Solving the Unit Commitment Problem with Improving Binary Particle Swarm Optimization

Unit commitment is a traditional mixed-integer non-convex problem and an optimization task in power system scheduling. The traditional methods of solving the Unit commitment problem have some problems, such as slow solving speed, low accuracy and complex calculation. Therefore, intelligent algorithms have been applied to solve the unit combination problem with continues and discrete feature, such as Particle Swarm Optimization, Genetic Algorithm. In order to improve the solution quality of Unit commitment, this paper proposes the adaptive binary Particle Swarm Optimization with V-shaped transfer function to solve the unit commitment problem, and adopts the policy of the segmented solution. By comparison with some classical algorithm in the same unit model, the experimental results show that solving the UC problem by using improved algorithm with segmented solution has higher stability and lower total energy consumption.