Reducing the cost of serving uncertain loads via the portfolio of supply contracts
Abstract
Uncertain electricity load variation could incur additional costs associated with some specific decisions to resolve its impacts. Price volatility also becomes one of the major concerns for the operation planning and scheduling problem, resulted from the emergence of electricity markets. Since electricity is generated and instantaneously consumed in real time, short-term operation is an important class of decision process, where a procurement strategy is implemented in order to mitigate operational risk from demand uncertainty and financial risk from price volatility while gaining better knowledge of demand and price. With the integration of supply contracts—contractual agreements to provide flexibility and reliability in electricity production—to the traditional unit commitment (UC), a generation company can hedge against such uncertainties while optimizing its operation portfolio value. The optimization model used to solve for the optimal exercising policy is referred to as stochastic unit commitment, which incorporates both the planning model from power systems control and hedging strategy from financial engineering, for a generation portfolio including self generation, forwards, options, ancillary services, and spot sales/purchases. The analyses of supply contracts are so essential that the true contract value should be revealed. With the valuation methods presented in this framework, the values of supply contracts can be interpreted as the incremental flexibility or reliability values provided to enhance the electricity operation portfolio—in other words, to reduce the cost of uncertainty under the exposure to operational and financial risks. The properties of supply contracts, e.g., additivity of contract values and parity of valuation methods, have been investigated to ensure the integrity of model responses. Furthermore, the expected value of perfect information (EVPI), a measure of a bound value on development of forecasting and information systems, will also be discussed in order to illustrate the superiority of the stochastic UC model as compared to the deterministic UC model.
Degree
Ph.D.
Advisors
Sparrow, Purdue University.
Subject Area
Industrial engineering|Operations research
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.