基于多粒度知识特征和Transformer网络的电力变压器故障声纹辨识方法  

A Voiceprint Classification Method for Power Transformer Fault Identification Based on Multi-granularity Knowledge Features and Transformer Network

作　　者：齐子豪 仝杰 张中浩 龙天航 唐鹏飞 黄灿 QI Zihao;TONG Jie;ZHANG Zhonghao;LONG Tianhang;TANG Pengfei;HUANG Can(China Electric Power Research Institute,Haidian District,Beijing 100192,China)

机构地区：[1]中国电力科学研究院有限公司,北京市海淀区100192

出　　处：《中国电机工程学报》2025年第4期1311-1322,I0008,共13页PROCEEDINGS OF THE CHINESE SOCIETY FOR ELECTRICAL ENGINEERING

基　　金：国家电网有限公司科技项目(5108-202218280A-2-398-XG)。

摘　　要：变压器在机械故障发生时伴随异常声响,基于声纹的机械故障辨识因其准确率高、发现及时和非侵入性等优势,成为当前研究热点。然而声纹信号易受噪声影响,分析处理速度慢且故障数据较难获取,因此如何在强干扰、小样本的情况下实现对机械故障声纹的快速准确辨识,成为当前研究难点。为解决上述问题,该文首先引入物理机理和经验知识,对变压器特征参数进行提取重组,构造多粒度知识特征向量并搭建改进型Transformer网络,大幅提高辨识方法的鲁棒性和抗噪声能力;其次,通过搭建卷积自编码器进行特征降维和模型压缩,缩短模型训练时间,提高机械故障辨识速度;最后,采用跨模态迁移学习技术,在ImageNet-lk数据集上进行预训练并进行知识迁移,解决训练样本不足的问题。相较于传统时序序列深度学习方法,所提方法在高噪声环境下(SNR=−16 dB),准确率均有所提高,实验结果证明,所提方法在准确性、鲁棒性、泛化性等方面均有显著提升,为在复杂环境下基于声纹实现变压器机械故障辨识提供了一种可靠解决方案。The power transformer makes abnormal sounds when mechanical failures occur.The mechanical fault identification based on voiceprint signals has become a current research hotspot due to its high accuracy,timely detection,and non-invasiveness.However,voiceprint signals are easily affected by noise and difficult to obtain,and the processing speed is slow.Therefore,how to achieve rapid and accurate identification of mechanical fault based on voiceprint signals in the presence of strong noise and small sample sizes has become a current research difficulty.To address the aforementioned issues,this paper first incorporates physical principles and empirical knowledge to extract feature and builds an improved Transformer network,significantly enhancing the noise resistance.Then,a convolutional autoencoder for model compression is constructed to shorten the training time.Finally,this paper employs cross-modal Transfer Learning by pretraining the model on the ImageNet-1k dataset to address the issue of limited training samples.Compared to traditional time-series deep learning methods,the proposed method achieves higher accuracy in a high-noise environment(SNR=−16 dB).Experimental results demonstrate significant improvements in accuracy,robustness,and generalization.This work provides a reliable solution for implementing power transformer mechanical fault identification based on voiceprint signals in complex environments.

关 键 词：机械故障 故障辨识 声纹识别 自编码器 Transformer 

分 类 号：TM71[电气工程—电力系统及自动化]