 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
Prepared Witness Testimony The House Committee on Energy and Commerce
Blackout 2003: How Did It Happen and Why?
Full Committee on Energy and Commerce
Mr. Larry Makovich
A definitive analysis of the contributing events and causes of the August 14, 2003, blackout will take months to complete. At this time, Cambridge Energy Research Associates' (CERA's) analysis indicates that a combination of normal component failures, transmission system deterioration, and an inability to respond and contain the problem at several levels of control caused the cascading blackout. At this time, it appears that the greatest power failure in US history began with normal component failures. For example, one failure on August 14 was an unplanned outage of a unit at the East Lake power plant that caused power flows to instantaneously reroute in the transmission network. Such unplanned power plant outages occur thousands of times each year and so too does the instantaneous rerouting of power flows. Such normal component failures and dynamic power flows are part of normal power system operations. Transmission system operators plan for normal component failures. To do this, they configure the electrical system-the real-time balancing of sources of power and uses of power and the limits on transmission line loadings in the system to withstand the effects of normal component failures. At a minimum, proper transmission network planning keeps the power system configured in such a way that it can withstand the effect of the most critical component in the system failing (first contingency planning). Automatic controls on generating plants and transmission lines allow the power system to isolate problems, protect equipment, and reconfigure itself to a stable condition within seconds following a normal component failure. As power system conditions change (supply, demand, weather, etc.), power flows reroute at close to the speed of light. Thus, when a generating unit and a transmission line trip and power reroutes, several transmission lines carry more power and, as expected, begin to sag. On August 14, one of these lines carrying more power near Cleveland sagged close enough to a tree to short circuit. Proper maintenance (tree trimming) should prevent such contact but, again, transmission line failures of various types are something power system operators also plan for. Nevertheless, when power rerouted along the remaining lines, additional overloading occurred and automatic protections for generating plants and transmission lines disconnected additional power plants and lines in the network. At some point, the multiple failures pushed the system past its limits to isolate and restabilize. Consequently, the problem expanded over a larger area of the power network as significant rerouting of power flows continued. When a power system is not configured to contain a normal component failure, the destabilization of a larger part of the power system quickly follows. Power surges spread through some parts of the network-Pennsylvania, New Jersey and Maryland, and AEP-that reacted (both automatically and with discretion) to isolate themselves in order to maintain stable system operations. However, such actions add to the rerouting dynamics of the remaining power network and begin to overwhelm the remaining parts such as eastern Michigan, Ontario, and finally New York. The root cause of the cascading blackout appears to be a breakdown in the planning, coordination, and communication necessary to control the interconnected power systems. The sequence of events in the blackout caused parts of the power system to act on their own rather than in a coordinated fashion. Such coordination has not gotten the proper investments of time, money, and systems in the past several years and this system deterioration-the cumulative effects of years of underinvestment in the varied needs of transmission networks-is a root cause of the blackout. Past Efforts to Prevent and Minimize Blackouts The blackouts of 1965 and 1977 in the Northeast and in 1996 in the West spurred efforts to prevent and minimize blackouts in the future. The lesson from 1965 was that greater integration of regional power systems created desirable day-to-day benefits from electric trade but required an associated higher level of planning, coordination, communication, and control to prevent cascading power outages. As a result, the formation of the North American Electric Reliability Council (NERC) and its regional reliability councils followed the 1965 blackout. The lesson from the two blackouts of 1996 in the West was that a breakdown in planning, coordination, communication, and control can allow normal events-again, in one case, a power line sagging into a tree-to cascade into a large regional system failure. In this case, the cascading failure began with federally owned transmission assets that were highly integrated with other publicly and privately owned transmission infrastructure. Following the 1996 blackouts, the western power system decreased the amount of power flowing on transmission lines (forgoing savings from increased power trade) in order to maintain the level of redundancy necessary to prevent a repeat of cascading failures following normal component failures. A year or more passed before the planning and coordination got to the point that these power transfer limits could return to pre-blackout levels. The blackouts of 1977 in New York and several years ago in Chicago highlighted the problem of underinvestment in power delivery systems. In Chicago the problem was underinvestment in distribution (the small wires near homes) rather than in transmission (the large wires that carry power long distances). Even the best planning and coordination to properly manage a power system cannot offset the problems created by continued underinvestment. Eventually the probability of multiple component failures and the increasing constraints on systems operators charged with configuring a reliable power system leads to a major blackout. This underinvestment affects more than just transmission lines and substations and includes computer systems, backup systems, software, instrumentation, data, rules, and organizations. What Worked on August 14? The conditions across the eastern power interconnection on August 14 were not highly stressful. The East was not in the throes of a prolonged heat wave or suffering from an abnormally high level of supply outages. Interregional power flows were providing benefits, as areas with higher-cost generation were able to draw upon areas of lower-cost generation. As the blackout cascaded through the Midwest, Ontario, and New York the automatic protective devices for power lines and power plants worked to prevent damage. Restoration of electric service reflected a well-thought-out and rehearsed sequence of procedures. The control centers of the electric systems appear to have captured the real-time data necessary to reconstruct the details of the cascading failure. The blackout exposed weakness in the US power grid but did not provide evidence that the US has a third world transmission infrastructure. Normal component failures should be expected even in a state-of-the-art transmission network. Quite to the contrary-the high degree of interconnection of the US grid exposed the need for better planning, coordination, communication, and control. Needed Improvements Defining the Transmission Mission Electric transmission is critical infrastructure in the US economy. The transmission network is a natural monopoly that is in the middle of an industry that is stuck halfway between regulation and the marketplace. Transmission remains in the center of integrated regulated power companies and public power entities as well as at center stage in emerging power markets, where it governs the interactions between consumers and producers. A properly structured transmission sector requires that the institutions and rules meet the needs of both of these existing industry structures. Transmission policy must adjust to the reality that regional power systems in the United States will operate for quite some time with very different structures-some relying greatly on market mechanisms and others relying on comprehensive regulation. Transmission institutions and rules must accommodate the different power industry structures that are interconnected and need to interface properly. Transmission Organizations Transmission organizations need to reflect the underlying reality of the transmission infrastructure. We do not have a seamless, national transmission grid and are not even close to having one. Instead, the US power system consists of a dozen regional transmission networks within three largely independent transmission interconnections, with varying levels of power transfer capability between regional networks and with networks in Canada. These networks cover multistate areas and need organizations that align with the physical extent of the grids to implement the necessary planning, coordination, communication, and control. Thus, the Federal Energy Regulatory Commission (FERC) should not allow movement to the market in regions that do not have proper alignment between the transmission organization and the network. Currently, the US Midwest network has two transmission organizations in formation and transition, rather than one, and suffers a misalignment between the organizations and the underlying extent of the regional network. On the other hand, if the FERC gains authority to order regional transmission organization participation in regions moving to the market, then it should also order proper alignment between transmission organizations and networks. Since these regional networks do have significant interconnections, the need also exists for an umbrella organization to coordinate operations and interdependencies within the interconnections. We want sufficient overall control to avoid situations in which one regional network protects itself by causing collapses in neighboring networks. The current NERC comes close to the envisioned umbrella organization but suffers from being a voluntary organization with limited enforcement authority. Mandatory Reliability Standards and Procedures Mandatory electric reliability standards and procedures would help address the breakdown in planning, coordination, and communication that are at the foundation of power system control. A system of rules and procedures is needed that provides real-time information flows such that all system operators have a clear view of not just their local power system but also the larger whole. Such standards and procedures need to be enforced by an agency with authority over both publicly and privately owned transmission assets in competitive as well as regulated industry structures. International agreements are also necessary to coordinate with Canadian power systems and, to a much smaller extent, Mexican power systems. An umbrella organization must ensure that contingency planning evaluates the power system as a whole-and is not just an uncoordinated set of regional contingency plans with a blind spot regarding their interdependencies. Resolving the Gridlock in Transmission Investment More investment is needed in the US transmission network. Many opportunities exist where the benefits of additional transmission infrastructure investments far exceed the costs, and this result is robust under a wide range of future conditions. The problem, as CERA identified in its 1999 report entitled Gridlock-Transmission Investment and Electric Restructuring, is that "[c]urrently there is no entity in the emerging industry structure-neither generators, transmission owners, independent system operators, distribution companies, traders, retail marketers, nor end users-facing the proper incentives to invest." Our conclusion four years ago was that "[s]ustained underinvestment in transmission may eventually threaten the reliability of the bulk power system." Underinvestment in transmission and the gridlock in transmission policy are longstanding problems. When I last testified before the Senate in July 2002, CERA warned that a continued lack of investment would lead to reliability problems: "A gridlock plagues most transmission investment decisions because incentives are misaligned." These investments "were not being undertaken because no one faced the full costs and benefits of AC network investments and was in a position to pursue these opportunities profitably." Over a year ago, the Department of Energy's National Transmission Grid Study provided a similar warning. And in CERA's Special Report Energy Restructuring at a Crossroads: Creating Workable Competitive Power Markets, 5 out of 12 recommendations on making power markets work involved transmission issues. CERA's currently ongoing study Grounded in Reality: Bottlenecks and Investment Needs in the North American Transmission System is finding that significant transmission congestion exists both within and between regions. The solution goes beyond higher allowed regulated rates of return, tax incentives, or accelerated depreciation. The payoffs already exist. The problem is settling who pays. The current principle is that whoever benefits ought to pay. However, implementation of this principle is very difficult. Benefits are robust under a wide variety of conditions but as conditions change, the incidence of those benefits can shift dramatically. Transmission investment is stymied by the complex arguments of who will benefit and thus who should pay. As a result, adequate investment is not yet being made. Transmission investment planning at the network level that guarantees cost recovery and prevents investment indecision due to gridlock on cost allocation and recovery mechanisms is sorely needed. One possibility is a policy that allows economic transmission investment identified by analyses at the network level to go forward with a default decision to spread the costs across the entire network. Reallocations and true-ups can follow later if necessary and substantiated.
|
|||||||||||
|
