The frequency of maintenance of electromechanical equipment depends on the plant, which uses and runs this equipment. Very often the maintenance frequency is poorly correlated with the actual state of the electromechanical equipment. Furthermore, traditional methods of diagnosis sometimes cannot work without stopping the process (for example, for equipment located in hard to reach places) and so the maintenance costs are increased. This problem can be solved using the indirect methods of diagnosing of the electromechanical equipment. The indirect methods often use the parameters in the real time and seldom use the parameters of traditional diagnostic methods for determination of the resource of electromechanical equipment. This article is dedicated to developing the structure of a special automated control system. This system must use the big flow of the information about the direct and indirect parameters of the equipment state from plants from different areas of industry and factories which produce the electromechanical equipment.

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.

DEM parameters calibration is the most important step in preparing a DEM model. At the same time, the lack of a universal approach to DEM parameters calibration complicates this process. The paper presents the author’s approach to creating a universal calibration approach based on the physical meaning of the friction coefficients and conducting symmetrical experiments at full scale and in a simulation, as well as the implementation of the approach in the form of a physical test rig. Several experiments were carried out to determine the DEM parameters of six material–boundary pairs. The resulting parameters were adjusted using a refinement experiment. The results confirmed the adequacy of the developed approach, as well as its applicability in various conditions. The limitations of both the approach itself and its specific implementation in the form of a test rig were identified.