<  Back to the Polytechnique Montréal portal

Using learned health indicators and deep sequence models to predict industrial machine health

Ido Amihai, Arzam Kotriwala, Diego Pareschi, Moncef Chioua and Ralf Gitzel

Paper (2021)

Open Acess document in PolyPublie and at official publisher
[img]
Preview
Open Access to the full text of this document
Published Version
Terms of Use: Creative Commons Attribution
Download (605kB)
Show abstract
Hide abstract

Abstract

In this paper, we describe a machine learning approach for predicting machine health indicators with a large time horizon into the future. The approach uses state-of-the-art neural network architectures for sequence modelling and can incorporate numerical-sensor and categorical data using entity embeddings. Moreover, we describe an unsupervised labelling approach where classes are generated using continuous sensor values in the training data and a clustering algorithm. To validate our approach, we performed an ablation study to verify the effectiveness of each of our model’s components. In this context, we show that entity embeddings can be used to generate effective features from categorical inputs, that state-of-the-art models, while originally developed for a different set of problems, can nonetheless be transferred to perform industrial asset health classification and provide a performance boost over simpler networks that have been traditionally used, such as relatively shallow recurrent or convolutional networks. Taken together, we present a machine health monitoring system that can accurately generate asset health predictions. This system can incorporate both numerical and categorical information, the current state-of-the-art for sequence modelling, and generate labels in an unsupervised fashion when explicit labels are unavailable.

Uncontrolled Keywords

neural networks; time series; sequence modelling; machine health monitoring; predictive maintenance

Subjects: 1800 Chemical engineering > 1800 Chemical engineering
2500 Electrical and electronic engineering > 2500 Electrical and electronic engineering
2500 Electrical and electronic engineering > 2509 Control systems
2700 Information technology > 2713 Algorithms
Department: Department of Chemical Engineering
Funders: ABB
PolyPublie URL: https://publications.polymtl.ca/9386/
Conference Title: 7th International Conference on Time Series and Forecasting (ITISE 2021)
Conference Location: Gran Canaria, Spain
Conference Date(s): 2021-07-19 - 2021-07-21
Journal Title: Engineering Proceedings (vol. 5, no. 1)
Publisher: MDPI
DOI: 10.3390/engproc2021005007
Official URL: https://doi.org/10.3390/engproc2021005007
Date Deposited: 16 Aug 2023 12:23
Last Modified: 28 Sep 2024 08:21
Cite in APA 7: Amihai, I., Kotriwala, A., Pareschi, D., Chioua, M., & Gitzel, R. (2021, July). Using learned health indicators and deep sequence models to predict industrial machine health [Paper]. 7th International Conference on Time Series and Forecasting (ITISE 2021), Gran Canaria, Spain (9 pages). Published in Engineering Proceedings, 5(1). https://doi.org/10.3390/engproc2021005007

Statistics

Total downloads

Downloads per month in the last year

Origin of downloads

Dimensions

Repository Staff Only

View Item View Item