Faced with the dual challenges of energy transition and renewable energy consumption, achieving low-carbon, efficient, and clean energy is the mainstream trend for future energy use. Based on the current park integrated energy system (PIES) model and multi-level thermal energy coupling, this paper proposes a multi-park integrated energy system (MPIES) collaborative scheduling model. In MPIES, each sub-park cooperates with each other to interact between electrical energy and multi-level thermal energy. To protect the privacy of each sub-zone, the ADMM algorithm is employed to decompose the optimization problem into two sub-problems: maximizing the joint operation total benefit of MPIES and maximizing the transaction benefit of PIES. The joint operation benefits after energy sharing are then allocated reasonably according to the Shapley value method based on the marginal contribution of each PIES to MPIES. Simulation results indicate that the operational costs of each PIES are reduced compared to independent operation, and this model is beneficial for enhancing the willingness of MPIES members to participate in energy sharing.