Cultural algorithm based long-term optimization scheduling of cascaded Hydro-Plant

A novel approach for long-term optimization scheduling of cascaded hydro-plant reservoirs using cultural algorithm (CA) is presented in this paper. By using of evolutionary programming(EP) model, the situation knowledge and the normative knowledge in belief space are constituted by refining and reasoning the experiences of excellent population which transmitted by the function of accept(). Which in turn to guide population evolution. A detailed mathematical model of a long-term scheduling mathematical model based on annual maximum electric power output of cascaded Hydro-Plant was established. The simulation result for three hydro-plants demonstrates that CA has more powerful global, local searching ability and more fast convergence velocity than Genetic algorithm (GA). The scheduling result can increase 2.32 hundred million kWh compared to GA. The CA can offer a new optimization method and thought for large-scale cascaded hydro-Plant scheduling.