Abstract: To address the issues of data privacy protection requirements, high computational overhead in distributed coordination, and operational uncertainty caused by electricity price fluctuations in multi-agent coordinated scheduling of regional integrated energy systems, a learning-enhanced federated distributed coordinated dispatch method is proposed. First, a federated distributed coordination optimization model for regional integrated energy systems is established, and federated coordination variables are introduced to represent the coupling constraints among subsystems, enabling each energy subsystem to perform local optimization while keeping operational data within the local domain. Then, the update process of the federated coordination variables is modeled as a Markov decision process, and reinforcement learning is introduced to enhance the coordination update strategy, by learning the update direction and magnitude of the coordination variables, the number of coordination iterations is reduced, thereby improving the efficiency of distributed solution. Furthermore, a rolling optimization framework is incorporated, and the fuzzy C-means method is used to perform online correction of electricity price to improve the adaptability of the dispatch strategy under electricity price uncertainty. Case studies show that the proposed method can achieve coordinated dispatch of multiple energy systems under typical operating scenarios. Compared with soft Actor-Critic (SAC) and deep deterministic policy gradient (DDPG) methods, the operating cost is reduced by approximately 0.42% and 0.49%, respectively, and the gap with the centralized optimization result is less than 0.5%. Meanwhile, reinforcement learning enhancement of the coordination update strategy reduces the computation time by approximately 71.6%. Under different electricity price disturbance scenarios, the system operating cost remains close to the ideal cost level, demonstrating good stability and robustness. The results indicate that the proposed method improves the efficiency of federated distributed coordination while ensuring local data storage and enhances the adaptability of integrated energy system scheduling under uncertain operating conditions.