传统上，在建筑能耗建模（BEM）项目中，占用率输入是确定性的，对真实场景的指示较少，这导致了建筑中模拟和实际能耗之间的差异。本文提出了一种基于agent的居住模拟器，该模拟器将每个居住者建模为一个agent，具有指定的运动事件和空间使用统计信息。为了减少数据输入量，模拟器允许用户将行为类似的乘客分组为乘客类型，将功能类似的空间分组为空间类型。它是一个基于web的应用程序，具有友好的图形用户界面、云计算和数据存储。 通过一个案例研究，展示了占用模拟器的使用及其与EnergyPlus和obFMU的集成。它首先显示所需的数据输入和占用模拟器的结果。然后，在EnergyPlus和obFMU模拟中使用生成的居住者时间表来评估居住者行为对建筑能源性能的影响。模拟结果表明，占用模拟器可以捕捉空间占用行为的多样性，而不是静态的每周剖面，并可以生成真实的占用时间表，以支持建筑性能模拟。引用: 2016年年度会议，密苏里州圣路易斯，会议论文

Traditionally, in building energy modeling (BEM) programs, occupancy inputs are deterministic and less indicative of real world scenarios, contributing to discrepancies between simulated and actual energy use in buildings. This paper presents an agent-based occupancy simulator, which models each occupant as an agent with specified movement events and statistics of space uses. To reduce the amount of data inputs, the simulator allows users to group occupants with similar behaviors as an occupant type, and spaces with similar function as a space type. It is a web-based application with friendly graphical user interface, cloud computing, and data storage. A case study is presented to demonstrate the usage of the occupancy simulator and its integration with EnergyPlus and obFMU. It first shows the required data inputs and the results from the occupancy simulator. Then, the generated occupant schedules are used in the EnergyPlus and obFMU simulation to evaluate the impacts of occupant behavior on building energy performance. The simulation results indicate that the occupancy simulator can capture the diversity of space's occupancy behavior rather than the static weekly profiles, and can generate realistic occupancy schedules to support building performance simulation.