Planning and Operation of Energy Storage Systems in Power Systems

Abstract

The increase in renewable power generation has greatly changed the current energy market dynamics. The cheaper, cleaner but intermittent characteristics of renewable power generation such as wind has accelerated the retirement of large coal plants and investments in flexible resources such as fast response generators, energy storage and transmission facilities. However, this transition requires tremendous effort on how to improve the market design to better handle the uncertainty on renewable generation, and to facilitate the deployment of more flexible resources such as energy storage. In this dissertation, we focus on the planning and operation of energy storage systems. To begin with, we discuss in detail different optimization methods that have been proposed to handle this growing uncertainty. Yearly simulations were performed to compare the solution accuracy and the computing efficiency. Then two stochastic multi-stage co-planning models are proposed to coordinate investments in battery energy storage and transmission expansion, and in battery energy storage and fast generation. These co-planning models have a 25-year horizon and consider not only the uncertainty on both wind capacity and load increase, but also the degradation of the batteries. A sensitivity analysis is performed to study the competition and cooperation relationship between these resources, their investment patterns under different geography, and the correlation between the profiles of wind generation and load. At the end of this dissertation, a stochastic energy and ancillary service co-optimization model is proposed to evaluate the contribution of storage to both energy arbitrage and ancillary services. The actual requirements for regulation reserve and spinning reserve are quantified by combining intra-hour system operation with day-ahead stochastic optimization.