To support the development of a new energy system and the achievement of China’s carbon peaking and carbon neutrality goals, smart energy management urgently needs to overcome key bottlenecks in multi-energy coordination, system regulation and control, and user response. This research area focuses on sensing, forecasting, and intelligent control in integrated generation–grid–load–storage systems. Combining artificial intelligence, big data, digital twins, behavioral science, and systems engineering, it seeks to move beyond the static paradigm of traditional energy management and establish a smart energy management system with a closed loop of “data–models–strategies–decisions,” supporting low-carbon transition and intelligent upgrading at the park, city, and regional scales.

Research includes, but is not limited to: (1) modeling and coordinated optimization of multi-energy systems to support the integrated operation and flexible scheduling of multiple energy carriers, including electricity, heating, cooling, and gas, under different scenarios, covering applications such as urban energy systems, park microgrids, and virtual power plants; (2) load forecasting and user response modeling based on machine learning and reinforcement learning to enable intelligent identification and flexible control of end-use energy consumption behavior; (3) multi-scale user behavior modeling and research on mechanisms underlying energy-saving behavior, exploring the dynamic evolution of behavioral characteristics and strategies for external intervention; (4) cross-domain modeling of coupled energy–behavior systems and development of simulation platforms integrating physical and virtual systems, providing dynamic assessment tools for energy-saving policies and system regulation and control; (5) optimization of regulation and control strategies incorporating market mechanisms and policy scenarios to support multidimensional responses and system scheduling within electricity spot markets and carbon markets.

Associated Faculty: Meng Shen and Te Han.