Detection and Fault Prediction in Electrolytic Capacitors Using Artificial Neural Networks

Abstract

Capacitors are electronic components that present a considerable variation in their characteristics during their useful life. After being submitted to several charge/discharge cycles, capacitors present losses in capacitance values and operate differently from the nominal characteristics. PHM (Prognostics and Health Monitoring) techniques can be used to monitor the evolution of a capacitor health condition and to predict its RUL (Remaining Useful Life). This paper uses artificial neural networks to monitor the degradation index of capacitors and predict the corresponding RUL. Different neural network architectures are investigated: MLP (Multilayer Perceptron), RBF (Radial Basis Function), and ELM (Extreme Learning Machine). The performances of the different architectures are compared in terms of the coefficient of determination (R 2 ) and the Mean Squared Error (MSE). The accuracy of RUL predictions are compared based on the Relative Accuracy (RA) indicator, which is a performance indicator proposed in the literature to evaluate PHM algorithms.

Fault Prediction for Capacitor of Power Converters Based on CEEMDAN and GWO-RELM

“…Its root mean square error and mean absolute error are decreased by 8.6% and 1.7%, respectively. Based on NASA data set, Mesquita applied different artificial neural networks to predict the remaining life time (RUL) [11], and Delanyo proposed a prediction method based on Bi-LSTM [12]. Broad learning system and LSTM are combined into a fusion network.…”