Preliminary Study of Fault Detection on an Islanded Microgrid Using Artificial Neural Networks

The adoption of electrical microgrids is growing. This form of grid can be connected to the main grid or can operate in an islanded mode. This study looks at the challenges faced when considering fault detection and location where microgrids are involved. Different Artificial Intelligence network protection solutions are explored and discussed. A three-phase islanded microgrid is modelled using MATLAB with symmetrical and asymmetrical faults introduced at different load values. Voltage magnitude and phase angles are recorded and used as training data for an Artificial Neural Network (ANN) tool to detect and classify faults on the islanded microgrid. The ANN tool is able to detect, classify and locate faults with good accuracy.