Chaotic Particle Swarm Optimization Algorithm with Niche and its Application in Cascade Hydropower Reservoirs Operation

Niche evolutionary strategy and chaotic searching were introduced into PSO, called as chaotic particle swarm optimization algorithm with niche (CNPSO) in this thesis. Restricted competition selection method was used to establish niche, in which each species excluded each other and dynamically formed their own searching spaces, effectively maintain the diversity of the species, so as to avoid local convergence. The chaotic searching further improved the global optimization searching precision. CNPSO was programmed to do the optimization regulation of 14 cascade hydropower stations with giant reservoirs by 48-year run-off series. With the result showing that CNPSO is highly efficient in optimization searching, capable of solving the complicated multi dimensional, strong-constraint, multi-states, multi-stages and non-linear problems such as optimization regulation of cascade hydropower stations with giant reservoirs.