A hybrid RBFNN–BBMO methodology for robust energy management in grid‐Connected microgrid

Abstract

This dissertation shows energy scheduling in grid-connected micro grid (MG) using an amazing hybrid method for optimal energy management. The hybrid method is the mixture of Radial Basis Neural Function Network (RBFNN) and Bumble Bees Mating Optimization (BBMO) and is therefore referred to as RBFNN-BBMO technique. The MG system consists of photovoltaic, wind turbine, battery storage and micro turbine systems. Here, batteries are used to maintain a constant and stable output power. The necessary load demand of the grid-connected MG scheme is continually monitored using the RBFNN method. BBMO optimizes the perfect combination of MG taking into account the expected load demand. In addition to decrease the impact of renewable energy prediction mistakes, a two strategy for MG energy management is adopted. The first approach is to program various renewable energy sources (RESs) during the operation of MG in order to minimize the electricity costs. The second approach is power flow balancing and forecasting errors, decreasing from the scheduled energy reference based on the summarized rule. Access of RESs, power supply and charging modules are the constraints. Two existing methods such as the Genetic Algorithm and the Cuttlefish Algorithm compare the supremacy of the proposed method.