Extensions for Benders cuts and new valid inequalities for solving the European day-ahead electricity market clearing problem efficiently

Abstract

We study the day-ahead electricity market clearing problem under the prevailing market design in the European electricity markets. We revisit the Benders decomposition algorithm that has been used to solve this problem. We develop new valid inequalities that substantially improve the performance of the algorithm. We generate instances that mimic the characteristics of past bids of the Turkish day-ahead electricity market and conduct experiments. We use two leading mixed-integer programming solvers, IBM ILOG Cplex and Gurobi, in order to assess the impact of employed solver on the algorithm performance. We compare the performances of our algorithm, the primal-dual algorithm, and the Benders decomposition algorithm using the existing cuts from the literature. The extensive experiments we conduct demonstrate that the price-based cuts we develop improve the performance of the Benders decomposition algorithm and outperform the primal-dual algorithm.