SIMULATION AND DIGITAL OPTIMIZATION OF PROCESSES IN AGRICULTURE

It is proposed to form mathematical models of processes in agriculture on the basis of statistical methods for planning multifactorial experiments, approximating the results of research by regression equations. The presence of the relationship between the factors characteristic of most agricultural tasks is shown. Therefore, it is established to determine the stationary values of the optimization parameter function together with additional conditions in the form of empirical connection equations, using the method of undetermined Lagrange multipliers. For planning experiments, a It is proposed to form mathematical models of processes in agriculture on the basis of statistical methods for planning multifactorial experiments, approximating the results of research by regression equations. The presence of the relationship between the factors characteristic of most agricultural tasks is shown. Therefore, it is established to determine the stationary values of the optimization parameter function together with additional conditions in the form of empirical connection equations, using the method of undetermined Lagrange multipliers. For planning experiments, a central orthogonal compositional design is recommended, which allows you to independently evaluate the coefficients of the regression equations and their variances, and exclude insignificant coefficients from the equations without recalculating the remaining coefficients. Reducing computational work and reducing the risk of random errors when using the method is achieved by coding factors with indicators that do not have dimensions. The practical expediency of applying the method is determined on the example of experimental data on the integrated optimization of soil tillage and fertilizer factors in the technology of winter wheat production. The influence of factors on the yield of crop grain, the protein content in it and the cost of production were studied in 2016-2020. in the field experiment of the Federal Agricultural Kursk Research Center (Medvensky district, Kursk region) on the slope of the northern exposure in a grain-grass- rowed crop rotation. Using the method of indefinite Lagrange multipliers, we determined the optimal dose of fertilizer N19P39K39 and the preference for the combined method of tillage, which correspond to the minimum cost of winter wheat grain 5.38 thousand rubles/t with a yield of 3.38 t/ha. Received grain of the 3rd class with a protein content of 12%.