A predictive maintenance model using Long Short-Term Memory Neural Networks and Bayesian inference

Abstract

The fourth industrial revolution is a profound transformation utilizing emerging technologies like smart automation, large-scale machine-to-machine communication, and the internet of things to change traditional manufacturing and industrial practices. The analysis of the huge amount of data collected in all modern industrial plants not only greatly benefited from modern tools of artificial intelligence but has also spurred the development of new ones. In this context, we present a new approach based on the combined use of Long Short-Term Memory (LSTM) neural networks and Bayesian inference for the predictive maintenance of an industrial plant. Hotelling’s T2 and Q metrics, assessing the degree of compatibility between the time-evolving industrial data and the output of the LSTM, trained on a reference period of good working condition, are used to update the Bayesian posterior probability about the good working condition of the plant. This method has successfully been applied to a real industrial case, and the results are presented and discussed.