An Automated Sensor Fusion Approach for the RUL Prediction of Electromagnetic Pumps

Ugochukwu Ejike Akpudo,Hur Jang-Wook

doi:10.1109/access.2021.3063676

Abstract

The remaining useful life (RUL) prediction of industrial cyber-physical system components demands the use of reliable prognostics parameters and frameworks. Against the traditional use of a single measure of degradation, data from multiple sensors provide abundant characteristic information for modeling, assessing, and extracting useful parameters via appropriate signal processing and sensor fusion methods. This study introduces a multi-sensor prognostics approach which merges highly prognosible statistical features from vibrational and pressure sensor measurements after a multi-level wavelet decomposition of the signals. The prognostic algorithm presented in this work for solenoid pump RUL prediction is a multi-objective genetic algorithm-optimized long short-term memory (MOGA-LSTM) which accepts the fused sensor features as input and returns the RUL of the pump as output. The framework was tested on a run-to-failure experiment on a VSC63A5 Solenoid pump following a significant pump malfunction caused by a clogged suction filter after the test. Using standard prognostic performance evaluation metrics, the performance of the prognostics framework was compared with other reliable state-of-the-art methods with a remarkable comparative advantage in addition to better automation potentials for real-time condition monitoring and RUL prediction.

Highlights

R ELIABILITY studies on industrial cyber-physical systems (ICPSs) have become one of the central research focus in academia owing to the fast-growing industry 4.0 revolution
Studies on ICPSs are currently receiving vast patronage and research interests covering vast modules ranging from industrial design technologies, manufacturing, smart control, robotics, engineering, etc.; regardless of the domain of interest, system reliability, safety, and maintainability are some of the key concerns which have motivated the on-going shift from traditional corrective to the more effective condition-based maintenance(CBM) with prognostics and health management (PHM) at its core [1]
The remaining useful life (RUL) prediction is achieved by first computing the ground truth and with the trained model, the RUL from TSP to end of life (EOL) is estimated

Summary

Introduction

R ELIABILITY studies on industrial cyber-physical systems (ICPSs) have become one of the central research focus in academia owing to the fast-growing industry 4.0 revolution. Deliberate attempts are being made to prevent occurrence of these undesired outcomes, absolute failure prevention and control seems unachievable with the major causes emanating from uncertainties, environmental factors, human errors, and equipment fatigue. These uncertainties and many past recorded accidents only validate the need for reliable process monitoring, fault diagnosis and prognostics system design schemes that can be applied to large-scale processes [2], [3]. Similar to discrtete Fourier transform (DFT) and short-time Fourier transform (STFT), wavelet transform can be viewed as the projection of a signal into a set of basis functions called wavelets Such basis functions offer localization in the frequency domain. The major difference is: Fourier transform decomposes the signal into sines and cosines whereas the wavelet transform uses functions (wavelets) that are localized in both the real and Fourier space, thereby providing more intuitive information in a signal

Objectives

Results

Conclusion