Arbitraging Variable Efficiency Energy Storage Using Analytical Stochastic Dynamic Programming

This paper presents a computation-efficient stochastic dynamic programming algorithm for solving energy storage price arbitrage considering variable charge and discharge efficiencies. We formulate the problem using model real-time prices as Markov process. Then we propose an analytical solution piecewise linear approximation of value-to-go function. Our achieves extreme computation performance solves proposed one operating day in less than second on personal computer. demonstrate our approach historical data from four zones New York Independent System Operator, with case studies comparing different models settings. results show that method captures 50% to 90% profit compared perfect forecasts. In particular, more 80% three out when batteries two-hour duration realistic degradation cost.