Cost Allocation Mechanism for Deep Peak Regulation in Power Systems Based on Cooperative Game Theory for Multiple Peak Regulation Demand Entities

Abstract

As the scale of renewable energy integration continues to expand, the uncertainty and variability of power system operation has increased. Meanwhile, peak regulation resources have become increasingly diverse, making the formation of peak regulation costs more complex. Existing allocation mechanisms therefore have difficulty directly linking the temporal power profile of each entity to its contribution to the optimized deep peak regulation cost and the corresponding cost responsibility. This paper applies the standard Shapley value framework to deep peak regulation cost allocation involving multiple peak regulation demand entities. An energy-preserving substitute scenario method is applied to heterogeneous demand entities. The actual profiles of coalition members are retained, whereas the profiles of nonmembers are replaced by their mean values, thereby removing temporal fluctuations while preserving total energy over the dispatch horizon. For each coalition-specific net load scenario, a dispatch optimization model considering the piecewise deep regulation capability of thermal units and the intertemporal constraints of pumped storage are re-solved, and the resulting minimum cost is used to construct the coalition cost function. The signed Shapley value is then used to calculate the average marginal cost effect of each entity across all coalitions, after which a practical positive contribution settlement rule converts the signed values into nonnegative and budget-balanced charges. A provincial system case study shows that, compared with energy proportional and peak power proportional allocation, the proposed method explicitly accounts for the temporal characteristics and interactions of heterogeneous power profiles and links them to changes in the optimized deep peak regulation cost through coalition-specific re-optimization.