Hybrid CNN-LSTM Model for Non-Invasive Fault Detection in Induction Motors Using Acoustic Data

Aruzhan Seralieva; Muhammad Ilyas; Anelya Gissa; Md Hazrat Ali

doi:10.1109/ICSPC63060.2024.10862542

Hybrid CNN-LSTM Model for Non-Invasive Fault Detection in Induction Motors Using Acoustic Data

Aruzhan Seralieva
, Muhammad Ilyas
, Anelya Gissa
, Md Hazrat Ali

Kazakh-British Technical University
Nazarbayev University

Research output: Contribution to journal › Conference article › peer-review

3 Citations (Scopus)

Abstract

Monitoring the health condition of industrial drives, especially, induction motors is crucial to maintaining the reliability and efficiency of industrial processes. Fault detection techniques tend to be invasive and expensive despite their effectiveness. This paper introduces a non-invasive methodology that leverages acoustic signals processed through sophisticated deep learning models, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs). We aim to accurately detect and classify different fault conditions in induction motors by extracting frequency features from acoustic data. Our results show marked improvements over conventional methods, offering a valuable predictive non-invasive maintenance tool.

Original language	English
Pages (from-to)	136-141
Number of pages	6
Journal	Proceedings of the IEEE Conference on Systems, Process and Control, ICSPC
Issue number	2024
DOIs	https://doi.org/10.1109/ICSPC63060.2024.10862542
Publication status	Published - 2024
Event	12th IEEE Conference on Systems, Process and Control, ICSPC 2024 - Malacca, Malaysia Duration: Dec 7 2024 → …

Keywords

convolutional neural network
deep learning
fault detection
industrial drives
Long Short-Term Memory networks

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Information Systems
Information Systems and Management
Safety, Risk, Reliability and Quality
Control and Optimization
Modelling and Simulation
Education

Access to Document

10.1109/ICSPC63060.2024.10862542

Cite this

@article{cc6c8df8b3374c4c8ca0f71a08041937,

title = "Hybrid CNN-LSTM Model for Non-Invasive Fault Detection in Induction Motors Using Acoustic Data",

abstract = "Monitoring the health condition of industrial drives, especially, induction motors is crucial to maintaining the reliability and efficiency of industrial processes. Fault detection techniques tend to be invasive and expensive despite their effectiveness. This paper introduces a non-invasive methodology that leverages acoustic signals processed through sophisticated deep learning models, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs). We aim to accurately detect and classify different fault conditions in induction motors by extracting frequency features from acoustic data. Our results show marked improvements over conventional methods, offering a valuable predictive non-invasive maintenance tool.",

keywords = "convolutional neural network, deep learning, fault detection, industrial drives, Long Short-Term Memory networks",

author = "Aruzhan Seralieva and Muhammad Ilyas and Anelya Gissa and Ali, \{Md Hazrat\}",

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 12th IEEE Conference on Systems, Process and Control, ICSPC 2024 ; Conference date: 07-12-2024",

year = "2024",

doi = "10.1109/ICSPC63060.2024.10862542",

language = "English",

pages = "136--141",

journal = "Proceedings of the IEEE Conference on Systems, Process and Control, ICSPC",

issn = "2769-8378",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2024",

}

TY - JOUR

T1 - Hybrid CNN-LSTM Model for Non-Invasive Fault Detection in Induction Motors Using Acoustic Data

AU - Seralieva, Aruzhan

AU - Ilyas, Muhammad

AU - Gissa, Anelya

AU - Ali, Md Hazrat

PY - 2024

Y1 - 2024

N2 - Monitoring the health condition of industrial drives, especially, induction motors is crucial to maintaining the reliability and efficiency of industrial processes. Fault detection techniques tend to be invasive and expensive despite their effectiveness. This paper introduces a non-invasive methodology that leverages acoustic signals processed through sophisticated deep learning models, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs). We aim to accurately detect and classify different fault conditions in induction motors by extracting frequency features from acoustic data. Our results show marked improvements over conventional methods, offering a valuable predictive non-invasive maintenance tool.

AB - Monitoring the health condition of industrial drives, especially, induction motors is crucial to maintaining the reliability and efficiency of industrial processes. Fault detection techniques tend to be invasive and expensive despite their effectiveness. This paper introduces a non-invasive methodology that leverages acoustic signals processed through sophisticated deep learning models, such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs). We aim to accurately detect and classify different fault conditions in induction motors by extracting frequency features from acoustic data. Our results show marked improvements over conventional methods, offering a valuable predictive non-invasive maintenance tool.

KW - convolutional neural network

KW - deep learning

KW - fault detection

KW - industrial drives

KW - Long Short-Term Memory networks

UR - https://www.scopus.com/pages/publications/105001487393

UR - https://www.scopus.com/pages/publications/105001487393#tab=citedBy

U2 - 10.1109/ICSPC63060.2024.10862542

DO - 10.1109/ICSPC63060.2024.10862542

M3 - Conference article

AN - SCOPUS:105001487393

SN - 2769-8378

SP - 136

EP - 141

JO - Proceedings of the IEEE Conference on Systems, Process and Control, ICSPC

JF - Proceedings of the IEEE Conference on Systems, Process and Control, ICSPC

IS - 2024

T2 - 12th IEEE Conference on Systems, Process and Control, ICSPC 2024

Y2 - 7 December 2024

ER -

Hybrid CNN-LSTM Model for Non-Invasive Fault Detection in Induction Motors Using Acoustic Data

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this