Comparative Analysis of Genetic Algorithms and Particle Swarm Optimization Algorithms for Optimal Reservoir Operation

Apart from traditional optimization techniques, modern heuristic optimization techniques, like genetic algorithms (GA), particle swarm optimization algorithm (PSO) have been widely used to solve optimization problems. This paper deals with comparative analysis of GA and PSO and their applications in a reservoir operation problem. Extensive component analysis, parameter sensitivity analysis of GA and PSO show that both GA and PSO can be used for optimal reservoir operation, but they display different features. GA can obtain very high approximate global optimal solutions of the problem with a high stability and a high computing efficiency, but it can't obtain the problem's accurate global optimal solutions. For GA, population size and mutation rate are two main parameters affect its solution qualities. Comparative to GA, PSO can obtain the accurate global optimal solutions of the problem with a higher computing efficiency, but with a less stability. For PSO, population size and velocity parameter are two main parameters affect its solution qualities.