Intelligent Flexible Priority List for Reconfiguration of Microgrid Demands Using Deep Neural Network

Abstract

The mounting demand on electrical energy and the surge of new forms of loads such as electric vehicles have added extra challenges to the current picture of the power systems. Generally more failures are occurring in the system and the largest portion of these faults come from the distribution network. The concept of the microgrid is considered to be a solution to this issue. Microgrids should achieve smart and robust load restoration, in which a decision is made on which load should be supplied first and what are the loads that follow. In this paper, a smart, dynamic load priority list will be modeled using artificial neural network (ANN), where different categories of loads such as residential, commercial, industrial and hospital will be prioritized for restoration based on the given time, reliability indices and amount of available energy. The ANN based priority list showed exceptional results in terms of flexibility and understanding of the current energy and reliability status. The results can be further used as an input direct load control functions to intelligently determine which loads should be curtailed and which ones are uninterruptible.