Abstract: The sources of energy supply and demand within heterogeneous energy systems are complex, so the conventional methods cannot build a generic model connecting different energy networks. To this end, a novel modeling approach for heterogeneous energy systems is proposed based on the modularity idea, and a corresponding control strategy is designed for them. Among them, a heterogeneous energy system model is established with a hybrid logic dynamic system representation considering photovoltaic power generation, cogeneration and energy storage technologies; a model predictive control strategy is constructed for optimizing investment, operation and maintenance cost. In the proposed strategy, energy trading, system nonlinearity or hybrid system behavior between heterogeneous energy systems and upstream energy systems are considered. The impacts of different user behaviors, changing climate conditions and prediction errors on the energy saving potential are analyzed by comparing it with traditional control strategies. The results of the algorithm analysis show that the proposed approach can effectively utilize variable clean energy and energy storage technologies to optimize the operation of heterogeneous energy systems, with annual cost savings of 3% to 6%.