Tripartite Hybrid Game-Theoretic Optimization for Integrated Vehicle-Station-Grid System With Charging Station Heterogeneity

Abstract

This paper proposes a tripartite hybrid game-theoretic pricing model for the vehicle-station-grid framework that accounts for the interest demands of all stakeholders. The model aims to balance interests among different stakeholders while guiding charging behavior of electric vehicle (EV) users within the context of grid peak shaving and valley filling. To this end, a tripartite hybrid game-theoretic framework is established, and the improved Discrete Momentum-Annealed Dynamic Evolutionary (DMADE) algorithm is employed to achieve equilibrium decisions among stakeholders. Furthermore, a competitive bidding model incorporating charging station heterogeneity is developed, which integrates factors such as charging speed to reflect their impact on user selection and bidding behaviors. This design not only satisfies grid requirements for peak shaving and valley filling but also effectively enhances market vitality and expands user choice diversity. Building upon this foundation, a bounded rationality evolutionary game model based on the Momentum-Annealed Dynamic Logit (MADL) mechanism is constructed to address user decision-making with differentiated demands under rapidly fluctuating environments. The proposed methodology integrates utility smoothing, temperature annealing, and momentum mechanisms to prevent estimation deviations in charging strategy parameters when handling non-stationary environmental decisions, thereby ensuring the effectiveness of charging station pricing strategies. The effectiveness of the proposed models and algorithms is verified through numerical case studies. Note to Practitioners—Practical EV charging is affected by multiple factors, including heterogeneous service capabilities of charging stations and bounded rational user responses to charging prices and queue information, rather than fully rational optimization. This paper develops a tripartite decision framework to model bounded rational EV user decisions toward differentiated charging station services and charging demand response in a non-stationary environment, and to obtain stable pricing and bidding equilibrium solutions that balance operator incentives with peak valley regulation benefits. The proposed framework is designed for emerging charging markets and serves as a mechanism analysis and decision support tool. It is applicable to market environments with differentiated charging services, station selection based on price and service related factors, and incentive based demand response. Two open issues remain in EV charging market modeling and are difficult to fully resolve in a single study: complex psychological heterogeneity in user choice and broader coordination among additional participants beyond the three parties considered in this paper. These issues will be further extended in future research.