基于粒子群算法的电力变压器噪声有源控制系统优化  

Optimization of AANC System in power transformer by particle swarm optimization

作　　者：刘姜涛[1,2] 应黎明[2] 

机构地区：[1]湖北第二师范学院物理与电子信息学院,湖北武汉430205 [2]武汉大学电气工程学院,湖北武汉430072

出　　处：《武汉大学学报（工学版）》2014年第4期498-501,共4页Engineering Journal of Wuhan University

基　　金：国家自然科学基金项目(编号:51377123)

摘　　要：误差传感器的位置对有源降噪系统理想降噪量影响很大,理论上评价降噪效果的优化准则是初级声源和次级声源的总辐射声功率最小.鉴于粒子群算法的找寻效率较高,采用粒子群算法来找寻使得总辐射声功率最小的误差传感器最佳位置.仿真研究表明,粒子群优化算法应用于误差传感器优化布放,收敛速度快,实现更灵活方便;两个误差传感器比单个误差传感器得到更高的降噪量,双次级声源比单次级声源得到更高的降噪量.对电力变压器自适应有源噪声控制系统的设计有理论指导意义.The location of the error sensors rains influence on ideal noise reduction.In theory,it is the optimization criterion to make total radiation acoustic power of primary source and secondary source achieve a minimum.In view of the high efficiency of particle swarm optimization（PSO）,it is used to search the optimal position of error sensors.The simulation study shows that the PSO applied to search optimal position of error sensors has fast convergence rate and is implemented more flexibly and conveniently than other algorithms.The adaptive active noise control（AANC）system with two error sensors and pairs of secondary sound source is superior to that with a single error sensor and a single secondary source.The study has theoretical significance to design an adaptive active noise control system for power transformer.

关 键 词：有源噪声控制 粒子群优化 电力变压器 声源 

分 类 号：TM912[电气工程—电力电子与电力传动]