Multitimescale Incentive-Based Demand Response Scheduling Models

Incentive-based demand response (IBDR) enhances grid stability, lowers carbon emissions, and optimizes economic returns. However, effectively scheduling IBDR resources is challenging due to uncertainties in user response, especially in large-scale real-time electricity markets. To address this issue, we propose a multitimescale scheduling model grounded in Stackelberg game theory, combining hourly level optimization with a minute-level dynamic incentive adjustment mechanism. This framework leverages the sample average approximation algorithm to manage user response uncertainty. It employs a two-stage variable-step-size privacy-preserving algorithm to balance computational efficiency with data privacy. Numerical simulations show that our approach significantly reduces power deviations by up to 74.98% in certain cases compared to fixed-incentive schemes and cuts optimization times. These findings underscore the importance of modeling uncertainty scenarios (with around 50 scenarios providing an optimal tradeoff) and demonstrate hierarchical decision making’s benefits for grid operators and load aggregators. For engineering managers and policymakers, our results offer actionable strategies for fine-tuning incentive structures, ensuring robust system performance, and supporting large-scale real-time demand response implementations. The proposed model provides a practical pathway to more stable and cost-effective smart grid operations by integrating long-term planning with real-time control.