METHODOLOGICAL MANAGEMENT TOOLS OF IRRIGATION WITH ARTIFICIAL INTELLIGENCE

Abstract

Purpose: to develop the concept of irrigation management systems using integrated artificial intelligence technologies. Materials and methods. The working hypothesis of the research was the assumption of the priority of the cluster application of artificial intelligence technologies in solving urgent problems of irrigation management. At the same time, the basis of the information-computing unit of the irrigation control system should be deterministic algorithms using proven solutions. Results. Within the framework of the proposed concept, a neural network can be used as the basis for solving a problem, or it can only be used to adapt the parameters of the models used. The impact points of traditional, deterministic - analytical irrigation management systems and artificial intelligence technologies are determined by the need to adapt parameters, fine-tune the coefficients of the methodological tools used. At the same time, the scope of application of artificial intelligence in solving problems of irrigation management is quite wide. Artificial intelligence technologies can be used to solve the problems of irrigation planning and operational management, image recognition in satellite monitoring of soil moisture, interpolation of soil moisture data by area, forecasting the soil moisture profile. An algorithm using artificial intelligence technologies to improve the reliability of forecasting the total water consumption of irrigated crops in the regional and landscape aspect is developed by research. The proposed algorithm allows the system to self-learn and refine the regional values of bioclimatic coefficients, with respect to cumulative influence of local factors. Conclusions: a concept has been developed and new scientific approaches have been proposed to effectively solve the problem of adapting the parameters of methodological irrigation management tools at the level of an irrigated field, and even taking into account intra-field variability.

Keywords

• irrigation management
• artificial intelligence
• areas of application
• integration methods
• neural network