Artificial Neural Network in Photovoltaic-Battery Microgrid System for Controlling the Battery Current in French Guiana : Battery Life Improvement

International audience This paper presents an artificial neural network applied to control a standalone microgrid in French Guiana. This microgrid is composed of a Photovoltaic (PV) source and a battery storage to supply a DC load. In this paper, different configurations of neural network associated with the Levenberg–Marquardt algorithm are tested to choose the best configuration to optimize the control of our system. Then, we focus on the advantage of using this type of control compared to the conventional proportional integral control (PI control). Finally, we present and discuss the results of our simulation obtained with Matlab Simulink software