改进遗传算法在梯级电站日优化运行中的应用

An Improved Genetic Algorithm for Shortterm OptimalCascaded Hydroelectric Stations Scheduling

针对基本遗传算法存在求解精度与收敛速度间的矛盾,研究了一种新的改进自适应遗传算法(IAGA),并用于求解梯级水电站日优化调度问题。基于提出的评价种群"早熟"程度新指标,自适应调整遗传算法的交叉概率和变异概率,同时依据当前最优个体解码所对应调度方案对约束条件的违反量,动态调整相应的惩罚系数,使调度方案获得期望的约束程度。仿真结果验证了算法的有效性和可靠性。

A novel Improved Adaptive Genetic Algorithm(IAGA) is applied to the problem of determining the optimal hourly schedule of power generation in a cascaded hydroelectric system. Based on the quantitative analysis of population diversity, i.e. the premature convergence of GA, the autotunings of the crossover probability and the mutation probability are considered. Moreover, according to the violation of the constraints, the corresponding penalty coefficients are regulated. As a result, the available scheme is not beyond the expected bounds. The simulation results verify that the proposed IAGA is effective and reliable.