To fully account for the dynamic coupling of heterogeneous energy and multi-entity interactive collaboration in hydrogen-integrated energy systems

this study proposes a multi-entity collaborative optimization scheduling method for multi-energy carriers based on multi-agent optimal compromise reinforcement learning. First

an optimal scheduling model for multi-entity hydrogen-integrated energy systems is established to minimize operational costs. Subsequently

a multi-agent model is constructed with individual agents as entities

and the optimization scheduling model is solved under a multi-agent reinforcement learning (MARL) framework. Addressing the issues of insufficient collaboration

action conflicts

and low optimization efficiency in traditional multi-agent proximal policy optimization algorithms caused by independent states and actions

this method enhances state-space collaboration by incorporating adjacent agents' action information. Furthermore

a multi-scheme evaluation mechanism based on the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is developed to screen optimal action combinations through compromise solutions

thereby forming a multi-agent optimal compromise reinforcement learning algorithm that improves agent collaboration and solution efficiency. Case study results verify the superior solution accuracy and optimization performance of the proposed model and algorithm.