Optimization operational framework of virtual power plant based on dynamic information entropy assessment approach

The inherent uncertainty and volatility in distributed energy resources pose significant challenges to the stable operation of virtual power plants. Existing optimization methods for virtual power plants may not adequately address the cumulative impact of uncertainties in operation. This study develops a hydrogen-buffered time-shifting regulation mechanism to suppress uncertainty accumulation, establishing an information entropy-driven optimization framework for the hydrogen-buffered multi-carrier virtual power plant. In the assessment phase, the information entropy approach is used to evaluate the confidence level of the forecasted output power from distributed energy resources, enabling dynamic adjustment of the optimization scheduling constraints. During the operation optimization phase, hydrogen energy is utilized as a backup power source, with hydrogen storage capacity and operational cost as the objective functions. A two-stage multi-objective optimization model is established and solved using the zero-sum game weighting method. This approach enhances system feasibility through dynamic constraint relaxation, effectively mitigating uncertainty accumulation. Finally, the α steady-state distribution is utilized to generate operational scenarios. Simulations are analyzed over multiple consecutive run cycles. The findings indicate that the proposed method is capable of reducing operating costs by 24.92%, while concurrently meeting the established reliability constraints. The proposed method has the capacity to develop flexible and efficient operation strategies in the context of uncertainty. • Proposed hydrogen-buffered VPP model improves cross-temporal energy scheduling flexibility. • Develops an entropy-based forecast evaluation method for DERs. • Designs an evaluation–operation optimization framework reducing long-term costs 24.92 • Constructs α -stable scenarios to overcome typical-day limits on uncertainty accumulation.