基于声纹特征识别的电力变压器运维检测技术研究及性能评估  

Research and Performance Evaluation of Power Transformer Operation and Maintenance Detection Technology Based on Voice Print Feature Recognition

作　　者：申国标 陈浩 李德成 刘寿光 李章勇 SHEN Guobiao;CHEN Hao;LI Decheng;LIU Shouguang;LI Zhangyong

机构地区：[1]云南电网有限责任公司文山供电局,云南文山663000

出　　处：《电力系统装备》2024年第5期110-111,114,共3页Electric Power System Equipment

摘　　要：文章开发了一种基于声纹特征识别的电力变压器运维检测技术,其利用梅尔频率倒谱系数结合深度置信网络和支持向量数据描述算法,提升了声纹信号分析的准确度和效率。试验表明,DBN–SVDD算法在变压器缺陷识别中准确率达97.94%,为智能电网的可靠运行提供了技术支持。The article develops a power transformer operation and maintenance detection technology based on voiceprint feature recognition,which combines Mel frequency cepstral coefficients with deep confidence networks and support vector data description algorithms to improve the accuracy and efficiency of voiceprint signal analysis.The experiment shows that the DBN-SVDD algorithm has an accuracy of 97.94%in transformer defect recognition,providing technical support for the reliable operation of smart grids.

关 键 词：电力变压器检测 声纹特征识别 MFCC DBN SVDD 

分 类 号：TM855[电气工程—高电压与绝缘技术]