Building energy consumption constitutes a significant portion of global final energy use, with elevator systems representing a major and growing load in high-rise buildings. Traditional methods for handling elevator regenerative braking energy, such as resistor-based dissipation, are inefficient. Meanwhile, the air conditioning system in the elevator car, a movable flexible load, has not been fully utilized for energy regulation. This study proposes an innovative Digital Twin (DT)-driven Virtual Energy Storage (VES) architecture specifically for the mobile elevator car air conditioning (AC) system. A high-fidelity DT simulation model is developed, integrating real-time elevator dynamics and car thermodynamics to model the thermal inertia of the car as a VES unit. A Model Predictive Control (MPC) coordination strategy, incorporating feedforward prediction of braking energy and feedback regulation, is designed to proactively schedule air conditioning power. Simulation results based on a 30-story high-rise building demonstrate that the proposed system effectively limits DC bus voltage fluctuations to within ±3.4%, reduces grid-side power consumption of the car AC by 17.2%, and lowers the required supercapacitor (SC) capacity by 25% compared to a conventional recovery approach. Over a 24-h operational period, the system achieves an overall electricity savings rate of 69.23% while maintaining the car temperature within the comfort range of 23–27°C with a 96.7% compliance rate. This research provides a novel paradigm for the collaborative optimization of heterogeneous energy units in dynamic built environments, showcasing the significant potential of DT-VES integration for enhancing energy efficiency and system stability.
Tian et al. (Mon,) studied this question.