基于交互式多目标遗传算法的无源滤波器优化设计

分析了传统无源滤波器优化设计方法的优缺点 ,提出了一种基于交互式多目标遗传算法的无源滤波器优化设计方法。该方法用遗传算法对无源滤波器的参数进行优化 ,通过非精确偏好信息引导遗传搜索 ,使得种群朝着多个目标的最佳综合性能方向进化。引入分享法来增加种群的多样性 ,最终获得与决策者意愿相吻合的解集。较好地实现了无源滤波器成本、电压电流谐波和无功功率补偿的多目标优化设计。实验结果表明此无源滤波器的成本可以降低约 3 5 % ,滤波效果却有明显的提高 ,具有良好的理论和工程应用价值

Optimal Design of Passive Power Filters Based on Interaction Multi-modal Genetic Algorithm

This paper analyzes the merits and demerits of the traditional passive power filter design method,presems an optimal passive power filter design method based on interaction multi modal genetic algorithm This method,which inducts population to evolve into multi modal optimal comprehensive function and introduces soaring method to raise the diversification of population,obtains the set of solution that is good fit the policymaker's demand by means of inaccurate preference information guiding genetic search,realizes the multi modal optimal design of the cost of passive power filter system,the voltage and current harmonics and the reactive power compensation.The experimental result shows that this design cuts down about 3 5% cost,obtains fine filter effect comparing with traditional design and is valuable in the field of theory and engineering application.