Energy Storage State-of-Charge Management in Real-Time Markets

The evolving resource mix will likely increase net load variability and forecast errors. While numerous studies show that grid-scale energy storage resources (ESRs) can mitigate reliability issues, they often assume perfect foresight of net load when optimizing their State-of-Charge (SoC). In practice, electricity markets operate on a rolling horizon, frequently updating short-term load forecasts. At each market clearing, operators must allocate SoC between immediate and later needs under forecast uncertainty. Through simulations of operator-managed ESRs within PJM's real-time markets under a future resource mix, we demonstrate that both a greedy strategy that prioritizes smoothing load volatility as it occurs and a conservative strategy that holds energy for peak load arbitrage can exacerbate reliability issues. We propose an SoC management framework that balances between the greedy and conservative strategies, improving reliability by absorbing near-term net load shocks while maintaining SoC for expected peak load beyond the horizon of real-time processes.