Memetic Algorithm for Solving Monthly Unit Commitment Problem Considering Uncertain Wind Power

Abstract

The continuous growth of wind power penetration has brought great challenges to the monthly unit commitment of power system. In order to deal with the monthly power generation schedule with large-scale wind power integration and maintain operating system economy and reliability, it is important to incorporate the interval prediction information of wind power under various confidence intervals into monthly unit commitment model considering risk costs caused by wind power curtailment and load shedding. To solve the above model, this paper proposes memetic algorithm based on a combination of global and local search, which introduces guided local search in the process of global search using genetic algorithm for the monthly unit commitment model with uncertain wind power. The excellent representatives selected from the local regions are used as the genetic operators to ensure the local extremum could be outputted in each iteration. In addition, the cost objective function is modified in time to change the terrain of search space, so as to avoid falling into the local optimal solution. Finally, the testing systems verify the validity and computation efficiency of the proposed method and algorithm.