ANN-Based Statistical Computation for Remote End Fault Monitoring of the IEEE 14 Bus Microgrid Network

This paper focuses on an artificial neural network-based fog computing approach for remote end fault super vision in microgrid systems. The work presented here considers an Artificial Neural Network-based adaptive Line-to-Ground (L-G) and Line-to-Line (L-L) fault monitoring system for a standard IEEE 14 bus microgrid. Discrete Wavelet Transformation (DWT)-based study of statistical parameters of the currents going out from the various generator buses is carried out both in healthy and faulty situations. Faults (L-G and L-L) are created at different load buses and also an algorithm is proposed for detecting fault location. The rule set proposed here is unaffected by variations in fault resistance, making it very much suitable for ground fault monitoring. It provided satisfactory results when examined with various unknown cases. A fog computing layer helps in fast execution and fewer data storage requirements in the cloud. This assessment may be expanded for other kinds of faults in a microgrid system.