基于自适应遗传算法的无源电力滤波器综合优化方法

ADAPTIVE GENETIC ALGORITHM BASED OPTIMAL DESIGN APPROACH FOR PASSIVE POWER FILTERS

无源电力滤波器经济而高效,是电网谐波治理的重要设备。现有设计方法普遍针对部分经济技术指标,依据工程经验或简化模型近似求解,故优化程度较低。该文综合考虑谐波抑制、无功补偿和初期投资等设计目标,提出一种基于改进遗传算法和PSpice仿真的优化设计新方法。实验结果表明,凭借高效实用的元件标称值编解码方案,权值动态调整多目标评估方法,以及兼顾种群多样性和遗传进程的遗传参数调整策略,该方法可获得综合优化的实用性设计结果。

Passive Power Filters (PPFs) have been widely used to absorb harmonics increasing in high power distribution/transmission systems, because of their low cost and high efficiency. Generally based on engineering experiences and/Or over-simplified models, conventional design approaches of PPFs focus on only parts of relevant technical and economical specifications, so they are not capable of finding out the optimal solution but some approximate ones. To meet the main optimizing targets of PPFs including harmonic rejection, reactive power compensation and original investment, this paper proposes a novel optimal design approach' based on an improved genetic algorithm and PSpice simulation。By virtue of an efficient and effective encoding-decoding scheme based on the standard series of component values, a multi-objective fitness evaluation mechanism with PSpice simulation and weight vectors dynamic adjusting, and an adaptation technique for genetic parameters to maintain the diversity of individuals and follow the genetic process, this approach has been validated by simulation results that it can present effective design results of PPFs in accordance with the optimizing targets.