Scheduling Battery Energy Storage Systems Under Battery Capacity Degradation Uncertainty

The increasing penetration of renewable energy resources and the decreasing cost of battery energy storage in recent years has led to a growing interest in using batteries to provide grid services like frequency regulation. In this paper, we discuss the advantages and disadvantages of different battery degradation models and the impacts that model choice can have on the assumed cost of energy capacity loss due to operation. We also explore the effects of modeling degradation as an uncertain process by extending a two-stage, multi-period optimization problem for scheduling the operation of a battery providing multiple services with risk aversion. We use stochastic dual dynamic programming to derive a policy for the problem. Case study results show that using a stochastic degradation model with risk aversion produces a policy for more conservative battery use and longer lifespan in comparison to that obtained with a deterministic degradation model.