Privacy-Aware Distributed Market Clearing for Multi-Regional Power Systems with Hybrid Energy Storage Using an Adaptive ADMM Approach

Multi-regional electricity markets increasingly struggle to balance data privacy requirements with the computational burden of centralized clearing. To address this issue, this study proposes a distributed joint-clearing framework based on the Alternating Direction Method of Multipliers (ADMM) to co-optimize pumped storage hydropower (PSH) and battery energy storage systems (BESS) across energy, frequency regulation, and reserve markets. A mixed-integer programming model is formulated to maximize social welfare, explicitly capturing the time-coupled, energy-oriented characteristics of PSH and the fast-response, power-oriented capabilities of BESS. The global problem is decomposed into regional subproblems that can be solved in parallel. An adaptive penalty parameter strategy is further introduced to dynamically balance primal and dual residuals, thereby improving convergence and robustness in the mixed-integer setting. To address the limited economic interpretability of dual variables in mixed-integer programming (MIP) models, an approximate marginal pricing mechanism based on subproblem sensitivity analysis is proposed. A two-region, 24 h case study shows that the proposed method converges in around 65 iterations and achieves a social welfare outcome within 0.61% of the centralized optimum. By minimizing information exchange, the framework offers a scalable and privacy-aware solution for future multi-regional market operations involving heterogeneous energy storage resources.