Lithium-Ion Energy Storage Systems in Smart Grid Applications: Modeling and Analysis

Nowadays, modern smart grids require the implementation of Energy Storage Systems (ESS) in order to optimize the production, consumption and distribution of electrical energy. In particular, ESS add a degree of freedom to the system which allows to cover the mismatch between production and demand while providing other grid support services. Lithium Ion Batteries have proven to be mature to serve in several applications involving different stakeholders. In fact, their characteristics in terms of aging, specific power and response times are superior to other energy storage technologies. The aim of the thesis is to develop and study a model which simulates a Lithium-Ion based storage system operating in some selected applications. The model is built in the MATLAB and Simulink environment, and it considers several aspects: electrical connections, thermal behavior, aging prediction, controlling and economic performance. Overall, the comprehensive model is developed thanks to large amounts of experimental data which make it accurate and flexible. Great relevance is given to the aging model which is designed starting from experimental results and fatigue algorithms: both calendar and cycle aging are considered. Three relevant energy storage system applications are selected and studied: RES-BESS, energy arbitrage and island grid. In all cases a simulation of one year is performed employing the developed model after an accurate design procedure. The operating conditions of the storage system are checked and analyzed throughout the simulation. Aging is the main index to be evaluated to asses the sustainability of the application. Finally, depending on the specific application, revenue sources are identified and calculated. With current prices of Lithium Ion batteries, all applications resulted to be hardly economically sustainable. However, considering the decreasing trend of prices, these applications may be very interesting in the medium and long term (about a decade) confirming the relevance of the research.