Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System
To investigate problems involving wind turbines that easily occur but are hard to diagnose, this paper presents a wind turbine (WT) fault diagnosis algorithm based on a spectrogram and a convolutional neural network. First, the original data are sampled into a phonetic form. Then, the data are trans...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-09-01
|
Series: | Energies |
Subjects: | |
Online Access: | http://www.mdpi.com/1996-1073/11/10/2561 |
id |
doaj-e014ff4a6b0a437999f975480707ac42 |
---|---|
record_format |
Article |
spelling |
doaj-e014ff4a6b0a437999f975480707ac422020-11-24T20:45:31ZengMDPI AGEnergies1996-10732018-09-011110256110.3390/en11102561en11102561Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine SystemWenxin Yu0Shoudao Huang1Weihong Xiao2College of Electrical and Information Engineering, Hunan University, Changsha 410082, ChinaCollege of Electrical and Information Engineering, Hunan University, Changsha 410082, ChinaSchool of Information Engineering, Hunan Province Cooperative Innovation Center for Wind Power Equipment and Energy Conversion, Xiangtan University, Xiangtan 411105, ChinaTo investigate problems involving wind turbines that easily occur but are hard to diagnose, this paper presents a wind turbine (WT) fault diagnosis algorithm based on a spectrogram and a convolutional neural network. First, the original data are sampled into a phonetic form. Then, the data are transformed into a spectrogram in the time-frequency domain. Finally, the data are sent into a convolutional neural network (CNN) model with batch regularization for training and testing. Experimental results show that the method is suitable for training a large number of samples and has good scalability. Compared with Back Propagation Neural Network (BPNN), Support Vector Machine (SVM), Extreme Learning Machine (ELM), and other fault diagnosis methods, the average diagnostic correctness rate is higher; so, the method can provide more accurate reference information for wind turbine fault diagnosis.http://www.mdpi.com/1996-1073/11/10/2561spectrogramconvolutional neural networkwind turbinefault diagnosis |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Wenxin Yu Shoudao Huang Weihong Xiao |
spellingShingle |
Wenxin Yu Shoudao Huang Weihong Xiao Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System Energies spectrogram convolutional neural network wind turbine fault diagnosis |
author_facet |
Wenxin Yu Shoudao Huang Weihong Xiao |
author_sort |
Wenxin Yu |
title |
Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System |
title_short |
Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System |
title_full |
Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System |
title_fullStr |
Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System |
title_full_unstemmed |
Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System |
title_sort |
fault diagnosis based on an approach combining a spectrogram and a convolutional neural network with application to a wind turbine system |
publisher |
MDPI AG |
series |
Energies |
issn |
1996-1073 |
publishDate |
2018-09-01 |
description |
To investigate problems involving wind turbines that easily occur but are hard to diagnose, this paper presents a wind turbine (WT) fault diagnosis algorithm based on a spectrogram and a convolutional neural network. First, the original data are sampled into a phonetic form. Then, the data are transformed into a spectrogram in the time-frequency domain. Finally, the data are sent into a convolutional neural network (CNN) model with batch regularization for training and testing. Experimental results show that the method is suitable for training a large number of samples and has good scalability. Compared with Back Propagation Neural Network (BPNN), Support Vector Machine (SVM), Extreme Learning Machine (ELM), and other fault diagnosis methods, the average diagnostic correctness rate is higher; so, the method can provide more accurate reference information for wind turbine fault diagnosis. |
topic |
spectrogram convolutional neural network wind turbine fault diagnosis |
url |
http://www.mdpi.com/1996-1073/11/10/2561 |
work_keys_str_mv |
AT wenxinyu faultdiagnosisbasedonanapproachcombiningaspectrogramandaconvolutionalneuralnetworkwithapplicationtoawindturbinesystem AT shoudaohuang faultdiagnosisbasedonanapproachcombiningaspectrogramandaconvolutionalneuralnetworkwithapplicationtoawindturbinesystem AT weihongxiao faultdiagnosisbasedonanapproachcombiningaspectrogramandaconvolutionalneuralnetworkwithapplicationtoawindturbinesystem |
_version_ |
1716814502200082432 |