Online Storage Control for Stable Wind Power Commitment via Lyapunov Optimization

Wind power, a prominent renewable energy resource, contributes to a substantial yet volatile energy supply to the power grid, posing significant risks to both supply and demand. On the supply side, wind farms often incur penalty costs due to the discrepancy between the actual and committed wind power generation specified in the electricity supply contracts. To address this issue, energy storage systems are widely adopted for real-time regulation of wind power delivery, effectively mitigating power fluctuations. However, effective storage control is hindered by the uncertainties surrounding the future renewable generation, and the control process often lacks theoretical guarantees. To tackle these issues, we design a Lyapunov drift-based storage control algorithm customized for wind power commitment considering a practical quadratic penalty. We also provide theoretical guarantees for the proposed algorithm regarding system penalty costs, system dynamics and the value of storage. The numerical study verifies our theory and highlights the performance of our proposed algorithm.