Wind turbine signal fault diagnosis using deep neural networks-inspired model

This work presents a deep neural network-inspired solution to intelligent signal fault diagnosis for wind turbine gearbox systems. A 1D convolution deep neural network architecture is proposed, constructed and validated. The proposed model was constructed of 1D signal for the input layer, ten differ...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Rababaah, Aaron (author)
منشور في: 2023
الوصول للمادة أونلاين:http://hdl.handle.net/11675/10926
https://www.inderscience.com/info/ingeneral/forthcoming.php?jcode=ijcat
http://www.scopus.com/inward/record.url?scp=85153848068&partnerID=8YFLogxK
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1870679723882840065
author Rababaah, Aaron
author_facet Rababaah, Aaron
author_role author
dc.creator.none.fl_str_mv Rababaah, Aaron
dc.date.none.fl_str_mv 2023-01-01
2024-02-05T08:32:40Z
2024-02-05T08:32:40Z
dc.identifier.none.fl_str_mv 10.1504/IJCAT.2022.129378
http://hdl.handle.net/11675/10926
https://www.inderscience.com/info/ingeneral/forthcoming.php?jcode=ijcat
http://www.scopus.com/inward/record.url?scp=85153848068&partnerID=8YFLogxK
dc.publisher.none.fl_str_mv Inderscience Enterprises Ltd
dc.relation.none.fl_str_mv Computer Science and Info Systems
International Journal of Computer Applications in Technology
dc.title.none.fl_str_mv Wind turbine signal fault diagnosis using deep neural networks-inspired model
dc.type.none.fl_str_mv Exhibition
info:eu-repo/semantics/publishedVersion
description This work presents a deep neural network-inspired solution to intelligent signal fault diagnosis for wind turbine gearbox systems. A 1D convolution deep neural network architecture is proposed, constructed and validated. The proposed model was constructed of 1D signal for the input layer, ten different learned kernels as signal features, convolution layer, activation layer using rectified linear unit function, max-pooling layer, flatten layer and traditional multi-perceptron neural network for classification with soft-max class assignment. The data was acquired from real-world experiments conducted on real wind turbine gearboxes and archived by the National Renewable Energy Labs of the US Department of Energy. Ten independent experiments were conducted on 2,400,000 data points and the proposed model produced a mean classification accuracy of 96.14% for normal signals with a standard deviation of 0.0027 and a mean classification accuracy of 99.87% for faulty signals with a standard deviation of 0.0016.
id AUKR_a612017f6111d737f144fbfbfca8d723
identifier_str_mv 10.1504/IJCAT.2022.129378
network_acronym_str AUKR
network_name_str AU Kuwait Rep
oai_identifier_str oai:dspace.auk.edu.kw:11675/10926
publishDate 2023
publisher.none.fl_str_mv Inderscience Enterprises Ltd
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Wind turbine signal fault diagnosis using deep neural networks-inspired modelRababaah, AaronThis work presents a deep neural network-inspired solution to intelligent signal fault diagnosis for wind turbine gearbox systems. A 1D convolution deep neural network architecture is proposed, constructed and validated. The proposed model was constructed of 1D signal for the input layer, ten different learned kernels as signal features, convolution layer, activation layer using rectified linear unit function, max-pooling layer, flatten layer and traditional multi-perceptron neural network for classification with soft-max class assignment. The data was acquired from real-world experiments conducted on real wind turbine gearboxes and archived by the National Renewable Energy Labs of the US Department of Energy. Ten independent experiments were conducted on 2,400,000 data points and the proposed model produced a mean classification accuracy of 96.14% for normal signals with a standard deviation of 0.0027 and a mean classification accuracy of 99.87% for faulty signals with a standard deviation of 0.0016.Inderscience Enterprises Ltd2024-02-05T08:32:40Z2024-02-05T08:32:40Z2023-01-01Exhibitioninfo:eu-repo/semantics/publishedVersion10.1504/IJCAT.2022.129378http://hdl.handle.net/11675/10926https://www.inderscience.com/info/ingeneral/forthcoming.php?jcode=ijcathttp://www.scopus.com/inward/record.url?scp=85153848068&partnerID=8YFLogxKComputer Science and Info SystemsInternational Journal of Computer Applications in Technologyoai:dspace.auk.edu.kw:11675/109262025-06-18T09:03:51Z
spellingShingle Wind turbine signal fault diagnosis using deep neural networks-inspired model
Rababaah, Aaron
status_str publishedVersion
title Wind turbine signal fault diagnosis using deep neural networks-inspired model
title_full Wind turbine signal fault diagnosis using deep neural networks-inspired model
title_fullStr Wind turbine signal fault diagnosis using deep neural networks-inspired model
title_full_unstemmed Wind turbine signal fault diagnosis using deep neural networks-inspired model
title_short Wind turbine signal fault diagnosis using deep neural networks-inspired model
title_sort Wind turbine signal fault diagnosis using deep neural networks-inspired model
url http://hdl.handle.net/11675/10926
https://www.inderscience.com/info/ingeneral/forthcoming.php?jcode=ijcat
http://www.scopus.com/inward/record.url?scp=85153848068&partnerID=8YFLogxK