An intermediate model to predict thermal comfort and air quality in a building.

In this paper, a zonal model used to predict the air movement, temperature distribution and airquality in a room is presented. It is based on a rough partitioning of the room: it is anintermediate approach between one-node models (that consider an homogeneous temperaturein each room, and, for that reason, do not permit to predict the thermal comfort in a room) andCFD models (that require great amount of simulation time). Where plumes, jets or thermallayers occur, air flow is described by empirical laws. In low velocity domains, flow rates arecalculated in respect to the pressure distribution. This air flow model is coupled with abuilding envelope model including the calculation of radiant and conductive exchanges. Thecomplete model is implemented in an object oriented environment, SPARK, in whichmodelling a room consists in connecting the different pre-design elementary models. Hence,the way of modelling is very modular, so that the zonal model can now be applied to a verylarge field of configurations. The strict syntax of SPARK permits having the simulationautomatically generated. Now, imagining the creation of a very flexible tool that allows torepresent a whole building is realistic.Results of simulations in a 3D-room will be given and it will be shown that this model yieldsrather accurate results even with a rough partitioning.