Modeling of a Distributed Generation System using Adaptive Neuro Fuzzy Inference Approach

A wind/photovoltaic/fuel cell distributed generation system (DGS) with battery storage that was built at UMass-Lowell consists of several power generating units and their controls. To study the response of such a system to various operating conditions, a complete model is developed. To develop the model, DGS is divided into subsystems, each subsystem has a sub model. The sub models are developed based on known information, and then the sub models are integrated to form an overall DGS model. The model is developed using an adaptive neuro fuzzy inference system (ANFIS) that is based on neuro-fuzzy approach. The ANFIS model of the DG system is used to predict the system performance. The model is validated by comparing the simulation results with the experimental ones.