AlexNet-Microgrid: Advanced Deep Learning for Islanding Detection in Microgrids

An essential component of microgrid operation is islanding detection, which guarantees the secure and dependable running of power systems-especially with the growing integration of renewable energy sources. Conventional islanding detection approaches frequently rely on passive and active procedures which can be slow or intrusive. This research employs the powerful convolutional neural network (CNN) architecture of Alex Net to provide a novel solution for islanding identification in microgrids using a deep learning-based framework. The suggested technique converts microgrid time-series data into visualizations that resemble images, which are subsequently input into AlexNet to distinguish between normal and islanding circumstances. The efficacy of the Alex Net-based method is demonstrated by extensive experiments conducted on simulated microgrid data, which yield excellent accuracy and quick detection times. The findings imply that islanding detection systems in contemporary microgrids can be greatly improved in terms of performance and dependability by using deep learning models, particularly AlexNet.