Strategic Investments of Large Scale Battery Energy Storage Systems in the Wholesale Electricity Market

TL;DR

This paper presents a three-phase strategic investment approach for large-scale battery energy storage systems in wholesale electricity markets, accounting for market price changes and optimizing configurations and operations.

Contribution

It introduces a novel three-phase method combining heuristics, Bayesian optimization, and iterative dispatch to improve BESS investment decisions considering market dynamics.

Findings

Bayesian optimization is 16 times faster than grid search.

Ignoring LMP changes leads to a 21% profit overestimation.

The model effectively captures market price impacts on BESS investments.

Abstract

In this paper, we study the strategic investment problem of battery energy storage systems (BESSs) in the wholesale electricity market from the perspective of BESSs owners. Large-scale BESSs planning without considering the possible wholesale market price change may result in possible locational marginal price (LMP) changes. To overcome such limits, we propose a three-phase approach for the BESS investment problem. In Phase-1, we conduct a search for the optimal BESS configurations via a congestion-based heuristics and Bayesian optimization. In Phases 2 and 3, we alternatively dispatch optimization tasks to optimize the wholesale market clearing for LMPs and identify the optimal schedule for BESSs operations. We validate the model using a ten-year simulation on the Electric Reliability Council of Texas (ERCOT) market. Experimental results show that the Bayesian optimization model runs 16 times faster than the grid search model in achieving the same solution quality. Further, the iterative method demonstrates that the model without considering possible LMP changes makes a 21% profit overestimation.