Architectures intégrées de gestion de l'énergie pour les multisystèmes autonomes utilisant le stockage par accumulateurs.

Abstract

Because of the global growth in energy demand, and because of the problems posed by fossil fuels, t : he depletion of these resources and the environmental problems caused by the emission of gases, renewable energy sources (photovoltaic, wind ...) represent an interesting alternative for the electrification of isolated sites. However, these sources have a disadvantage of being dependent on weather conditions. In order to reduce fluctuations in production caused by the random nature of these resources and to satisfy the requirements of the load the solution to be retained is the assembly of the different sources of electric power generation. This work is a contribution to the study of an autonomous hybrid energy system, which consists of systems: photovoltaic (PV), wind, PEM fuel cell and a storage system on several aspects: modeling, supervision and simulation. Energy storage is a key factor in an isolated site hybrid energy system. However, it is characterized by a very limited lifespan and high cost. In order to avoid its degradation by deep discharges or overloads on the one hand and on the other hand to optimally manage the energy obtained from the various available energy resources, a management strategy based on The state of charge of the battery (SOC) is designed to control the proposed autonomous multi-source system. Where PV and wind systems are considered primary sources, while the PEMFC and the battery are used as emergency sources, the electrolysis is used to consume the excess power available when the battery is fully charged. The modeling and simulation of the PV and wind systems were presented with an MPPT command to maximize the power delivered. To maintain the DC bus voltage at its reference value, a modified control algorithm is proposed which requires the control of the bidirectional converter to DC / DC current between the battery and the DC bus. The results obtained with the Matlab / Simulink software show the efficiency of the various commands used.