As multiple types of loads and distributed resources such as electric vehicles, energy storage, and new energy power generation are widely accessible in the park, the characteristics of “source-grid-load-storage” interaction and multi-energy complementary coupling are increasingly apparent, and the traditional scheduling cannot meet the needs of accurate scheduling and real-time analysis of operation situation. The digital twin technology will greatly improve the intelligent level of the park’s comprehensive energy system by carrying out data acquisition and transmission, data processing and feature mining, operation scheduling and high-level decision-making for the park’s energy and power, and realizing the holographic mapping of the park’s comprehensive energy in the virtual information world. This paper firstly analyzes the principle and process of “source-grid-load-storage” operation situation deduction and multi-level scheduling of the park’s integrated energy system based on the digital twin technology, and secondly systematically expounds the key technologies of “source-grid-load-storage” collaborative regulation in the park from the aspects of aggregation modelling, rolling prediction, situation deduction, and optimal scheduling. Also, it points out the direction for the practical application of digital twin technology of integrated energy systems in the park. Finally, based on a comprehensive smart energy project on a campus in Shanghai, this article conducts an energy management experiment on a certain day, which proved that the system can maximize the consumption of new energy, make the heating mode more energy-saving and environmentally friendly, and effectively achieve peak shaving, valley filling, energy conservation and emission reduction.