The conventional control methods of variable air volume (VAV) air conditioning systems usually assume that the indoor air is well mixed, and consider each building zone as one node with homogeneous temperature distribution. The average temperature is subsequently used as the controlled parameter in the VAV cascade control process, which might cause uneven temperature distribution and unsatisfactory thermal comfort. This paper presents a coupled simulation of computational fluid dynamics (CFD) and building energy simulation (BES) for the VAV system in an office building located in Shanghai for the purpose of simulating the building, the VAV control system, and indoor thermal environment simultaneously. An external interface is developed to integrate the CFD and BES models based on quasi-dynamic coupling approach. Based upon the developed co-simulation platform, the novel VAV control method is further proposed by fusing information from multiple sensors. By adding two temperature sensors to constrain the thermal comfort of the occupied zone, the supply air temperature setpoint of the VAV terminal unit can be reset in real time. The novel control method is embedded into the co-simulation platform and compared with the conventional VAV control approach. The results illustrate that the temperature distribution under the proposed method is more uniform. At most times of the typical test day, the air diffusion performance indexes (ADPIs) for the proposed method are above 80%, while the ADPIs for the conventional control method are between 60% and 80%. Due to multi-sensor information fusion, the proposed VAV control approach has better ability to ensure the indoor thermal comfort.

可变风量（VAV）空调系统的常规控制方法通常假定室内空气混合均匀，并将每个建筑物区域视为温度分布均匀的一个节点。随后，将平均温度用作VAV级联控制过程中的受控参数，这可能会导致温度分布不均和热舒适性不理想。本文针对位于上海的一幢办公楼中的VAV系统，提供了计算流体动力学（CFD）与建筑能耗模拟（BES）的耦合仿真，目的是同时模拟建筑物，VAV控制系统和室内热环境。开发了一个外部接口以基于准动态耦合方法集成CFD和BES模型。基于已开发的协同仿真平台，通过融合来自多个传感器的信息，进一步提出了新颖的VAV控制方法。通过添加两个温度传感器来限制占用区域的热舒适度，可以实时重置VAV终端单元的送风温度设定点。该新颖的控制方法被嵌入到协同仿真平台中，并与传统的VAV控制方法进行了比较。结果表明，所提方法的温度分布更为均匀。在典型测试日的大多数时间，建议方法的空气扩散性能指标（ADPI）高于80％，而常规控制方法的ADPI则在60％至80％之间。由于多传感器信息融合，