Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion

Mechanical seals are critical components in the mechanical industry, and their operational status directly impacts the performance of pumps, compressors, and other machinery. Therefore, conducting research on the fault diagnosis of mechanical seals is essential. To enhance the accuracy of the assess...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaoran Zhu, Jiahao Wang, Binhui Wang, Hao Wang, Ren Sheng, Baozun Zhai
Format: Article
Language:English
Published: SAGE Publishing 2025-02-01
Series:Advances in Mechanical Engineering
Online Access:https://doi.org/10.1177/16878132251319141
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850025802255564800
author Xiaoran Zhu
Jiahao Wang
Binhui Wang
Hao Wang
Ren Sheng
Baozun Zhai
author_facet Xiaoran Zhu
Jiahao Wang
Binhui Wang
Hao Wang
Ren Sheng
Baozun Zhai
author_sort Xiaoran Zhu
collection DOAJ
description Mechanical seals are critical components in the mechanical industry, and their operational status directly impacts the performance of pumps, compressors, and other machinery. Therefore, conducting research on the fault diagnosis of mechanical seals is essential. To enhance the accuracy of the assessment model, we propose an integrated approach that leverages the fusion of multiple graph neural networks (GNNs). Firstly, recognizing the diversity among different sensors, we utilize multi-channel data to comprehensively represent the operational state of the mechanical seal. These channels include various types of sensors such as acoustic emission and force sensors. Secondly, we employ multiple methods to transform the original multi-channel data into graph data, thereby continuously increasing the diversity of the datasets used for training. Finally, after training GNNs, we output the data of these networks through data fusion to obtain evaluation results. The effectiveness of our assessment approach is demonstrated using mechanical seal test data.
format Article
id doaj-art-243198a702da4e90bf9005db0bad8359
institution DOAJ
issn 1687-8140
language English
publishDate 2025-02-01
publisher SAGE Publishing
record_format Article
series Advances in Mechanical Engineering
spelling doaj-art-243198a702da4e90bf9005db0bad83592025-08-20T03:00:44ZengSAGE PublishingAdvances in Mechanical Engineering1687-81402025-02-011710.1177/16878132251319141Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusionXiaoran Zhu0Jiahao Wang1Binhui Wang2Hao Wang3Ren Sheng4Baozun Zhai5School of Mechanical Engineering, Yellow River Conservancy Technical Institute, Kaifeng, Henan, ChinaSchool of Mechanical Engineering, North China University of Water Resources and Electric Power, Zhengzhou, Henan, ChinaSchool of Mechanical Engineering, North China University of Water Resources and Electric Power, Zhengzhou, Henan, ChinaSchool of Mechanical Engineering, North China University of Water Resources and Electric Power, Zhengzhou, Henan, ChinaSchool of Mechanical Engineering, Yellow River Conservancy Technical Institute, Kaifeng, Henan, ChinaSchool of Mechanical Engineering, Yellow River Conservancy Technical Institute, Kaifeng, Henan, ChinaMechanical seals are critical components in the mechanical industry, and their operational status directly impacts the performance of pumps, compressors, and other machinery. Therefore, conducting research on the fault diagnosis of mechanical seals is essential. To enhance the accuracy of the assessment model, we propose an integrated approach that leverages the fusion of multiple graph neural networks (GNNs). Firstly, recognizing the diversity among different sensors, we utilize multi-channel data to comprehensively represent the operational state of the mechanical seal. These channels include various types of sensors such as acoustic emission and force sensors. Secondly, we employ multiple methods to transform the original multi-channel data into graph data, thereby continuously increasing the diversity of the datasets used for training. Finally, after training GNNs, we output the data of these networks through data fusion to obtain evaluation results. The effectiveness of our assessment approach is demonstrated using mechanical seal test data.https://doi.org/10.1177/16878132251319141
spellingShingle Xiaoran Zhu
Jiahao Wang
Binhui Wang
Hao Wang
Ren Sheng
Baozun Zhai
Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion
Advances in Mechanical Engineering
title Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion
title_full Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion
title_fullStr Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion
title_full_unstemmed Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion
title_short Fault diagnosis of mechanical seals using graph neural networks with multi-sensor data fusion
title_sort fault diagnosis of mechanical seals using graph neural networks with multi sensor data fusion
url https://doi.org/10.1177/16878132251319141
work_keys_str_mv AT xiaoranzhu faultdiagnosisofmechanicalsealsusinggraphneuralnetworkswithmultisensordatafusion
AT jiahaowang faultdiagnosisofmechanicalsealsusinggraphneuralnetworkswithmultisensordatafusion
AT binhuiwang faultdiagnosisofmechanicalsealsusinggraphneuralnetworkswithmultisensordatafusion
AT haowang faultdiagnosisofmechanicalsealsusinggraphneuralnetworkswithmultisensordatafusion
AT rensheng faultdiagnosisofmechanicalsealsusinggraphneuralnetworkswithmultisensordatafusion
AT baozunzhai faultdiagnosisofmechanicalsealsusinggraphneuralnetworkswithmultisensordatafusion