In order to reduce the carbon emissions of park integrated energy systems and to explore the demand response potential of different types of loads，a low-carbon economic dispatch method for multi-park integrated energy systems considering electricity-carbon coupling and differentiated demand response is proposed.Considering the electricity-carbon coupling among different parks and the characteristics of the external carbon trading market，a multi-park carbon trading model with a stepped external carbon trading mechanism is constructed；in view of the differences in energy demands and demand response capabilities of users in different parks，a differentiated demand response model including flexible and adjustable electrical and thermal loads is established.With the joint park operator as the leader，and user load aggregators and energy storage providers as followers，a multi-agent master-slave game mechanism and the decision-making models of each agent are established，and an improved particle swarm parameter optimization algorithm is used to solve the master-slave game model.Based on the example simulation of a multi-park integrated energy system in Hunan Province，the results show that compared with traditional optimal dispatch methods，the proposed optimization strategy reduces the total carbon emissions of the system by 9.31% and improves the total revenue by 21.32%.