Thus, utilities need to develop emergency response plans. Although it is not possible to cover all possible emergency scenarios, the planning and drill process is invaluable in building a capability in responding to actual events because it provides a basic framework and foundation. The following should be considered as part of future contingency response development:
• Evaluation of existing facilities based on their criti-cality and development of plans for recovery in the event of the loss of all key equipment in several of these facilities (e.g., the loss of entire substations or loss for an extended period of one or several key transmission lines). Such contingency analysis should be conducted to determine the impact of this loss on other facilities.
• For new designs or upgrades, a planning/engineering process that considers how to make facilities more robust in the face of possible attack, and development of strategies to quickly restore or bypass such facilities if they sustain signifcant damage.
• Sharing by utilities of ideas and designs that may improve performance. Organizations such as the Edison Electric Institute (EEI) and the Association of Edison Illuminating Companies (AEIC) are excellent forums for such sharing. Benchmarking with other utilities, especially those in countries that have had experience in addressing terrorist threats and attacks, will provide valuable lessons and ideas. For example, the Infrastructure Security Initiative sponsored by the Electric Power Research Institute (EPRI) produced Counterterrorism Measures and the Protection and Restoration of an Electric Grid (EPRI, 2005a), a report that describes Israel Electric Corporation (IEC) programs and procedures for maintaining the integrity of Israel’s power transmission and distribution system, as well as related restoration efforts. However, there is a decided limit to how much special investment private utilities can be expected to make to protect against low-probability threats to every key element of their system.
To prepare for the possible need to mount a restoration of service, utilities should carefully address several important issues:
• Black-start capability (that is, the ability to supply limited amounts of power to generators and other power-system equipment before they can be brought back online);
• Line/cable charging strategies and other means of voltage and reactive power control;
• Need to disable or adjust certain protective systems, such as those for undervoltage, underfrequency, synchronization checks, and so on;
• Use of restoration panels; and
• Development of restoration policies, including islanding requirements and monitoring of voltages, frequencies, and phase angles.
In anticipation of catastrophic events leading to a system-wide blackout, utilities are required to develop plans that will enable their operators to break up the normally synchronized grid into “isolated” islands that are self-supportive. Such advanced planning can be valuable, but in the event of any specifc outage, these plans will require real-time adjustments based on existing conditions, such as availability of equipment, load conditions, reactive power supply/control capability, availability of synchronizing equipment, and governing control while maintaining voltages and frequency at acceptable operating levels.
Plans for the restoration of a transmission and distribution system should consider two basic approaches. One is based on the availability of power from other external providers through tie lines. A second, or “island,” approach considers restoration of the system from generation internal to its service territory. The latter approach could be signifcantly strengthened with the greater deployment of various types of distributed generation, including micro-grids. Today, however, there are considerable regulatory impediments3 to the deployment of such systems, and distribution system operators typically do not have plans to make use of such resources in emergency situations.
With some important exceptions, many distribution circuits serve both socially critical facilities such as police stations, schools, and flling stations, together with many less critical facilities. If the supply of power were to become seriously limited, it would be highly desirable to temporarily restrict service to just critical loads. Advanced distribution automation (see Chapter 6) could make it possible to rapidly and selectively supply service to a few such key facilities. However, many systems still do not have distribution automation, and in the case of those that do, most have not been confgured to facilitate such selective load shedding within a single distribution feeder. In the absence of such capabilities, reconfguring distribution feeders to serve just a few loads would typically be a slow, labor-intensive operation (sending line crews out to open or close breakers at customer service drops), as would be restoring service to dropped customers along such feeders as power supplies once again became more plentiful.
3These impediments include informal diffculties that many distributed resources still experience when trying to connect to the utility system (Alderfer et al., 2000), interconnection rules that currently require all distributed resources to disconnect from the grid the moment any problems arise (IEEE, 2003), and laws that grant legacy utilities exclusive service territories, making the installation of small micro-grids that serve several customers diffcult or impossible in much of the country (King, 2006; Morgan and Zerriff, 2002). There is additional discussion of some of these issues in Chapter 9.