Calibration and validation of a greenhouse climate control model

While thermal structures such as buildings or greenhouses are complex to model under transient conditions because of coupled conductive, convective and radiative heat exchanges between their different parts, systems theory allows a simple analysis of the input/output behaviour. In particular, reduced dynamic models can be derived from in-situ measurements on these systems. This paper describes an application of the systems approach to a greenhouse climate model which incorporates crop transpiration, the effects of heating, natural ventilation and evaporative cooling. It represents quite well the complex coupling mechanisms between heat and mass exchanges and allows computation of the inside climate from the state of the actuators such as vent opening, heating intensity and fog system, and the outside climate. System parameters were on-line estimated from sequences of input-output data recorded over the crop production period from December to July. Good agreement between measured and computed values of air temperature, air humidity and crop transpiration was observed. Applications of this approach to on-line climate control are discussed.