基于稳态繁殖的遗传算法在水库优化调度中的应用

Application of Steady-state Genetic Algorithm to Optimal Operation of Reservoir

针对遗传算法用于求解水库优化调度时收敛局部最优解的问题,以葛洲坝水库为例,提出了基于稳态繁殖的遗传算法,并与基本遗传算法做了比较。结果表明,该方法避免了个别种群迅速繁殖及早熟现象,扩大了搜索空间,更好地保持了种群多样性,并使收敛速度加快。

The traditional genetic algorithm （GA） usually converges to a local optimal solution for solving the optimal operation of reservoirs. A new method based on steady-state breeding GA is proposed to solve optimal operation of Gezhouba Reservoir. Compared with simple GA, the results show that the improved GA not only avoids rapid propaga- tion of population and individual premature phenomena, but also has fast convergence speed by expanding search space and maintaining the diversity of population.