A Stochastic Programming Approach to Power Portfolio Optimization

A STOCHASTIC PROGRAMMING APPROACH TO POWER PORTFOLIO OPTIMIZATION Suvrajeet Sen Lihua Yu Talat Genc Raptor Laboratory, SIE Department, University of Arizona, Tucson, AZ 85721,U.S.A. September 20,2004 We consider a power portfolio optimization model that is intended as a decision aid for scheduling and hedging (DASH) in the wholesale power market. Our multi-scale model integrates the unit commitment model with financial decision-making by including the forwards and spot market activity within the scheduling decision model. The methodology is based on a multi-scale stochastic programming model that selects portfolio positions that perform well on a variety of scenarios generated through statistical modeling and optimization. When compared with several commonly used fixed-mix policies, our experiments demonstrate that the DASH model provides significant advantages.