ANN based optimization of price-based demand response management for solar powered nanogrids

The quest for cheap, green and most importantly sustainable supply of electrical power has boosted the employment of renewable energy resources (RERs) in the existing power systems around the globe. The new millennium has witnessed enormous research in the optimized utilization of RERs in the power system to achieve maximum benefits of the RERs. The present body of knowledge records a handsome number of research efforts regarding the optimization techniques of demand response (DR). The main focus of the research efforts presented so far is the use of fixed pricing schemes in the different optimization algorithms. In contrast, the current research effort aims to present real time price (RTP) based demand response. The scheme is implemented considering the residential power system of a modern locality equipped with state of the art loads. A nanogrid (NG) has been developed in MATLAB for testing of the proposed optimization scheme. The simulation model comprises a grid-connected solar PV system supplying the house loads, which are categorized according to their utilization as fixed and flexible loads. An artificial neural network (ANN) is used to schedule flexible loads to maximize profit. The simulation results confirm the fruitfulness of the proposed dynamic pricing scheme, being profitable over the conventional fixed pricing scheme.