FUZZY CONTROL IN AGRICULTURE: SIMULATION SOFTWARE

ABSTRACT In the present document the software called “System of Fuzzy Control (SCD)” is described as the platform that allows to simulate some types of control systems based in fuzzy logic. This software allows the user to create and modify the elements that make a fuzzy knowledge-based controller in an intuitive and clear way. The knowledge base of SCD is formed by a set of rules of the conditional type being possible with the choice of connective operators and negation between antecedents, which are fuzzy comparisons. The program is very flexible, visualizing in a graphical form the char-acteristics of each fuzzy variable next to the simulation data. This software can be applied to any scope of the industrial simulation. In our case we have applied it to the climatic control on industrial greenhouses. 1. INTRODUCTION Ever since Mamdani [9] showed the first application of the fuzzy logic to the control of a particular process, an extensive range of applications, from household-electric to robots and systems of industrial control [6,7,11] have been developed under the premises of the fuzzy control. At the present time we can find a great variety of soft-ware [1,12], which is able to control processes being based on its structu ral approach to the form of the human thought and to the natural language of an expert opera-tor. The software that we present here tries to be a contr i-bution with substantial improvements with respect to other contributions [3]. It covers the purpose of being able to create, in the most possible general form, fuzzy knowledge-based controllers. In addition, the application to the fuzzy logic to the agricultural production systems [2, 4, 10] is an appropriate approach in the search of solutions because it would serve like a tool for “storing” expert knowledge of aid to the decision. For this reason we have applied the program to the crop of peppers in a greenhouse in Almeria (Spanish Southeas tern). In the first place we will expose a description of software by listing its main characteristics. Next we will show the phases of creation of the system of climatic control in the application. Once the system is created we will come to simulate, in order to do so, we will see the different phases that the application makes to obtain the control actions. Finally, we present some conclusions, future works and ref erences.