Skip to Main Content
This paper presents an improved binary particle swarm optimization algorithm (IBPSO) to solve short-term thermal unit commitment. Unit commitment (UC) is a challenging optimization problem in the power system operation. The NP-Hardness of the UC motivates us to develop metaheuristics algorithm to solve it approximately. PSO is one of relatively current metaheuristics. When implementing the PSO to UC, we derived two strategies to improve the binary particle swarm optimization algorithm. One is asynchronous time-varying learning strategy and another is a new repairing strategy for particles. In order to verify the performance of the proposed PSO, Lagrangian relaxation is used to find lower bound of UC. A computational experiment is carried out on randomly generated instances. The numerical results show that the IBPSO may obtain better solution within reasonable computational time.