Multi-purpose operation optimization of cascade hydroelectric power stations based on improved accelerating genetic algorithm

In this study, improved accelerating genetic algorithm was applied in the mid-and-long term operation optimization of cascade hydroelectric power stations. A mid-and-long term operation optimization model was constructed with the target functions being the maximization of cascade power capacity and probability of power generation. Combined with simulated annealing thinking, penalty factor was established to turn the two targets into an integrated one. With this algorithm, local searching method was pitched into accelerating genetic algorithm to improve the global optimization ability of this algorithm, and the possibility of curse of dimensionality when applying dynamic planning in the solving of complex optimization problems can be minimized. It was shown by the calculation results that schedule schemes with high timeliness, even distribution and good convergence can be obtained with this algorithm. Through coupling analysis of the relationship between scheduling targets, scientific guide can be provided for the multi-objective scheduling of cascade hydroelectric power station.