Hybrid knowledge modeling for an intelligent greenhouse

The quality of control provided by greenhouse control systems can be improved by applying model based and intelligent control. To this aim a good model of a greenhouse is needed. For a large variety of industrial or recreational greenhouses the derivation of an analytical model is not feasible, due to the large amount of identification data and expertise needed. A black-box model of the whole system is neither feasible, as it would require a huge number of teaching samples. The only way to tackle this problem is the decomposition of the greenhouse system using hybrid modularized models, where each module represents a relatively loosely coupled component of the system. This paper discusses such decomposition and a model under development.