Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章检索

首页 >> 文章检索 >> 最新索引

一种基于SAE-RF算法的配电变压器故障诊断方法

来源:电工电气发布时间:2021-02-23 09:23 浏览次数:680

一种基于SAE-RF算法的配电变压器故障诊断方法

陈锦锋1,张军财1,卢思佳2,高伟2,范贤盛1,陈致远3
(1 国网福建南平供电公司,福建 南平 353000;2 福州大学 电气工程与自动化学院,福建 福州 350108;
3 上海宏力达信息技术股份有限公司,上海 200030)
 
    摘 要:为有效解决配电变压器故障诊断中面临的数据特征人工提取、机器学习调参困难等问题,提出了一种基于堆栈自编码器(SAE)和随机森林(RF)组合的配电变压器故障诊断方法。建立SAE配电变压器故障特征自动挖掘模型,利用大量的无标签数据对SAE模型中的每一个自编码器进行逐层无监督训练,通过贝叶斯优化算法自动选择模型的最优参数;通过有标签数据对模型参数进行有监督细调,挖掘出能够代表各种故障本质属性的特征量;创建一个RF分类器对故障类型进行辨识,调参过程同样实现参数的自动寻优。试验结果表明,所提方法对配电变压器故障诊断准确率达96.67%,显著优于单独使用SAE和RF的分类结果。
    关键词:配电变压器;故障诊断;堆栈自编码器;随机森林;贝叶斯优化
    中图分类号:TM407;TM421     文献标识码:A     文章编号:1007-3175(2021)02-0017-07
 
A Novel Fault Diagnosis Method for Distribution Transformer Via Automatic
Feature Mining and Automatic Parameter Optimization
 
CHEN Jin-feng1, ZHANG Jun-cai1, LU Si-jia2, GAO Wei2, FAN Xian-sheng1, CHEN Zhi-yuan3
(1 State Grid Nanping Power Supply Company, Nanping 353000, China;
2 College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108, China;
3 Shanghai Holystar Information Technology Co., Ltd, Shanghai 200030, China)
 
    Abstract: In order to effectively solve the problems of manual extraction of data features and difficulty of machine learning parameter adjustment in distribution transformer fault diagnosis, a fault diagnosis method for distribution transformer via the combination of stacked autoencoder (SAE) and random forest (RF) is proposed. First, a SAE model is established to realize automatic mining of distribution transformer fault characteristics, and a large number of unlabeled data is used to perform layer-by-layer unsupervised training of each auto-encoder in the model. After that, the optimal parameters of the model are automatically selected by Bayesian optimization algorithm. And then, fine-tune the model parameters through labeled data to mine features that can represent the essential attributes of various faults. Finally, an RF classifier is created to identify the fault type, and the parameter tuning process also realizes automatic parameter optimization. The test results show that the proposed method has an accuracy of 96.67% for distribution transformers fault diagnosis, which is significantly better than the results using SAE and RF alone.
    Key words: distribution transformer; fault diagnosis; stacked auto-encoder (SAE); random forest (RF); Bayesian optimization
 
参考文献
[1] 赵莉华,丰遥,谢荣斌,等. 基于交叉小波的变压器振动信号幅频特征量提取方法[J] . 高电压技术,2019,45(2):505-511.
[2] 王逸萍. 基于最小二乘支持向量机的变压器故障诊断[J]. 电工电气,2016(6):24-27.
[3] 孔德钱,张新燕,童涛,等. 基于差分进化算法与BP 神经网络的变压器故障诊断[J] . 电测与仪表,2020,57(5):57-61.
[4] 丛日立,赵明宇,周洋,等. 基于参数优化的电力变压器故障诊断模型[J] . 电测与仪表,2019,56(22):84-88.
[5] 齐波,王一鸣,张鹏,等. 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(1):23-32.
[6] 周光宇,马松龄. 振动法在线监测变压器的研究现状与发展[J]. 电工电气,2019(3):1-6.
[7] 杨毅,王丰华,段若晨,等. 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.
[8] 武立平,马维青,程胤璋,等. 基于振动信号指标能量的变压器机械故障检测[J] . 电测与仪表,2020,57(6):126-131.
[9] ZHAO Miaoying, XU Gang.Feature extraction of power transformer vibration signals based on empirical wavelet transform and multiscale entropy [J]. IET Science , Measurement & Technology, 2018, 12(1):63-71.
[10] 赵妙颖,许刚. 基于经验小波变换的变压器振动信号特征提取[J]. 电力系统自动化,2017,41(20):63-69.
[11] 王杰,张曦煌. 基于图卷积网络和自编码器的半监督网络表示学习模型[J] . 模式识别与人工智能,2019,32(4):317-325.
[12] 王怀远,陈启凡. 基于堆叠变分自动编码器的电力系统暂态稳定评估方法[J] . 电力自动化设备,2019,39(12):134-139.
[13] 崔广新,李殿奎. 基于自编码算法的深度学习综述[J]. 计算机系统应用,2018,27(9):47-51.
[14] 刘振兴,陈震. 一种基于堆栈自编码器的水声信道均衡算法[J]. 通信技术,2019,52(11):2605-2610.
[15] 徐锋,方彦军. 基于贝叶斯优化XGBoost 的现场校验仪误差预测[J] . 电测与仪表,2019,56(18):120-125.
[16] 葛阳,郭兰中,牛曙光,等. 基于t-SNE和LSTM的旋转机械剩余寿命预测[J] . 振动与冲击,2020,39(7):223-231.