电测与仪表Issue(10)：6-9,32,5.
基于灰度矩特征的局部放电模式识别
Partial Discharge Pattern Recognition Based on Gray Moment
摘要
Abstract
A novel feature extraction method is put forward in this paper because he number of features needed for successful classification depends on the discriminatory quality of the chosen partial discharge (PD) features. First the φ-q-n pattern is fragmented, shifted and remerged to form two parts, each with a 180°phase angle. Then both parts are projected onto the φ-q plate to transform to a 2D digital image. Finally the moments and central moments are calculated. BP Neural Network (BPNN) is invoked as the classifier with 6 input vectors (5 gray moments and q). Tests are carried out on three typical PD types, that is, bubble discharge, surface discharge and corona discharge. The recognition rates are 100%.关键词
局部放电/模式识别/BP神经网络/灰度矩Key words
partial discharge/pattern recognition/BP Neural Network/gray moment分类
信息技术与安全科学引用本文复制引用
王瑜,苑津莎,靳松..基于灰度矩特征的局部放电模式识别[J].电测与仪表,2013,(10):6-9,32,5.基金项目
国家自然科学基金资助项目（61204027）；中央高校基本科研业务费专项资金资助项目 （）
