Development of Computational Intelligent Units for Optimal Generation Scheduling

Abstract

In this research, the optimal generation scheduling based on particle swarm optimization (PSO) is proposed. The generation scheduling concerns the operation between thermal and hydro power plants. The research aims to minimize the production cost of generators with satisfying technical constraints such as generation limits, power balance, water and reservoir limits, etc. The PSO-based approach is compared to other stochastic optimizers, i.e., genetic algorithm (GA), simulated annealing (SA), and evolutionary programming. The research finds that PSO could provide better solutions for the optimal generation scheduling problem compared to the others. Especially, Self-organizing hierarchical particle swarm optimization with time-varying coefficient (SPSO-TVAC) could provide the least operation cost of the electricity generation. The proposed approach could be applied to a short term planning for power industry. And it could be used to provide an optimal solution for other applications such as energy management.