The article discusses the extraction of solid mineral resources (SMR) from the bottom, such as ferromanganese nodules (FMN) and cobalt-manganese crusts (CMC), occurring in both offshore and deep regions of the World Ocean. For separating and lifting the water surface is used mining complex, which includes a hydraulically dragheads and intermediate capsule with atmospheric pressure introduced into the of extraction process in order to increase energy efficiency through the organization of a two-stage hydroascent, the first stage is carried out by external hydrostatic pressure. To determine the optimum depth of immersion of the intermediate capsule composed mathematical model. It is based on the sustainable hydroascent provided at the first stage transport (from the bottom to capsule), and reducing power consumption slurry pump performing the second stage transport (from the capsule to the sea surface). In determining the rational immersion depth were taken into account design features of slurry pipeline, mine geological parameters, the parameters of extracted minerals, performance of dragheads, the slurry flow parameters, such as concentration, consistency, density. Presents an example calculation and determined the optimal capsule insertion depth for offshore geological environments.

The paper presents a methodology for training neural networks for vision tasks on synthe-sized data on the example of steel defect recognition in automated production control systems. The article describes the process of dataset procedural generation of steel slab defects with a symmetrical distribution. The results of training two neural networks Unet and Xception on a generated data grid and testing them on real data are presented. The performance of these neural networks was assessed using real data from the Severstal: Steel Defect Detection set. In both cases, the neural networks showed good results in the classification and segmentation of surface defects of steel workpieces in the image. Dice score on synthetic data reaches 0.62, and accuracy—0.81.

Currently in the Russian Federation there is a large number of methods for determining the economic damage from environmental pollution. They make it possible to assess the quantitative damage caused by the industry to the environment and consider the environmental pollution factor when calculating the economic performance of enterprises. In general, all methods can be reduced to three main types, whose main advantages and disadvantages, as well as applicability for adjusting the performance of enterprises, and inclusion in environmental performance assessment are discussed in this article. The authors proposed a step-by-step procedure for including the factor of negative impact on the environment in the evaluation of companies’ performance. First, a preliminary consolidated assessment is made by the method of specific damage. Secondly, the economic effect is calculated by the method of generalized indirect estimates. If necessary, clarification of calculations by the method of direct calculation is carried out. When calculating production efficiency indicators, considering the nature protection activities of the enterprise, a method of accounting for the prevented damage is proposed.