Chance-Constrained Optimal Hedging Rules for Cascaded Hydropower Reservoirs

Abstract

A multiobjective model with chance constraints is proposed to obtain hedging rules for hydropower reservoirs based on initial reservoir operating rules derived using stochastic dynamic programming. In the model, hedging rules are evaluated through simulations and the maximizations of minimum powers in simulation horizon of hydropower reservoirs are taken as objectives. The maximum-minimum objectives are substituted by aggregated functions. A genetic algorithm is used to solve the model, with a fuzzy optimal method evaluating the individuals. A case study shows that the proposed approach is feasible and efficient to produce a series of hedging rules with different reliability levels, and valuable to balance the magnitudes of generation below firm power and the chances of such events.