DISINFECTION AND USAGE OF ANTHROPOGENIC WASTE IN RESOURCE-SAVING CONSTRUCTION TECHNOLOGIES

Relevance. The environmental protection problem is one of the most important one in the present time. Multilateral pollution of all geospheres with non-utilized wastes led to a sharp deterioration of ecological systems, to the death of some unique natural complexes, to the reduction and disappearance of populations of certain plant and animal species, to the danger of unpredictable irreversible consequences which can be the result of man's technogenic activity. That is why the utilization of industrial wastes of industries and their use in resource-saving technologies, of course, are very relevant. The aim of the work is to show the possibility of utilization, control and use of waste products, in particular fluorine-anhydrite in resource-saving construction technologies. Methods: chemical waste neutralization, disintegration of the obtained raw materials and production of bulk clinker material from it, methods of its control and application in various formulations in construction technologies; laser scanning of the surface of the received bulk mixture; approximation of the surface model using radial basis function neural networks; modeling of a non-linear surface function, by the Kolmogorov theorem, using the superposition of radial basis functions; calculation of the volume of a substance limited by the obtained function using the cubature function by the Gauss-Kronrod method and the Monte Carlo method. Results. The paper considers the technology of calcium sulfate neutralization and technological scheme of its disintegration and use in drilling fluids and introduces the system for controlling the volume of a disintegrated substance, which is necessary to account its quantity and dosing when used in formulations of various technologies. It is shown that when controlling and estimating the amount of a disintegrated substance, the best way is to approximate the nonlinear function, since it is possible to control the error by the number of partitions of the function or the number of experiments, and also the model of the surface of the substance is taken into account. The usage of the radial basis function neural network is advisable for obtaining a model of the surface of bulk materials in order to improve the accuracy of measuring volumes in silo farms and warehouses of manufacturing enterprises. Methods of modeling surface functions and measuring their volume by quadrature or Monte Carlo methods are the most effective ones. Usage of the Gauss-Kronrod method in this case is preferable.