Journal
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
Volume 18, Issue 9, Pages 5859-5869Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TII.2021.3131471
Keywords
Feature extraction; Soft sensors; Logic gates; Data mining; Time series analysis; Frequency modulation; Data models; Deep learning; industrial big data; industrial intelligence; product quality prediction; self-supervised learning; soft sensor
Categories
Funding
- National Key Research and Development Program of China [2019YFB1703903]
- National Science Foundation of China [92167108, 62173023, 61836001]
Ask authors/readers for more resources
In this article, a data-driven self-supervised deep learning model is proposed for industrial soft sensor, which explores diverse industrial data characteristics through pretraining and finetuning stages. Experimental results demonstrate that the proposed method achieves state of the art performance on real-world datasets.
Soft sensor, as an important paradigm for industrial intelligence, is widely used in industrial production to achieve efficient monitoring and prediction of production status including product quality. Data-driven soft sensor methods have attracted attention, which still have challenges because of complex industrial data with diverse characteristics, nonlinear relationships, and massive unlabeled samples. In this article, a data-driven self-supervised long short-term memory-deep factorization machine (LSTM-DeepFM) model is proposed for industrial soft sensor, in which a framework mainly including pretraining and finetuning stages is proposed to explore diverse industrial data characteristics. In the pretraining stage, an LSTM-autoencoder is first unsupervised pretrained. Then, based on two self-supervised mask strategies, LSTM-deep can explore the interdependencies between features as well as the dynamic fluctuation in time series. In the finetuning stage, relying on pretrained representation, the temporal, high-dimensional, and low-dimensional features can be extracted from the LSTM, deep, and FM components, respectively. Finally, experiments on the real-world mining dataset demonstrate that the proposed method achieves state of the art comparing with stacked autoencoder-based models, variational autoencoder-based models, semisupervised parallel DeepFM, etc.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available