Ensemble Learning for AC Microgrid Fault Detection and Classification: A Random Forest Perspective

As the need for energy is increasing, it results in the dependence on fossil fuel, which creates risks such as global warming, depletion of resources and causes pollution. Microgrids are self-sufficient energy systems that discourage use of fossil fuel rather they use renewable sources like sun, wind etc. Because of bidirectional current flow in microgrids it makes fault detection more difficult and a challenge to conventional protection techniques. Integrating the dynamic generator also increases complexity. Here in this paper we have used the Random forest approach for detection and classification of the fault. It operates on the labeled datasets and belongs to the ensemble learning category which combines the outputs of multiple decision trees. Ensemble learning enhances resource allocation and enhances fault detection and classification capabilities in AC microgrid systems. Hence, the integration of the Machine learning application in microgrids helps to contribute significantly to a resilient and sustainable energy future.