Decision Making in Problem of Irrigation Based on Z-Information

In this paper an intelligent irrigation system is considered. Irrigation regimes for agricultural crops provide the determination of scientifically based norms of irrigation, depending on the soil, physiological, climatic and technical characteristics of the irrigated array. In modern irrigation systems, the parameters of irrigation regimes are calculated similarly to traditional methods. To solve the problem of efficient irrigation, it is necessary to take into account the weak formalizability, which makes it difficult to use accurate mathematical models to describe the control system. The solution to this problem would be partially possible if there was a large number of qualified experts who, taking into account the linguistic nature of the variables, are able to determine the optimal irrigation rates, using their experience without constructing any functional dependencies. Due to the absence of such reference experts in intelligent irrigation planning systems, mathematical models based on Soft Computing should come forward. Our model is based on Z-information, which has such an important qualitative attribute as reliability. The concept of Z-number relates to the issue of reliability of information, especially in the realms of decision analysis. Z-number valuations take into account the uncertаinty of the experts` opinion in the estimаtion of the options. We apply an interpolation approach for Z-rules for the solution of problem of approximate reasoning. A practical example of evaporation is given to show the applicability of approach in irrigation process.