REDES BAYESIANAS APLICADAS A UN MODELO CFD DEL ENTORNO DE UN CULTIVO EN INVERNADERO BAYESIAN NETWORKS APPLIED IN A CFD MODEL OF THE CROP IN GREENHOUSE

The advances in computer systems and resources make it possible to develop models to simulate the behavior of the fluids in greenhouses. However, the prediction of the gradients of mass and energy in the greenhouses with the crop and natural ventilation is difficult due to the stochastic nature of the wind and the relationships of dependence among temperature, CO 2 and relative humidity. There are heuristic techniques, such as the Bayesian Networks, which help to know the relationships among the variables that cannot be determined with statistical tools. The objective of the present study was to determine temperature, CO 2 concentration and relative humidity with respect to crop height, in a greenhouse with natural ventilation, by means of Bayesian Networks applied to a model of Computational Fluid Dynamic. The Bayesian Network made it possible to determine the spaces of the greenhouse with adverse environmental conditions for the crop development and the most probable climatic states, from the relationships among the variables studied.