Competence building - Capacity shortage

The objective of the project was to increase knowledge about central approaches aimed at solving the peaking capacity problem in restructured power systems. Specifically, the project assesed the following fields:

• Market based solutions to provide reserve capacity
• Consequences of specific reserve requirements

The project had a Nordic perspective with relation to the applicability of proposed solutions. However, the project had an international perspective with respect to solutions, methods and cooperation.

The project was financed by the Research Council of Norway, The Norwegian Electricity Industry Association EBL, and Statkraft, Statnett, Norske Skog, Nordpool and Norsk Hydro.
 
The characteristic property of peaking power is that it is needed only a few hours annually, and moreover that the need for it is uncertain. Partly because of this, it is unclear if existing market mechanisms are sufficient to provide peaking capacity. This uncertainty is still endangering the whole restructuring process. In some cases it has clearly come to the surface (the evident example is California), in others there seem to be no great problems so far, but this does not mean that the problem has been solved. Failure to solve the peaking power problem will sooner or later result in involuntary load-shedding as was experienced in California. If a satisfactory solution to this problem is not found, the whole restructuring process may be at stake.

The problem has considerable international interest, illustrated by numerous publications and a several conferences dedicated to the subject, e.g. Market Design 2001, and Peak Production Capability and Peak Load in the Nordic Electricity Market. Many countries are presently skeptical to the type of market solutions that is in use in e.g. Scandinavia, because there are doubts with respect to their ability to guarantee system security. Because the Scandinavian restructuring process started in a situation with considerable excess capacity, the problem was not evident until recently, but developments in the last two years have shown that this is changing quickly. Problems similar to those experienced in California may also occur in the Nordic market, albeit to a lesser extent, because of a more sound market design (specifically a better “throughput” of prices from the wholesale to the retail market).

Sound solutions to the peaking capacity problem are vital to a continuing satisfactory performance of restructured electricity markets. Without such solutions, there will occur situations where demand has to be shed involuntary to maintain system security. This is economically inefficient and at the same time socially unacceptable.

Background
In a free electricity market it should in principle not be necessary to worry about the availability of peaking capacity. If there is a deficit, prices increase. This reduces demand and gives a signal to the supply side to invest in new capacity. However, there are good reasons why this mechanism does not guarantee an optimal solution in the electricity market:

• Electricity cannot be stored economically. This means that there has to be a continuous minute-to-minute balance between demand and supply.
• There are very high requirements to reliability. The reason for this is the high costs to consumers of power interruptions. Consumers’ preferences for electricity can be viewed as “asymmetric”: the cost of an interruption is much higher than the general willingness to pay.
• The power system is one large interconnected system, where problems under high-load conditions can result in new problems other places. The final consequence of a component outage somewhere in the network can be system collapse, resulting in long and extended interruptions of demand and huge costs.
• The traditional tariff systems completely protect consumers from short-term variations of the spot-market price.

As a result, the availability of peaking capacity is a potential (and in some places already real) problem in restructured electricity markets. Also in the Nordic market experiences from the last two winters show that this is becoming a very real problem.

Activities

Market based solutions to provide reserve capacity
The objective of this Activity is to do a thorough analysis of reservation, activation and pricing of reserves on a competitive basis. The way reserves are treated has significant impact on market prices, and gives both direct and indirect signals to potential investors. The significance of efficient pricing of reserves goes beyond the reserve markets, because its impact on the far greater volume of the (energy) spot market. A central aspect is the interaction between the transmission network and market based solutions.

Consequences of specific reserve requirements
Operational reliability (security) of the power system is ensured by adhering to reserve requirements. These are defined with various sets of time/frequency response (e.g. primary, secondary and tertiary reserves) that have to be available in real time. Strictly speaking, reserve requirements are substitutes for the underlying security requirements; they are an indirect way to ensure that these requirements are satisfied. The reason is that it is difficult to apply the security requirements directly in an operational context. This is partly because existing methods are not good enough, partly because of very high requirements to data availability and partly because even today’s computing capacity is not satisfactory. In principle, reserve requirements are the result of the reliability requirements, but the latter do not have a sound economic basis, and their justification is normally “engineering judgement”.

In a restructured market some parties may wish to reduce reserve requirements in certain periods. This is evident in situations where it would be necessary to shed load in order to satisfy the requirements, but it may also occur in other situations where market participants would be better off with reduced reserve requirements. Moreover, it is clear that reserve requirements in an inflexible market should not be identical to a market with more flexible demand. Another issue is the level of the requirements in relation to the characteristics of the generation side, e.g. a development in the direction of more decentral generation.

Capacity balance simulations
The balance between supply and demand of capacity is of vital importance for system security. Therefore, a tool to analyze the development of this balance for the medium term (1-5 years ahead) can be very useful when designing measures to avoid capacity shortage. If it appears that insufficient capacity will be available, quantitative information is a valuable basis for taking the best decisions. Also for market participants it can be of great interest to be able to analyze the market ahead specifically with respect to capacity shortage, as a basis for deciding upon investment in new generation.

In this Activity we propose to develop a market simulator for the Nordic market, which focuses on capacity. It will quantify both peaking incidents, their duration and the uncertainty involved. The result of this activity will be a model well-suited for the analysis of demand peaks and concrete results for the Nordic power system.

Capacity subscription
The root of the peaking power problem lies in the inflexibility of prices or rather tariffs paid by consumers. While in other markets demand is reduced by high prices in times of shortage, this mechanism does not work in the electricity market, due to a combination of historical, technological and economic factors. Thus it can become necessary to resort to random rationing of demand, which is expensive and socially unacceptable.

One way to solve the problem would be to let consumers subscribe on their anticipated need for capacity during system-peak conditions. Capacity Subscription implies that consumers “subscribe” to their anticipated need for capacity during system-peak demand. Physically this is obtained by the use of a fuse-like device that is activated by the System Operator when the system is approaching its capacity limit. Under all other circumstances the “fuse” is inactive and consumption is limited by technical constraints only.

Capacity Subscription has the following attractive properties:

• Willingness to pay for fuses directly mirrors consumers’ preference for quality of supply.
• Capacity installation can be based on consumers’ revealed preferences, instead of surveys where consumers answer hypothetical questions.
• Quality of supply becomes a private good instead of a public good.
• Introduction of demand for fuses enables the creation of a capacity market.
• Although definitely more complicated than an energy-only market, the transaction costs are lower than e.g. for full spot pricing.
• The potential creation of a market for smart devices for local control of demand (“market pull”). This may have a powerful effect on the development of this kind of technology. Thus Capacity Subscription has the property of dynamic efficiency.

Because of these properties, Capacity Subscription has the potential to solve many of the problems related to peaking power in today’s markets. However, it is new, has not been used in practice, and many questions remain. Thus it is a very well suited theme for a competence building project.