TABLE OF CONTENTS Greenhouses are mainly used to improve the environmental conditions in which plants are grown. Greenhouse environmental control (GEC) provides a means to further improve these conditions in order to optimise the plant production process. GEC methods aimed at efficiently controlling the greenhouse climate environment and optimising the crop production must take into account the influences of the outside weather, the actuators and the crop, which is achieved by the use of models. The production of most crops is primarily affected by the greenhouse air temperature, humidity and CO 2 concentration.
This section presents an example application of two of the most widely used soft- computing approaches (neural computation and evolutionary computation) for the identification of predictive greenhouse climate models and outside weather models.
Relevant outside weather variables available for this system identification problem are the air temperature ( t w), global solar radiation (sr w), wind speed (ws w) and wind direction (wd w). Greenhouse climate variables available are the air temperature ( t c), relative humidity ( h c) and global solar radiation (sr c). Throughout the rest of this chapter, relative humidity will be simply denoted as humidity.
The data was acquired at a 1 min sampling rate in a plastic covered greenhouse during 12 days. Considering that climate variables change slowly over time, the amount of data is reduced by changing the sampling rate from 1 to 5 min, with the benefits of reducing...
