Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator

<p dir="ltr">The <u>gas turbine engine</u> is a predominant <u>prime mover</u> in the transport and energy sectors, and ensuring its reliable operation holds paramount significance. While intelligent fault diagnosis (FD) approaches have seen successful advance...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Fatemeh Negar Irani (22302835) (author)
مؤلفون آخرون: Mohammadjavad Soleimani (22302838) (author), Meysam Yadegar (16410089) (author), Nader Meskin (14147796) (author)
منشور في: 2024
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513513462956032
author Fatemeh Negar Irani (22302835)
author2 Mohammadjavad Soleimani (22302838)
Meysam Yadegar (16410089)
Nader Meskin (14147796)
author2_role author
author
author
author_facet Fatemeh Negar Irani (22302835)
Mohammadjavad Soleimani (22302838)
Meysam Yadegar (16410089)
Nader Meskin (14147796)
author_role author
dc.creator.none.fl_str_mv Fatemeh Negar Irani (22302835)
Mohammadjavad Soleimani (22302838)
Meysam Yadegar (16410089)
Nader Meskin (14147796)
dc.date.none.fl_str_mv 2024-04-20T12:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.apenergy.2024.123256
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Deep_transfer_learning_strategy_in_intelligent_fault_diagnosis_of_gas_turbines_based_on_the_Koopman_operator/30197188
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Engineering
Aerospace engineering
Mechanical engineering
Information and computing sciences
Artificial intelligence
Gas turbine
Fault diagnosis
Data-driven
Transfer learning
Koopman operator
Geometric
dc.title.none.fl_str_mv Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">The <u>gas turbine engine</u> is a predominant <u>prime mover</u> in the transport and energy sectors, and ensuring its reliable operation holds paramount significance. While intelligent fault diagnosis (FD) approaches have seen successful advancements within the <u>gas turbine</u> FD landscape, many existing methods operate under the assumption of identical health states during both data collection and the FD process. Moreover, most previous studies have overlooked the diagnosis of both sensors and <u>actuators</u>. Another critical challenge lies in isolating simultaneous and multiple faults and providing compromising FD performance, especially in the face of continued system <u>performance</u> <u>degradation</u>. Aiming at these problems, this study develops a novel unsupervised data-driven FD strategy based on leveraging the potential of <u>transfer learning</u> and the Koopman operator. A <u>deep neural network</u>-based transfer learning framework is proposed for realizing a precise adaptive linear model called the deep transfer linear (DTL) model enabling reliable prediction of the system’s behavior in various situations and designing structured fault residuals. To this end, a deep neural network framework is used to obtain a precise Koopman model realization using the richly collected data in the source domain. Subsequently, the realized model is fine-tuned for the target domains associated with the degraded system, mitigating the adverse effects of domain shift and addressing the rich data scarcity problem in the target domain. In addition, the dedicated and generalized <u>residual sets</u> are designed and generated employing the geometric approach for fault isolation and a decision-making analysis is developed to diagnose simultaneous faults. The reliability of the proposed strategy is demonstrated through various experiments in the presence of noise and performance degradation, and a comparative performance analysis is conducted between the proposed strategy and another data-driven method showcasing the superiority of the proposed approach.</p><h2>Other Information</h2><p dir="ltr">Published in: Applied Energy<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.apenergy.2024.123256" target="_blank">https://dx.doi.org/10.1016/j.apenergy.2024.123256</a></p>
eu_rights_str_mv openAccess
id Manara2_e07beb79fd641c4a882625ac52ecb061
identifier_str_mv 10.1016/j.apenergy.2024.123256
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30197188
publishDate 2024
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operatorFatemeh Negar Irani (22302835)Mohammadjavad Soleimani (22302838)Meysam Yadegar (16410089)Nader Meskin (14147796)EngineeringAerospace engineeringMechanical engineeringInformation and computing sciencesArtificial intelligenceGas turbineFault diagnosisData-drivenTransfer learningKoopman operatorGeometric<p dir="ltr">The <u>gas turbine engine</u> is a predominant <u>prime mover</u> in the transport and energy sectors, and ensuring its reliable operation holds paramount significance. While intelligent fault diagnosis (FD) approaches have seen successful advancements within the <u>gas turbine</u> FD landscape, many existing methods operate under the assumption of identical health states during both data collection and the FD process. Moreover, most previous studies have overlooked the diagnosis of both sensors and <u>actuators</u>. Another critical challenge lies in isolating simultaneous and multiple faults and providing compromising FD performance, especially in the face of continued system <u>performance</u> <u>degradation</u>. Aiming at these problems, this study develops a novel unsupervised data-driven FD strategy based on leveraging the potential of <u>transfer learning</u> and the Koopman operator. A <u>deep neural network</u>-based transfer learning framework is proposed for realizing a precise adaptive linear model called the deep transfer linear (DTL) model enabling reliable prediction of the system’s behavior in various situations and designing structured fault residuals. To this end, a deep neural network framework is used to obtain a precise Koopman model realization using the richly collected data in the source domain. Subsequently, the realized model is fine-tuned for the target domains associated with the degraded system, mitigating the adverse effects of domain shift and addressing the rich data scarcity problem in the target domain. In addition, the dedicated and generalized <u>residual sets</u> are designed and generated employing the geometric approach for fault isolation and a decision-making analysis is developed to diagnose simultaneous faults. The reliability of the proposed strategy is demonstrated through various experiments in the presence of noise and performance degradation, and a comparative performance analysis is conducted between the proposed strategy and another data-driven method showcasing the superiority of the proposed approach.</p><h2>Other Information</h2><p dir="ltr">Published in: Applied Energy<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.apenergy.2024.123256" target="_blank">https://dx.doi.org/10.1016/j.apenergy.2024.123256</a></p>2024-04-20T12:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.apenergy.2024.123256https://figshare.com/articles/journal_contribution/Deep_transfer_learning_strategy_in_intelligent_fault_diagnosis_of_gas_turbines_based_on_the_Koopman_operator/30197188CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/301971882024-04-20T12:00:00Z
spellingShingle Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
Fatemeh Negar Irani (22302835)
Engineering
Aerospace engineering
Mechanical engineering
Information and computing sciences
Artificial intelligence
Gas turbine
Fault diagnosis
Data-driven
Transfer learning
Koopman operator
Geometric
status_str publishedVersion
title Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
title_full Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
title_fullStr Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
title_full_unstemmed Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
title_short Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
title_sort Deep transfer learning strategy in intelligent fault diagnosis of gas turbines based on the Koopman operator
topic Engineering
Aerospace engineering
Mechanical engineering
Information and computing sciences
Artificial intelligence
Gas turbine
Fault diagnosis
Data-driven
Transfer learning
Koopman operator
Geometric