WU Zi-jing, ZHENG Dian-chun, ZHANG Zheng-han, et al. The WFCM Recognition Method of Gas Corona Discharge Based Image Gray Features[J]. 2010, 15(2): 14-18.
WU Zi-jing, ZHENG Dian-chun, ZHANG Zheng-han, et al. The WFCM Recognition Method of Gas Corona Discharge Based Image Gray Features[J]. 2010, 15(2): 14-18.DOI:
基于图像灰度矩特征的气体电晕放电WFCM辨识方法
摘要
模拟了以气体为绝缘介质的高压电器的放电现象
根据气隙几何构型差异
构建了极不均匀场、不均匀场及稍不均匀场电极模型
并根据图像的中心矩不随坐标变换而改变的特性
采集了模型电极间隙的气体电晕放电信息
根据其信息的时空分布
计算了放电信息图像的灰度矩
归一化处理后作为气体介质失效诊断的模式辨识特征量.在分析了传统算法中模糊聚类算法的基础上
详细论述了加权模糊c-均值聚类算法
实验结果表明此方法充分利用了放电信息样本点的密度信息量
有效弥补了传统算法中存在的不足
提高了模式辨识算法收敛速度.
Abstract
In this paper
three kinds of models are structured to simulating the GCD(gas corona discharge) phenomena and sampling spatio-temporal information of the GCD signals under different electric field strengths
and the fuzzy C-means(FCM) and the weighted fuzzy C-means(WFCM) clustering algorithm are explicitly represented and compared.Moreover
the grey intensity moment features of GCD signal waveforms are extracted and taken as input vector matrix for realizing the pattern recognition