Формализация параметров модуля фертигации в системах капельного орошения

Abstract

Purpose: to formalize the parameters on which the choice of optimal fertilizer application rates depends with fuzzy sets for constructing a fuzzy model and a control system for the fertigation module. Materials and methods. The use of Soft Computing methods is required by the need to represent the input variables values in the form of fuzzy sets, with further structural and parametric identification of a fuzzy model for controlling the concentration of nutrients in the solution. This will enable to compile fuzzy knowledge bases in the form of “IF – THEN” rules and, using fuzzy logic methods, to determine the well-defined values of output variables: the necessary application rates of nitrogen, phosphorus and potassium with irrigation water. Results. As input variables, the parameters on which the choice of optimal fertilizer rates depends are determined. At the output of the model, three variables determining the norms of nitrogen, phosphorus and potassium of the nutrient solution are used. For each of the selected input and output variables, the measurement ranges and the linguistic term values are selected and membership functions are constructed. Conclusions. The conducted formalization of input and output values will make it possible to build the structure of the fertigation module control model, to form fuzzy rule bases and to implement fuzzy logical inference according to the Mamdani algorithm. This will allow defuzzification of the output variables using the center of gravity method in order to obtain a clear numerical value at the output. It is also possible to carry out parametric identification of the obtained model from the training sample by adjusting the membership functions parameters of the. Thanks to the use of IoT technologies, it will be possible to carry out external control and, if necessary, manual control of the fertigation system.