National audience ; The development of decentralized, distributed, stationary electric microgrids producing energy as close as possible to the consumers, is a key challenge of the energy transition. Microgrids can contribute to the massive renewables energies integration and allow the optimization of the energy use according to several economic and ecological criteria. Batteries are the most widely used Energy Storage Systems (ESS) in microgrids, but are among the most expensive components with a relatively short lifetime. Therefore, optimize the sizing of sources and batteries (sizes, technologies and topologies) considering the energy management strategies is the general objective of the optimal design of a microgrid. For this purpose, many tools and studies exist using battery behavior and aging models. However, few works evaluate the accuracy of these models in real operating conditions, nor their impacts on the resulting sizing or energy management decisions. The research path chosen in this thesis consists in the analysis of different ESS models, considering their operation and degradation and used in optimal microgrid design methods. The objective is to evaluate the ability of these models to be integrated in a modeling environment dedicated to multi-criteria optimization (cost, use of locally produced energy, lifetime.) to size microgrids and develop energy management strategies. To validate this work, a test bench of a low voltage DC microgrid type has been built. This bench allows to emulate a microgrid from real production and consumption data and thus to evaluate the behavior of the ESS under controlled conditions but close to a real operation. A comparative study of battery models has been then made to find the most appropriate ones for microgrids design. An analysis of the accuracy of these models was carried out in order to study their ability to represent the battery behavior and the typical variations related to the stresses impacting their life span. For this purpose, usage profiles related to ...