Hydropower Producer Day-ahead Market Strategic Offering Using Stochastic Bi-level Optimization

This paper proposes a bi-level stochastic optimization problem (a Stackelberg game) to generate optimal bids for a profit maximizing hydropower producer and presents a mathematical approach to solve it. The first level represents the strategically acting hydropower producer also called a Stackelberg leader, while the second level represents the transmission system operator (TSO) also called a Stackelberg follower. To solve the bi-level stochastic optimization problem, the second level is replaced by its KKT (Karush-Kuhn-Tucker) optimality conditions, which results in a stochastic MPEC (mathematical program with equilibrium constraints). Finally, the stochastic MPEC is reformulated as a stochastic MILP (mixed integer linear program) using linearization and SOS1 variables (special ordered sets of type 1). Results are reported studying a small case, which point out the impact on the market outcomes when a hydropower producer behaves strategically.