(137e) The Effect of Distributed Grid-Level Energy Storage On Transmission Congestion and Real-Time Dispatch: A Case Study of the Pjm Grid

Some U.S. regional transmission organizations (RTOs) are studying the possibility of integrating offshore wind generation into the electrical grid, where aggregate wind power generation may be as high as 20GW and may drop off to zero in as little as an hour. As penetration levels rise, grid-level energy storage will become a necessity in order to mitigate the uncertainty in wind power generation. As a result, RTOs will need to reformulate their generation planning algorithms to include storage.

We present an application of an approximate dynamic programming (ADP) algorithm for the control of grid-level energy storage devices distributed across the electricity grid, under uncertainty in future wind power generation and grid congestion. Our case study is based on the direct-current approximation the PJM electrical grid, which includes over 9,000 buses and 14,000 transmission lines. Our study analyzes the spatiotemporal patterns of storage in the presence of grid congestion, as well as the impact of using mixed energy storage portfolios of different efficiencies, power and energy capacities on electricity transmission.