The proposed comprehensive energy realizes the mutual coupling between different energy sources, and realizes the high-efficiency cascade utilization of energy. Based on this, this paper establishes a day ahead day internal scheduling model of multi-regional virtual power plant comprehensive energy, which takes into account the system network loss. First of all, this paper establishes a system model with the minimum operating cost as the goal under the premise of satisfying the constraints of cogeneration unit and energy storage output by using the day ahead day multi time scale scheduling method. Then, considering that the model is a nonlinear optimization problem, the Hyper-Spherical Search algorithm is used to find the optimal solution of the system. Finally, a park is taken as a simulation object. Simultaneous interpreting and analyzing the output and operation cost under different conditions of the collaborative operation and the separate operation of the virtual power plant, the voltage of the important node or tie line port of the power grid and the heat network is studied. The Hyper-Spherical Search algorithm is compared with the operation results of the traditional particle swarm algorithm, and the feasibility of the proposed model and algorithm is finally verified.