基于智能算法的局部放电模式分层识别综合实验设计  

Comprehensive experiment design of partial discharge pattern layered recognition based on intelligent algorithm

作　　者：严家明 周一恒 徐瑞东[1] 邓先明[1] YAN Jiaming;ZHOU Yiheng;XU Ruidong;DENG Xianming(School of Electrical Engineering,China University of Mining and Technology,Xuzhou 221116,China;School of Information and Control Engineering,China University of Mining and Technology,Xuzhou 221116,China)

机构地区：[1]中国矿业大学电气工程学院,江苏徐州221116 [2]中国矿业大学信息与控制工程学院,江苏徐州221116

出　　处：《实验技术与管理》2023年第2期48-56,共9页Experimental Technology and Management

基　　金：教育部第二批新工科研究与实践项目(E-DZYQ20201416);中国矿业大学教学研究重点项目(2021ZD01);中国矿业大学实验技术研究与开发重大项目(S2020Z003);国家重点研发计划项目(2017YFF0210600);煤炭行业教育研究课题(2021MXJG082)。

摘　　要：该文设计了局部放电模式识别实验,改进了基于智能算法的样本分层实验分析方法,建立了基于教研融合的实验教学新模式。通过设计三种放电缺陷采集局部放电信号,利用小波变换清洗局部放电数据,基于统计学构建局部放电相位分布(PRPD)图谱并提取特征量,对比分析了利用BP神经网络(BPNN)传统算法和改进的样本分层算法对放电类型识别的效率。实验结果表明,样本分层智能算法能够提高局部放电模式识别效率。该实验基于教师科研成果,综合运用了电气、信息、统计学及计算机等多学科交叉,建立了“基于实践、递进启发、学科交叉”的教研融合实验教学新模式,锻炼了学生的实践动手能力和知识综合运用能力,培养了学生的探索热情和创新意识。In this paper,a partial discharge(PD)pattern recognition experiment was designed,a sample-layered experiment analysis method based on intelligent algorithm was improved,and a new experiment-teaching mode based on teaching and research integration was established.Three kinds of discharge defects were designed to acquire partial discharge signals,partial discharge data was cleaned by wavelet transform,partial discharge phase distribution(PRPD)maps were constructed based on statistics and their statistics features were extracted,efficiency of discharge pattern recognition by using traditional BP neural network(BPNN)algorithm and improved sample-layered algorithm was compared and analyzed.The experiment results show that sample-layered intelligent algorithm can improve efficiency of discharge pattern recognition.Based on teacher’s scientific research achievements,this experiment comprehensively utilizes interdisciplinary ways of electricity,information,statistics,computer and so on,and establishes a new experiment teaching and research integration mode of“based on practice,progressive inspiration,interdisciplinary crossing”,which trains students’practice ability and comprehensive application ability of knowledge,and cultivates students’exploration enthusiasm and innovation consciousness.

关 键 词：放电缺陷 PRPD模式 统计特征 样本分层识别 教研融合 

分 类 号：TM835[电气工程—高电压与绝缘技术] G642.423[文化科学—高等教育学]