Integrating Energy Storage Systems into Renewable Grid Applications: A Model-based Approach

Thesis (Ph.D.)--University of Washington, 2019Accurate and economical sizing of stand-alone power system components, including batteries, has been an active area of research in renewable grid applications, but current approaches to integrating batteries into the entire renewable grid components do not make them economically feasible. Typically, batteries are treated as a black box that does not account for their internal states in current renewable grid simulations. This might lead to under-utilization and over-stacking of batteries. In contrast, detailed physics-based battery models, accounting for internal states, can save a significant amount of energy and cost, utilizing batteries with maximized life and usability. Therefore, it is important to identify how efficient physics-based models of batteries can be included and addressed in grid control strategies. In this dissertation, an efficient battery/grid integration framework will be studied and analyzed. Simple examples for microgrids and the direct simulation of the same including physics-based battery models will be presented. A representative microgrid example, which integrates stand-alone PV arrays, a maximum power point tracking controller, lithium-ion batteries, and power electronics, is illustrated. The results of the proposed approach are compared with the conventional approaches and improvements in performance and speed are reported. Also, Vanadium redox flow batteries are promising energy storage systems for renewable grid applications. Open source codes, batch-cell models, and experimental data are provided to enable people’s access to robust and accurate models and optimizers for an efficient battery/grid integration framework