Unsupervised Minimum Redundancy Maximum Relevance Feature Selection for Predictive Maintenance

Abstract

Identifying and selecting optimal prognostic health indicators in the context of predictive maintenance is essential to obtain a good model and make accurate predictions. Several metrics have been proposed in the past decade to quantify the relevance of those prognostic parameters. Other works have used the well-known minimum redundancy maximum relevance (mRMR) algorithm to select features that are both relevant and non-redundant. However, the relevance criterion is based on labelled machine malfunctions which are not always available in real life scenarios. In this paper, we develop a prognostic mRMR feature selection, an adaptation of the conventional mRMR algorithm, to a situation where class labels are a priori unknown, which we call unsupervised feature selection. In addition, this paper proposes new metrics for computing the relevance and compares different methods to estimate redundancy between features. We show that using unsupervised feature selection as well as adapting relevance metrics with the dynamic time warping algorithm help increase the effectiveness of the selection of health indicators for a rotating machine case study.

Predicting Machine Failures from Multivariate Time Series: An Industrial Case Study

“…SVM is another approach commonly used in supervised failure prediction [12,15,16,25,26,43,[46][47][48]. The work in [26] presents an application of SVM to railway systems and evaluates the performances using False Positive Rate (FPR) and recall.…”

A Review of Multisensor Data Fusion Solutions in Smart Manufacturing: Systems and Trends

“…The goal of this method is to find those indicators that minimise the redundancy of the data, as the removal of one feature from highly mutually dependent sets will not lead to a change in the information given by them; at the same time, the method must maximise the relevance to the target classes. There is an unsupervised version of this algorithm (UmRMR) that has been used for predictive maintenance in rotating machinery [120] and in structural health monitoring [121].…”

Long Horizon Anomaly Prediction in Multivariate Time Series with Causal Autoencoders

“…However, the existing data-driven approaches for PdM revolve around the development of supervised models which aim at specific labeled data or/and rely on feature extraction signal processing tools such as variants of Fourier transform, Wavelet transform, statistical based and principal component analysis (PCA) (Tang et al, 2019;Huang et al, 2016;X. Li et al, 2020;Langone et al, 2020;Wang et al, 2018;Hadj-Kacem et al, 2020;Hamaide & Glineur, 2021). In (Hadj-Kacem et al, 2020), a machine learning-based anomaly prediction model was proposed using forecasting future time steps mechanism for mobile networks.…”