An Artificial Neural Network Framework for Automated Fault Diagnosis in Industrial DC-DC Converters

Abstract

The operational longevity of electrical equipment is often limited by gradual aging and degradation resulting from various stressors, which leads to a decrease in system reliability over time. In industrial environments, DC machines are commonly powered by chopper voltage sources, making them susceptible to a range of faults that affect switches, capacitors, and coils. Early detection and localization of such defects are crucial to ensure efficient maintenance and prevent faults. To address the risk of severe failures in DC-DC converters, advanced diagnostic techniques are required. This study explores the application of artificial neural networks (ANN) for automatic fault detection and diagnosis in DC-DC converters. The proposed ANN-based methodology demonstrates effective identification of converter faults, offering a reliable and practical solution for real-time monitoring and predictive maintenance in industrial settings