基于多种群遗传算法的波浪发电最大功率跟踪控制

波浪发电系统遗传算法最大功率点跟踪过程中,因群体中的所有个体较快趋于单一化而停止进化,导致难以获得最优解,为此引入多种群遗传优化新算法。在初始阶段,新算法引入多个种群同时进行搜索,并对每个种群赋予不同的交叉、变异概率,使算法能够兼顾全局与局部搜索;同时加入用于维持种群间联系的移民算子及可用来建立精华种群的人工选择算子,并以精华种群作为算法收敛的判据。仿真结果表明,与传统遗传算法相比,该算法能够提高波浪发电系统的波浪能捕获率。

Maximum power point tracking algorithm based on multiple population genetic algorithm for wave power systems

All individual of the population tends to the same state quickly and stop evolution in genetic algorithm (GA),therefore,GA has difficulty in discovering the optimum solution in the maximum power point tracking (MPPT) of the wave generation system.A novel multiple population genetic algorithm (MPGA) was proposed to solve the problem of the traditional GA.MPGA is introduced to search simultaneously at the beginning.Different MPGAs were given different crossover probabilities and mutation probabilities so that the novel algorithm can balance global search and local search.At the same time,immigration operator was added to maintain the connection between population and artificial selection operator was used to establish quintessence population.The criterion for the convergence of the algorithm was based on the quintessence population.The simulation results show that this algorithm can improve the capture rate of wave energy of the wave generation system.