СОВЕРШЕНСТВОВАНИЕ ДОЖДЕВАЛЬНОЙ ТЕХНИКИ НА ОСНОВЕ МЕТОДОВ НЕЙРОУПРАВЛЕНИЯ

Abstract

Purpose: of ways to improve operational efficiency of center-pivot sprinkling machines on models of neural network control. Materials and methods. Research and field data collection were carried out in Engels district Saratov region. The object of research is center-pivot sprinkling machines. Most center-pivot sprinklers use ON/OFF controllers. These controllers cannot provide optimal results for different time delays, different system parameters and external influences. Modern methods of intelligent data analysis are applied, namely, methods of neurocontrol of dynamic objects. Results. As a result of research, it was found that traditional approaches based only on physical modeling of technical processes and connections often complicate the search for effective solutions. Intelligent control of irrigation technology is essential for maximum efficiency and productivity. An approach based on intelligent data analyses model is proposed, namely, the control of a sprinkler machine using a neural controller. Conclusions. The algorithm for neural control by speed (neural controller) is proposed, which minimizes the deviation of the actual values of irrigation rates from the specified ones, arising under the influence of operational, stochastic factors, up to 1–3 %, and methods of its implementation into control systems to improve the efficiency of control of existing equipment and in the development of modern sprinkler machines. The proposed controller based on an artificial neural network is created using MATLAB. The main parameter of modeling is speed. Improving sprinkler equipment based on intelligent control methods is a new trend in increasing the efficiency of Russian sprinkler equipment.