In this paper, we address the optimal energy storage management and sizing problem in the presence of renewable energy and dynamic pricing. We formulate the problem as a stochastic dynamic programming problem that aims to minimize the long-term average cost of conventional generation used as well as investment in storage, if any, while satisfying all the demand. Storage is modeled with ramp constraints, conversion losses, dissipation losses and investment cost. Assuming the renewable energy, demand and prices evolve as exogenous stochastic processes, we show the existence of the optimal storage management policy. The policy is a dual-threshold dependent balancing policy with thresholds that decrease with prices. Under this policy, we provide structural results that help in computing the optimal storage size efficiently. We characterize how the cost and efficiency of storage, dynamic pricing and the uncertainty in the system impact the size of optimal storage and its gain. In a surprising result we prove that under constant prices for storage to be profitable the ratio of the amortized cost of storage to the price of electricity should be less than 1/4 . We present a numerical study on real-world data that demonstrates the implementation of the optimal policy and provides realistic savings using storage.
Citation
July, 2012. The full paper is published in IEEE Transactions on Power Systems, Volume 30, Issue 3, pp 1164-1181, May 2015. Please use this for future citations.