Predictive Maintenance Model Based on Multisensor Data Fusion of Hybrid Fuzzy Rough Set Theory Feature Selection and Stacked Ensemble for Fault Classification
No Thumbnail Available
Date
2022
Journal Title
Journal ISSN
Volume Title
Publisher
Hindawi
Abstract
With the rising demand for integrated and autonomous systems in the eld of engineering, e cient frameworks for instant
detection of performance anomalies are imperative for improved productivity and cost-e ectiveness. is study proposes a
systematic predictive maintenance framework based on the hybrid multisensor fusion technique of fuzzy rough set feature
selection and stacked ensemble for the e cient classi cation of fault conditions characterised by uncertainties. First, a feature
vector of time-domain features was extracted from 17 multiple sensor signals. en, a comparative study of six di erent Fuzzy
Rough Set Feature Selection (FRFS) methods was employed to select the various combinations of optimal feature subsets for
various faults classi cation tasks. e determined optimal feature subsets then served as inputs for training the stacked ensemble
(ESB(STK)). In the ESB(STK), Support Vector Machine (SVM), Multilayer Perceptron (MLP), k-Nearest Neighbour (k-NN), C4.5
Decision Tree (C4.5 DT), Logistic Regression (LR), and Linear Discriminant Analysis (LDA) served as the base classi ers while the
LR was selected to be the metaclassi er. e proposed hybrid framework (FRFS-ESB(STK)) improved the classi cation accuracy
with the selected combinations of optimal feature subset size whiles reducing the computational cost, over tting, training runtime,
and uncertainty in modelling. Overall analyses showed that the FRFS-ESB(STK) proved to be generalisable and versatile in the
classi cation of all conditions of four monitored hydraulic components (i.e., cooler, valve, accumulator, and internal pump
leakage) when compared with the six base classi ers (standalone) and three existing ensemble classi ers (Stochastic Gradient
Boosting (SGB), Ada Boost (ADB), and Bagging (BAG)). e proposed FRFS-ESB(STK) showed an average improvement of
11.28% and 0.88% test accuracies when classifying accumulator and pump conditions, respectively, whiles 100% classi cation
rates were obtained for both cooler and valve.
Description
This article is published by Hindawi,2022 and is also available at https://doi.org/10.1155/2022/4372567
Keywords
Citation
Hindawi Mathematical Problems in Engineering Volume 2022, Article ID 4372567, 24 pages