基于思维进化算法的变压器局部放电源定位新方法

A New Method to Locate Partial Discharge Sources within Transformer Based on Mind Evolutionary Computation

为准确对变压器局部放电源的位置定位,寻求先进的优化计算方法是解决该问题的有效途径。把思维进化算法(MEC)用于变压器局部放电源"声-声"定位。实验结果表明,与基于标准遗传定位算法比较,基于思维进化方法的定位具有精度高、收敛速度快、避免早熟等特点,做到对放电源位置实时在线定位。

Mind Evolutionary Computation is a new evolutionary computing algorithm.The algorithm inherits ＂group＂ and ＂evolution＂ of evolutionism,and adopts ＂similartaxis＂ and ＂dissimilation＂ operators to achieve global search.This paper used the mind evolutionary algorithm for ＂sound-sound＂ ultrasonic locating of partial discharge.The experimental results indicate that,compared with standard genetic algorithm,the MEC algorithm had fast convergence,high precision,and prematurity prevention in real-time on-line location of discharge source.