Cargando…

Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes

In the manufacturing process, equipment failure is directly related to productivity, so predictive maintenance plays a very important role. Industrial parks are distributed, and data heterogeneity exists among heterogeneous equipment, which makes predictive maintenance of equipment challenging. In t...

Descripción completa

Detalles Bibliográficos
Autores principales: Ahn, Jisu, Lee, Younjeong, Kim, Namji, Park, Chanho, Jeong, Jongpil
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10490086/
https://www.ncbi.nlm.nih.gov/pubmed/37687787
http://dx.doi.org/10.3390/s23177331
_version_ 1785103760240934912
author Ahn, Jisu
Lee, Younjeong
Kim, Namji
Park, Chanho
Jeong, Jongpil
author_facet Ahn, Jisu
Lee, Younjeong
Kim, Namji
Park, Chanho
Jeong, Jongpil
author_sort Ahn, Jisu
collection PubMed
description In the manufacturing process, equipment failure is directly related to productivity, so predictive maintenance plays a very important role. Industrial parks are distributed, and data heterogeneity exists among heterogeneous equipment, which makes predictive maintenance of equipment challenging. In this paper, we propose two main techniques to enable effective predictive maintenance in this environment. We propose a 1DCNN-Bilstm model for time series anomaly detection and predictive maintenance of manufacturing processes. The model combines a 1D convolutional neural network (1DCNN) and a bidirectional LSTM (Bilstm), which is effective in extracting features from time series data and detecting anomalies. In this paper, we combine a federated learning framework with these models to consider the distributional shifts of time series data and perform anomaly detection and predictive maintenance based on them. In this paper, we utilize the pump dataset to evaluate the performance of the combination of several federated learning frameworks and time series anomaly detection models. Experimental results show that the proposed framework achieves a test accuracy of 97.2%, which shows its potential to be utilized for real-world predictive maintenance in the future.
format Online
Article
Text
id pubmed-10490086
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-104900862023-09-09 Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes Ahn, Jisu Lee, Younjeong Kim, Namji Park, Chanho Jeong, Jongpil Sensors (Basel) Article In the manufacturing process, equipment failure is directly related to productivity, so predictive maintenance plays a very important role. Industrial parks are distributed, and data heterogeneity exists among heterogeneous equipment, which makes predictive maintenance of equipment challenging. In this paper, we propose two main techniques to enable effective predictive maintenance in this environment. We propose a 1DCNN-Bilstm model for time series anomaly detection and predictive maintenance of manufacturing processes. The model combines a 1D convolutional neural network (1DCNN) and a bidirectional LSTM (Bilstm), which is effective in extracting features from time series data and detecting anomalies. In this paper, we combine a federated learning framework with these models to consider the distributional shifts of time series data and perform anomaly detection and predictive maintenance based on them. In this paper, we utilize the pump dataset to evaluate the performance of the combination of several federated learning frameworks and time series anomaly detection models. Experimental results show that the proposed framework achieves a test accuracy of 97.2%, which shows its potential to be utilized for real-world predictive maintenance in the future. MDPI 2023-08-22 /pmc/articles/PMC10490086/ /pubmed/37687787 http://dx.doi.org/10.3390/s23177331 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Ahn, Jisu
Lee, Younjeong
Kim, Namji
Park, Chanho
Jeong, Jongpil
Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes
title Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes
title_full Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes
title_fullStr Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes
title_full_unstemmed Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes
title_short Federated Learning for Predictive Maintenance and Anomaly Detection Using Time Series Data Distribution Shifts in Manufacturing Processes
title_sort federated learning for predictive maintenance and anomaly detection using time series data distribution shifts in manufacturing processes
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10490086/
https://www.ncbi.nlm.nih.gov/pubmed/37687787
http://dx.doi.org/10.3390/s23177331
work_keys_str_mv AT ahnjisu federatedlearningforpredictivemaintenanceandanomalydetectionusingtimeseriesdatadistributionshiftsinmanufacturingprocesses
AT leeyounjeong federatedlearningforpredictivemaintenanceandanomalydetectionusingtimeseriesdatadistributionshiftsinmanufacturingprocesses
AT kimnamji federatedlearningforpredictivemaintenanceandanomalydetectionusingtimeseriesdatadistributionshiftsinmanufacturingprocesses
AT parkchanho federatedlearningforpredictivemaintenanceandanomalydetectionusingtimeseriesdatadistributionshiftsinmanufacturingprocesses
AT jeongjongpil federatedlearningforpredictivemaintenanceandanomalydetectionusingtimeseriesdatadistributionshiftsinmanufacturingprocesses