Distributed Collaborative Optimal Scheduling of Regional Integrated Energy Systems with Multi-agent Participation

In response to the global transition toward carbon neutrality, Regional Integrated Energy Systems (RIES) have emerged as a critical technical pathway for multi-energy synergy. This paper provides a comprehensive review of the collaborative optimal scheduling of RIES, structured across four dimensions. First, we establish the physical modeling foundation based on the bus method, characterizing the deep coupling of electricity, heat, and gas flows, with a specific focus on the thermodynamic constraints of Compressed Air Energy Storage (CAES) and the flexibility of demand response. Second, we analyze the distributed hierarchical architectures, evaluating how Analytical Target Cascading (ATC) and game theory resolve the inherent conflict between global efficiency and stakeholder privacy. Third, we review the algorithmic evolution of swarm intelligence—such as PSO, GWO, and BOA—specifically focusing on chaos mapping and adaptive mutation mechanisms used to navigate non-linear constraints. Finally, we examine uncertainty mitigation strategies, synthesizing multi-time-scale coordination and robust optimization as key tools for hedging against source-load fluctuations. By identifying current research gaps, this review outlines a future roadmap toward digital-twin integration and system resilience, providing a strategic reference for modern energy interconnections.