Variability and the Locational Marginal Value of Energy Storage

TitleVariability and the Locational Marginal Value of Energy Storage
Publication TypeConference Paper
Year of Publication2014
AuthorsSubhonmesh Bose, Eilyan Bitar
Conference Name2014 IEEE 53rd Annual Conference on Decision and Control (CDC)
Date Published12/2014
Conference LocationLos Angeles, CA, USA
ISBN Number978-1-4799-7746-8
Keywordsenergy storage, Locational marginal value, reliability and markets, RM11-006

Given a stochastic net demand process evolving over a transmission-constrained power network, we consider the system operator's problem of minimizing the expected cost of generator dispatch, when it has access to spatially distributed energy storage resources. We show that the expected benefit of storage derived under the optimal dispatch policy is concave and non-decreasing in the vector of energy storage capacities. Thus, the greatest marginal value of storage is derived at small installed capacities. For such capacities, we provide an upper bound on the locational (nodal) marginal value of storage in terms of the variation of the shadow prices of electricity at each node. In addition, we prove that this upper bound is tight, when the cost of generation is spatially uniform and the network topology is acyclic. These formulae not only shed light on the correct measure of statistical variation in quantifying the value of storage, but also provide computationally tractable tools to empirically calculate the locational marginal value of storage from net demand time series data.