Fruit growing
and viticulture of South Russia
The article provides a comparative analysis of existing methods and models for designing healthy food products with a given composition: experimental statistical modeling, linear programming, object-oriented approach, fuzzy logic models, neural networks. For each technique, their key features, as well as shortcomings and boundaries of effective application, are identified. The choice of a basic technique for subsequent practical application is justified based on the results of the analysis linear programming: the simplex method and its modifications. Further, the paper presents a description of mathematical models and algorithms for solving three problems that implement the design of the component composition of functional foods. For each technique, sets of input and output data, constraints, objective functions, as well as generalized sequences of actions in calculations in the form of flowcharts of algorithms are defined. The component composition optimization model will allow you to design the selected healthy food products based on its basic recipe, optimizing it according to a given criterion: cost, energy value, degree of content of a certain chemical component, etc. The methodology for optimizing nutritionally balanced food recipes with desired properties and composition is based on the use of norms of physiological needs for nutrients for different population groups. It allows to evaluate a specific formulation of a healthy nutrition product for its compliance with these standards for a given population group. A distinctive feature of the proposed method of system modeling of multicomponent food products of functional nutrition is the possibility of designing a recipe for given ratios of components and chemicals. The described models will further be the basis for the development of digital tools for designing healthy food products with a given composition.
At the present stage, with an increase in the volume of consumption of grape-growing products, it is necessary to carry out a monitoring forecast of the possibility of its production for each individual variety or scion-rootstock combination, depending on edaphoclimatic conditions and cultivation technology. This is possible only if predictive models of the behavior of the grape variety or its scion-rootstock combination are developed in the grafted culture in various ecoagrobiocenoses. The purpose of the study was to consider methodological approaches to the creation of mathematical models for predicting the behavior of an individual variety or groups of grape varieties, depending on the abiotic and agrotechnological characteristics of cultivation. To achieve this goal, a previously created database was used, obtained during an experiment conducted on the basis of uterine plantations and an open grape school of the Institute "Agrotechnological Academy" of the V.I. Vernadsky Crimean Federal University, collected in the period from 2018 to 2021 and subjected to multidimensional regression analysis using the developed program. The total number of items included in the database is 1,860. (31 parameters). The research proved the possibility of developing regression models for predicting productivity using nonparametric digital introduction of varieties, as well as environmental factors. It is established that regression models characterizing the quality of the vine, taking into account varietal characteristics and weather conditions, can vary depending on a particular variety. Thus, a similar model for the Cabernet Sauvignon variety is fundamental with a multiple correlation coefficient of R = 0.9866, and for the Syrah variety it is logarithmic at R= 1.0000. Promising possibilities and ways of developing digital (mathematical) models characterizing individual varieties or groups of varieties by origin according to their productivity, depending on edaphoclimatic conditions, production technology, as well as quality parameters of manufactured products are considered.
The necessity of developing multifactorial models for managing the sustainability of reproductive processes in perennial agrocenoses with the participation of orchard crops and grapes is substantiated. The factors influencing the sustainability and efficiency of the production of fruit products and grapes in the Krasnodar region have been identified. Cognitive models for managing the sustainability of reproductive processes in perennial agrocenoses based on technological processes of elements of the fruit and grape-growing management system have been developed. Models for managing the sustainability of reproductive processes are based on cognitive methods of analysis and digital technologies, taking into account the identified systemic relationships in reproductive processes in functional areas of impact according to criteria of efficiency; biologization; environmental management. Cognitive models have made it possible to identify multifunctional relationships in reproductive processes in functional areas of influence, the optimization of which is the basis for the development of digital technologies. The results of cognitive modeling will be the basis for solving applied problems to increase the productivity of fruit agrocenoses, reduce the level of chemical and anthropogenic load on their elements and optimize the costs of production of industrial products. normative parametric model of the stability of reproductive processes in perennial agrocenoses with the participation of orchard crops and grapes has been developed. The tools for managing biotechnological processes in the elements of agrocenosis are proposed.
The article presents the results of agrobiological and physiological-biochemical studies to ensure the stability of grape plantations in unstable weather conditions and anthropogenic intensification of production. Studies in the agroecological conditions of the Krasnodar region on the southern chernozems revealed the most productive norm of loading of grape bushes with shoots for the technical varieties Kurchanskiy, Dmitriy, Vladimir, Granatovyi, Antaris and Alkor, on leached chernozems for the table seedless variety Centennial Seedless. According to the results of electrophoretic separation of peroxidases in polyacrylamide gel and an increased content of ascorbic acid, increased frost resistance was established in the varieties Kristall, Vostorg, Krasnostop AZOS. The Vostorg and Zarif varieties proved to be more adaptive to the stresses of the summer period in terms of leaf hydration, bound water content, the ratio of chlorophylls and carotenoids, the amount of proline and water-soluble sugars, membrane permeability. The role of physiologically active substances in increasing the adaptive potential of grape plants has been established. The treatment of Merlot vines with methyl jasmonate and epibrassinolide during deep dormancy period, as well as proline, salicylic acid and epibrassinolide during induced dormancy period had a positive effect on the condition of the vine: there was a decrease in electrolyte yield. Sustainable grape cultivation is achieved with the systemic use of the Bioconcentrate-Z effluent by the non-root method. Against the background of nonroot treatments of Merlot and Cabernet Sauvignon varieties, a higher level of plant water supply, an increase in potassium content, green pigments, organic acids and grape yields were established. Physiologically active compounds and the preparation Turinbash have a positive effect in intensive own-root plantations of native and promising introduced grape varieties against the background of phylloxera infection in the southern part of Dagestan. The created yield forecasting method based on the NDVI of the grape plant and the actual data of multispectral images of the phenophases of flowering and growth allows to increase the accuracy of the calculated yield capacity to the actual one with a deviation from 0 to 0.8 kg per bush, the accuracy of the forecast increases by 2 times by the end of the growth phase and by the beginning of the maturation phase.