基于多传感器融合声纹特征图谱的变压器铁芯松动故障诊断方法

Abstract
Figure/Table
References (23)
Related Citation (15)

Download: PDF (3559 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract The core looseness fault leaves a huge hidden trouble on the safe and stable operation of the power transformer, a practical and reliable diagnostic method has not yet been presented. In this paper, a multi-sensor-fusion voiceprint based diagnosis method core looseness fault of power transformer is proposed. Firstly, four sensors are used to collect the time-domain series of the voiceprint, which are converted to voiceprints by the wavelet transform. The weight of different voiceprints, which are achieved by different sensors, are calculated by the entropy weight method. By weighted fusion, a multi-sensor-fusion voiceprint is generated. Secondly, the multi-sensor-fusion voiceprint is input into the improved ShuffleNetV2 model, where the fault degree is identified by group convolution and channel shuffling. Finally, the effectiveness of the method is verified experimentally. Experimental results show that the proposed method can reliably diagnose the winding faults with looseness of 25%, 50%, 75% and 100%, and the accuracy rate is up to 99.6%. Compared with conventional voiceprints, such as FFT, GAF, MTF, and RP, the diagnosis accuracy is increased by 12.2%. Compared with single-sensor voiceprint the diagnosis accuracy is increased by 5.8%. Compared with other neural network models, such as AlexNet , MobilleNetV2, GoogleNet and ResNet, the diagnosis accuracy is increased by 2.7%.

Key words： power transformer core looseness fault voiceprint multi-sensor fusion neural network models

Received: 22 March 2022 Published: 15 August 2023

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	LI Nan1
	MA Hongzhong1
	DUAN Dawei1
	ZHU Hao1
	HE Ping2

Cite this article:

LI Nan1,MA Hongzhong1,DUAN Dawei1, et al. Fault diagnosis method for transformer core looseness based on multi-sensor fusion voiceprint feature map[J]. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(15): 129-137.

URL:

http://jvs.sjtu.edu.cn/EN/ OR http://jvs.sjtu.edu.cn/EN/Y2023/V42/I15/129

[1] 杨毅,刘石,张楚,韩丹,孟源源,胡异炜,郑婧,黄海.基于振动分布特征的电力变压器绕组故障诊断[J].振动与冲击,2020,39(01):199-208.
Yang Yi, Liu Shi, Zhang Chu, Han Dan, Meng Yuanyuan, Hu Yiwei, Zheng Jing, Huang Hai. Fault diagnosis of power transformer windings based on vibration distribution characteristics [J]. Vibration and Shock, 2020, 39(01): 199 -208.
[2] 鲁文波,曲光磊.油浸式自耦变压器振动噪声研究[J].振动与冲击,2019,38(15):273-280.
Lu Wenbo, Qu Guanglei. Research on vibration and noise of oil-immersed autotransformer [J]. Vibration and Shock, 2019, 38(15): 273-280.
[3] 胡勇,程蕾.大型电力变压器故障实例统计分析[J].电力安全技术,2003(01):20-22.
Hu Yong, Cheng Lei. Statistical Analysis of Large-scale Power Transformer Fault Cases [J]. Electric Power Safety Technology, 2003(01): 20-22.
[4] 邓永辉.变压器类设备典型故障案例汇编2006-2010[M].北京：中国电力出版社,2012.
Deng Yonghui. Compilation of Typical Fault Cases of Transformer Equipment 2006-2010 [M]. Beijing: China Electric Power Press, 2012.
[5] 杨毅,王丰华,段若晨,杜胜磊,刘石,杨贤.基于自适应筛选EMD和CFDC的变压器绕组状态检测[J].振动与冲击,2017,36(19):106-111+144.
Yang Yi, Wang Fenghua, Duan Ruochen, Du Shenglei, Liu Shi, Yang Xian. Transformer winding state detection based on adaptive screening EMD and CFDC [J]. Vibration and Shock, 2017, 36(19): 106-111+144 .
[6] 刘丽龙,刘武能,何耿利,冯旭,罗长兵.变压器绕组和铁芯故障检测方法研究[J].电力设备管理,2020(11):174-175.
Liu Lilong, Liu Wuneng, He Gengli, Feng Xu, Luo Changbing. Research on fault detection method of transformer winding and iron core [J]. Power Equipment Management, 2020(11):174-175.
[7] 李鹏,毕建刚,于浩,许渊.变电设备智能传感与状态感知技术及应用[J].高电压技术,2020,46(09):3097-3113.
Li Peng, Bi Jiangang, Yu Hao, Xu Yuan. Intelligent sensing and state sensing technology and application of substation equipment [J]. High Voltage Technology, 2020, 46(09): 3097-3113.
[8] 齐波,王一鸣,张鹏,温钊,李成榕,王红斌.基于自决策主动纠偏的电力变压器油色谱诊断模型[J].高电压技术,2020,46(01):23-32.
Qi Bo, Wang Yiming, Zhang Peng, Wen Zhao, Li Chengrong, Wang Hongbin. A chromatographic diagnostic model of power transformer oil based on self-decision and active rectification [J]. High Voltage Technology, 2020, 46(01): 23-32.
[9] 丁登伟,张星海,兰新生.HVDC单极运行对500kV交流变压器的振动影响分析研究[J].振动与冲击,2016,35(17):201-206.
Ding Dengwei, Zhang Xinghai, Lan Xinsheng. Analysis and research on the influence of HVDC unipolar operation on the vibration of 500kV AC transformer [J]. Vibration and Shock, 2016, 35(17): 201-206.
[10] 王荣昊,李喆,孙正,胡赵宇,孙汉文,江秀臣.基于FISVDD与GRU的变压器声纹识别技术[J/OL].高电压技术:1-12[2022-03-04].
Wang Ronghao, Li Zhe, Sun Zheng, Hu Zhaoyu, Sun Hanwen, Jiang Xiuchen. Transformer Voiceprint Recognition Technology Based on FISVDD and GRU [J/OL]. High Voltage Technology: 1-12 [2022-03-04].
[11] 耿琪深,王丰华,金霄.基于Gammatone滤波器倒谱系数与鲸鱼算法优化随机森林的干式变压器机械故障声音诊断[J].电力自动化设备,2020,40(08):191-196+224+197-199.
Geng Qishen, Wang Fenghua, Jin Xiao. Sound diagnosis of dry-type transformer mechanical fault based on Gammatone filter cepstral coefficient and whale algorithm optimization of random forest [J]. Electric Power Automation Equipment, 2020,40(08):191-196+ 224+197-199.
[12] 张重远,罗世豪,岳浩天,王博闻,刘云鹏.基于Mel时频谱-卷积神经网络的变压器铁芯声纹模式识别方法[J].高电压技术,2020,46(02):413-423.
Zhang Chongyuan, Luo Shihao, Yue Haotian, Wang Bowen, Liu Yunpeng. Transformer core voiceprint pattern recognition method based on Mel-time spectrum-convolutional neural network [J]. High Voltage Technology, 2020, 46(02): 413-423.
[13] 王丰华,王邵菁,陈颂,袁国刚,张君.基于改进MFCC和VQ的变压器声纹识别模型[J].中国电机工程学报,2017,37(05):1535-1543.
Wang Fenghua, Wang Shaojing, Chen Song, Yuan Guogang, Zhang Jun. Transformer voiceprint recognition model based on improved MFCC and VQ [J]. Chinese Journal of Electrical Engineering, 2017, 37(05): 1535-1543.
[14] 周东旭,王丰华,党晓婧,张欣,刘顺桂.基于压缩观测与判别字典学习的干式变压器声纹识别[J].中国电机工程学报,2020,40(19):6380-6390.
Zhou Dongxu, Wang Fenghua, Dang Xiaojing, Zhang Xin, Liu Shungui. Voiceprint recognition of dry-type transformers based on compression observation and discriminative dictionary learning [J]. Chinese Journal of Electrical Engineering, 2020, 40(19): 6380-6390.
[15] Zhang X , Zhou X , Lin M , et al. ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City : IEEE ,2017.
[16] Ma N , Zhang X , Zheng H T , et al. ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design[J]. Springer, Cham, 2018.
[17] 张占龙,肖睿,武雍烨,蒋培榆,邓军,潘志城.换流变压器振动信号多层次特征提取模型研究[J].中国电机工程学报,2021,41(20):7093-7104.
Zhang Zhanlong, Xiao Rui, Wu Yongye, Jiang Peiyu, Deng Jun, Pan Zhicheng. Research on multi-level feature extraction model of converter transformer vibration signal [J]. Chinese Journal of Electrical Engineering, 2021, 41(20): 7093-7104.
[18] 刘宝稳,汤容川,马钲洲,马宏忠,许洪华.基于S变换D-SVM AlexNet模型的GIS机械故障诊断与试验分析[J].高电压技术,2021,47(07):2526-2538.
Liu Baowen, Tang Rongchuan, Ma Zhengzhou, Ma Hongzhong, Xu Honghua. GIS mechanical fault diagnosis and test analysis based on S-transform D-SVM AlexNet model [J]. High Voltage Technology, 2021, 47(07): 2526-2538.
[19] 曾全昊,王丰华,郑一鸣,何文林.基于卷积神经网络的变压器有载分接开关故障识别[J].电力系统自动化,2020,44(11):144-151.
Zeng Quanhao, Wang Fenghua, Zheng Yiming, He Wenlin. Fault identification of transformer on-load tap-changer based on convolutional neural network [J]. Automation of Electric Power Systems, 2020, 44(11): 144-151.
[20] 洪翠,连淑婷,黄晟,郭谋发.基于改进经验小波变换和改进多视角深度矩阵分解的直流配电网故障检测方案[J/OL].电力自动化设备:1-9[2022-03-17].
Hong Cui, Lian Shuting, Huang Sheng, Guo Moufa. Fault detection scheme based on IEWT and IMDMF for DC distribution network [J/OL]. Electric Power Automation Equipment: 1-9 [2022] -03-17].
[21] Wang Ting, Chen Kun, Zhang Kanjun, Du Zheng’an,Li Jun,Dai Di. Mixed Weibull distribution model of DC protection system based on entropy weight method[C]// 2015 IEEE Conference on Electrical Insulation and Dielectric Phenomena. Ann Arbor :CEIDP,2015.
[22] Howard A , Sandler M , Chen B , et al. Searching for MobileNetV3[C]// 2019 IEEE/CVF International Conference on Computer Vision.Seoul :ICCV, 2020.
[23] GB/T 1094.10-2003. 电力变压器　第10部分：声级测定[S]. 1987
GB/T 1094.10-2003. Power transformers Part 10: Sound level determination[S]. 1987

[1]	. [J]. JOURNAL OF VIBRATION AND SHOCK, 2024, 43(9): 13-.