2油纸绝缘热老化阶段识别.docVIP

基于随机森林的油纸绝缘老化阶段评估* 张建文，王曼，解浩，严家明，张寰宇 (中国矿业大学 电气与动力工程学院，江苏 徐州，221008) 摘要：为了更准确地评估变压器油纸绝缘老化阶段，提出了一种基于EMD-SVD特征和随机森林分类器相结合的识别方法。通过实验平台采集到对应于不同热老化时间局部放电信号的气隙缺陷样本，在消噪后用EMD-SVD两次分解并提取相应老化特征量，并分别利用随机森林分类器与传统分类器BP神经网络和支持向量机对EMD-SVD特征进行分类识别。结果显示随机森林分类器识别效果优于传统分类器，对于提取的油纸绝缘局部放电信号EMD-SVD特征，随机森林分类器分类能力更强。分析首次将EMD-SVD特征与随机森林分类器相结合应用在识别油纸绝缘热老化阶段方面能够取得更好的效果。 关键词：局部放电信号；EMD-SVD特征；随机森林；油纸绝缘热老化识别 中图分类号：TM855 文献标识码：A 文章编号：1001-1390(2017)00-0000-00 Aging stage evaluation of oil-paper insulation based on random forest Zhang Jianwen, Wang Man, Xie Hao, Yan Jiaming, Zhang Huanyu (School of Electrical and Power Engineering Engineering, China University of Mining and Technology, Xuzhou , 221008, Jiangsu, China) Abstract: This paper proposes the diagnostic method which is based on EMD-SVD feature extraction and random forest classifier in order to more accurately assess the aging stage of transformer oil paper insulation aging stage. The air gap defect samples corresponding to the partial discharge signals at different thermal aging time were collected through the experimental platform. After denoising, the The EMD-SVD was decomposed twice and the corresponding aging features were extracted after de-noising. The EMD-SVD feature are is diagnosed by random forest classifier with the traditional BP neural network classifier and support vector machine respectively, and the results show that random forest classifier recognition result is superior to the traditional classifier. Compared with the traditional classifier, the classification ability of random forest classifier classification ability is better for EMD-SVD characteristics classification. The paper demonstrates that the EMD-SVD features combined with random forest classifier applied in oil paper insulation thermal aging phase recognition effect is better. Keywords: partial discharge signal, EMD-SVD feature, random forest, oilpaper insulation thermal aging identification 引 言 *基金项目：十三五国家重点研发专项(2016YFC0801808);中国