An Integrated Cost-Aware Dual Monitoring Framework for SMPS Switching Device Diagnosis

Abstract

The ability of a switch-mode AC/DC power supply to shrink supplies is a benefit and a requirement for most electronic devices with limited space. Major failures in switch-mode power supply (SMPS) during adverse working conditions are subject to mostly the switching devices and capacitors. For effective condition monitoring of the SMPS, dual (or multiple) sensing provides a more reliable standpoint against the traditional single sensing techniques as it provides a more comprehensive paradigm for accurate condition monitoring. This study proposes an integrated approach to SMPS condition monitoring by exploiting statistically extracted features from current and voltage signals for system fault diagnosis based on electrical stress. Following a correlation-based feature selection approach, salient features were utilized for improved fault detection and isolation (FDI) using ML-based classifiers. Diagnostic results by the classifiers reveal that the random forest and gradient boosting classifiers are highly reliable but computationally expensive when compared with the others while the decision tree was quite cost-efficient with reliable diagnostic results. The proposed framework is effectively applicable for use in diagnosing the switching devices and classification at different states.

Towards Data-Driven Fault Diagnostics Framework for SMPS-AEC Using Supervised Learning Algorithms

“…The positive linear matrix shown in the correlation means there is a strong relationship between the variables extracted. A similar correlation matrix can be seen in [11,13] and in statistical feature extraction application [42][43][44][45].…”

“…As most machine learning classifiers need discriminant qualities as inputs to attain acceptable diagnostic accuracy, there are significant correlations between derived condition health indicators/features. The feature selection findings are shown in Figures 9,11,13 and 15, which should provide adequate and satisfactory diagnostic results and serve as the right condition indicators for the aluminium electrolytic capacitors used in SMPS. For ML-based diagnosis training and assessments, the variation-dimensional selected features are taken from the training and test datasets, respectively.…”

“…Designers can attempt to reduce reliability difficulties by studying the relationship between the electrical and thermal properties of the semiconductor devices used in a power electronic circuit. Common factors that can reduce the reliability of power electronics are temperature (which could result in loss of electrolyte and leakage current), over current (which could result in dry out/electrolyte loss, over voltage (could result in oxide degradation) and environmental conditions [8][9][10][11]. The combined effect of the listed factors could result in a decrease in capacitance, an increase in ESR, and changes in the dissipation factor.…”

LSTM-Autoencoder for Vibration Anomaly Detection in Vertical Carousel Storage and Retrieval System (VCSRS)

“…The Pearson, Spearman, and Kendall mathematical expressions are represented by Equations ( 1 )–( 3 ), respectively. Several of these correlation coefficient techniques have been used in the literature to reduce computational costs, extract meaningful features for diagnostics and prognostics of various industrial systems, and also ensure that discriminative features are chosen for training and testing machine learning models [ 37 , 38 , 39 , 40 , 41 ]. …”

RUL Prediction of Switched Mode Power Supply Using a Kalman Filter Assisted Deep Neural Network

“…A total mathematical model is required to perform model-based PHM, which is often unavailable to many systems. On the other hand, due to the advancement of resource availability, the data-driven PHM approach is widely used nowadays [1][2][3][4][5][6].…”