Annual Review of Environment and Resources 29: 421-469.
Galvin Electricity. 2006. Phase 1 Summary, The Galvin Electricity Initiative. Available at www.galvinelectricity.org.
King, D.E. 2006. “Electric Power Micro-grids: Opportunities and Challenges for an Emerging Distributed Energy Architecture.” Ph.D. Thesis, Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, Pa.
Morgan, M.G., and H. Zerriffi. 2002. “The Regulatory Environment for Small Independent Micro-Grid Companies.” Electricity Journal 15(9): 52-57.
Treasury (U.S. Department of the Treasury). 1986. Scientific Research Under IRC 501(c)(3). Treasury Regulation 1.501(c)(3)-1(d)(5)(ii). Available at http://www.irs.gov/pub/irs-tege/eotopico86.pdf. Accessed November 2007.
Zerriffi, H. 2004. “Electric Power Systems Under Stress: An Evaluation of Centralized Versus Distributed System Architectures.” Ph.D. Thesis, Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, Pa.