Integrating AE-CNN with Smart Relaying and SSCB for Enhanced Three-Phase Fault Detection and Mitigation

Amidst escalating demands for electricity, the complexity of power systems has significantly increased, making fault detection increasingly challenging. This research addresses the critical need for efficient fault detection and mitigation in three-phase power systems, proposing a novel solution integrating autoencoder-based Convolutional Neural Networks (AE-CNN) with Smart Relaying and Solid-State Circuit Breakers (SSCB). We developed and tested a Hybrid Neural Network model, trained with data processed from MATLAB simulations, to accurately detect faults in these systems. Our AE-CNN model achieved a remarkable 99.99% accuracy in fault detection, underscoring its effectiveness. Post-detection, the SSCB technology swiftly mitigated faults. This research not only demonstrates the precise detection and rapid mitigation of faults in three-phase power systems but also exemplifies the successful adoption of cutting-edge technologies. Our work makes a significant contribution to the evolution of modern power systems, offering a highly reliable and innovative solution to maintain the integrity and stability of electrical networks in the face of increasing complexities.