AI-Based Fault Detection and Isolation for UPQCs: Modelling, Simulation, and Power Quality Improvement

Abstract Unified Power Quality Conditioners (UPQCs) enhance power quality in electrical distribution networks. However, faults in UPQCs can lead to power quality issues and system outages. This paper investigates the application of artificial intelligence (AI) techniques for advanced fault detection and isolation in UPQCs. We propose modeling and simulating various AI-based approaches, including artificial neural networks (ANNs), fuzzy Logic, and hybrid neuro-fuzzy systems. The primary objective is to develop an efficient, accurate, and real-time fault detection system that improves the overall performance and reliability of UPQCs. Simulation results demonstrate that AI techniques can accurately detect and classify faults. While AI-based fault detection in UPQCs is promising, challenges such as real-time data processing and interpretation must be addressed for widespread adoption. Future research directions include deep learning architectures, unsupervised learning, Internet of Things (IoT) integration, and explainable AI to further enhance fault detection capabilities in UPQCs. Addressing these challenges will contribute to developing intelligent and self-healing power distribution networks, with AI-based fault detection playing a pivotal role.