基于深度学习的风机故障智能诊断

doi:10.1004-7530/2019-36-2-45

江苏科技信息 ›› 2019, Vol. 36 ›› Issue (2): 45-48.doi: 10.1004-7530/2019-36-2-45

基于深度学习的风机故障智能诊断

吕志远,马笑潇

观为监测技术无锡股份有限公司,江苏无锡 214000

出版日期:2019-01-20 发布日期:2019-07-09
通讯作者: 马笑潇
作者简介:吕志远（1984— ）,男,吉林通榆人,工程师,本科;研究方向：机械故障诊断,信号处理。

Intelligent fault diagnosis of wind power generator based on deep learning

Zhiyuan Lyu,Xiaoxiao Ma

Guanwei Monitoring Technology Wuxi Co., Ltd., Wuxi 214000, China

Online:2019-01-20 Published:2019-07-09
Contact: Xiaoxiao Ma

摘要/Abstract

摘要：

风力发电场往往建设在地方人稀的地区,分布范围很广,而且由于安装高度很高,导致很难直接实施观测与故障诊断。通过安装在线传感器,可以对数据进行实时监测,然而风机结构复杂,信号包含了大量噪声,对其进行高效的故障监测是一个难点。针对此,文章将深度学习引入故障识别中,通过使用堆栈式自动编码器建立智能识别网络,将信号频谱直接输入网络,不需要人工提取特征,实现风机的智能诊断。这种方式有效减少了人工工作量,大大提高了诊断的效率。

关键词: 深度学习, 故障诊断, 风机

Abstract:

Wind farms are often built in the sparsely populated areas with a wide distribution, and because of the high installation height, it is difficult to directly implement observation and fault diagnosis. Luckily, its condition can be monitored by installing on-line sensors. However, the structure of wind power generator is complex, and the signal contains a lot of noise, so it is difficult to monitor it effectively. In view of this, this paper introduces deep learning into fault recognition. By using stacked autoencoders to establish an intelligent recognition network, the signal spectrum is directly input into the network, and the intelligent diagnosis of wind power generator is realized without manual feature extraction. This method effectively reduces the manual workload and greatly improves the efficiency of diagnosis.

Key words: deep learning, fault diagnosis, wind power generator

中图分类号:

TN165.3

吕志远,马笑潇. 基于深度学习的风机故障智能诊断[J]. 江苏科技信息, 2019, 36(2): 45-48.

Zhiyuan Lyu,Xiaoxiao Ma. Intelligent fault diagnosis of wind power generator based on deep learning[J]. Jiangsu Science & Technology Information, 2019, 36(2): 45-48.

图/表 8

表1

表2

图1

图2

图3

图4

图5

图6

参考文献 10

[1]	WANG X Q, ZHU H J, WANG D , et al. The diagnosis of rolling bearing based on the parameters of pulse atoms and degree of cyclostationarity[J].Journal of Vibroen-gineering, 2013(3):1560-1575.
[2]	ZHU H J, WANG X Q, ZHAO Y , et al. Sparse representation based on adaptive multiscale features for robust machinery fault diagnosis[J].Joural of Mecha-nical Engineering Science, 2015(12):2303-2313.
[3]	HINTON G E, OSINDERO S, THE Y W . A fast learning algorithm for deep belief nets[J].Neural Compu-tation, 2006(7):1527-1554.
[4]	余凯, 贾磊, 陈雨强 , 等. 深度学习的昨天,今天和明天[J].计算机研究与发展, 2013(9):1799-1804.
[5]	ZHU H J, RUI T, WANG X Q , et al. Fault diagnosis of hydraulic pump based on stacked autoencoders[C]. Qingdao: 2015 12th IEEE International Conference on Electronic Measurement & Instruments(ICEMI), 2015.
[6]	孙志军, 薛磊, 许阳明 , 等. 深度学习研究综述[J].计算机应用研究, 2012(8):2806-2810.
[7]	徐姗姗, 刘应安, 徐昇 . 基于卷积神经网络的木材缺陷识别[J].山东大学学报(工学版), 2013(2):23-28.
[8]	TAMILSELVAN P, WANG P F . Failure diagnosis using deep belief learning based health state classifi-cation[J].Reliability Engineering & System Safety, 2013(115):127-135.
[9]	LEE H, GROSSE R, RANGANATH R, et al. Convo-lutional deep belief networks for scalable unsupervised learning of hierarchical representations [C]. Association for Computing Machinery, 2009: 609-616.
[10]	林妙真 . 基于深度学习的人脸识别研究[D]. 大连:大连理工大学, 2013.

《江苏科技信息》杂志社
Jiangsu Science & Technology Information Magazine Agency

基于深度学习的风机故障智能诊断

Intelligent fault diagnosis of wind power generator based on deep learning

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 10

相关文章 5

Metrics

本文评价

推荐阅读 10

[1]	史裕问,韦怀坤,朱嘉晨. 离心式空气净化器的降噪分析[J]. 江苏科技信息, 2019, 36(10): 48-50.
[2]	孙琳. 新形势下人工智能产业发展现状研究 ——以江苏为例[J]. 江苏科技信息, 2018, 35(9): 22-24.
[3]	庄伟栋, 邱晓侨. 水利工程泵站机电设备故障诊断方法分析[J]. 江苏科技信息, 2018, 35(31): 66-69.
[4]	陆钟华. 110 kV主变局放检测现场试验技术研究[J]. 江苏科技信息, 2017, 34(5): 23-26.
[5]	张峰, 屠秉慧, 郑思源. 基于智辅平台的变电站视频设备故障诊断子系统的研究[J]. 江苏科技信息, 2015, 32(25): 65-66.