Optimizing the Capacity Cost Recovery Mechanism of Shanxi Power Market through Improved Double-layer Stackelberg Game Model

Abstract

Aiming at the problem of insufficient capacity cost recovery caused by the fluctuation of renewable energy and information asymmetry in Shanxi power market, this paper constructs an improved two-layer Stackelberg game model. The model constructs a sequential decision-making framework of leaders and followers. The backward induction method is used to solve the equilibrium, and the information entropy is innovatively introduced to quantify the uncertainty of enterprise data, and the dynamic adjustment equation of compensation coefficient is established. When the bid deviation exceeds the threshold, the cost review is automatically triggered, and the segmented subsidy function is designed simultaneously to convert the deviation of renewable energy consumption rate into the increase or decrease of thermal power subsidy, forming a "step punishment-excess reward" mechanism. The non-convex problem is handled by improving the ADMM (Alternating Direction Method of Multipliers) algorithm, and the residual ε≤1e-4 is used as the convergence criterion. Experiments show that the cost recovery rate of the model in the "high demand-low cost" and "high demand-high cost" scenarios is 93% and 89%, and the proportion of days achieving the renewable energy consumption target is increased to 82.74%±1.24%, an increase of 192% over the benchmark, which verifies the effectiveness of the mechanism.