A hybrid of real coded genetic algorithm and artificial fish swarm algorithm for short-term optimal hydrothermal scheduling

Abstract

The short-term hydrothermal scheduling (SHS) is a complicated nonlinear optimization problem with a series of hydraulic and electric system constraints. This paper presents a hybrid algorithm for solving SHS problem by combining real coded genetic algorithm and artificial fish swarm algorithm (RCGA–AFSA), which takes advantage of their complementary ability of global and local search for optimal solution. Real coded genetic algorithm (RCGA) is applied as global search, which can explore more promising solution spaces and give a good direction to the global optimal region. Artificial fish swarm algorithm (AFSA) is used as local search to obtain the final optimal solution for improving the exploitation capability of algorithm. The water transport delay between connected reservoirs is taken into account in this paper. Moreover, new coarse and fine adjustment methods without any penalty factors and extra parameters are proposed to deal with all equality and inequality constraints. To verify the feasibility and effectiveness of RCGA–AFSA, the proposed method is tested on two hydrothermal systems. Compared with other methods reported in the literature, the simulation results obtained by hybrid RCGA–AFSA are superior in fuel cost and computation time.