Experimental Validation of Neuro-fuzzy Energy Management for a DC Electrical Micro-network
Abstract
Energy management is an important component for the profitability of energy production systems in a context where the use of photovoitaic (PV) systems with storage is expanding. It allows to control the energy produced by the solar system and at the same time to supervise the storage system. The objective of this work is to propose a system based on the Adaptive Neuro-Fuzzy Technique (TANF) with particle swarm for the energy management system (EMS) of a hybrid PV battery system (HPVB). The proposed EMS consists of two parts. A part concerning the optimization of the power produced by the PV module (PVM) and a part concerning the supervision of the battery. The same hybrid method is used for both separate applications. The methods are validated under a dSPACE DS1202. The results obtained confirm that the proposed method makes it possible to optimize the power of the solar source. While ensuring the protection of the battery against deep discharges and overloads whatever the weather conditions.