This article deals with solving the urgent scientific problem of the diagnostics of drives of technological robotized workplaces with support of sensors. The dependence of diagnostic parameters on the technical state of drives of automated technological systems, which is of great economic importance for industrial enterprises, is being investigated. Diagnostic models have been developed based on sensory systems to diagnose drive models of technological robotized workplaces. The use of these models may also include monitoring systems in which it is possible to build a system for identifying detected changes. These systems identify many contradictory changes and thereby reduce the false alarm frequency of monitoring sensory systems. Numerous methods for solving technical diagnostics problems are often based on methods based on mathematical models describing work processes, as well as on spectral analysis of measured parameters, such as vibrations, noise, and electric current. A fuzzy inference system for assessing the technical condition, a system for estimating the residual resource of drives, and asystem for calculating diagnostic intervals based on fuzzy knowledge have been developed. Based on the historical trend of the diagnostic parameters, the intelligent diagnostic system determines the current technical condition of the actuator and predicts future technical condition changes, determines the remaining service life and the time intervals for diagnostics. The analysis of the time spent on planned preventive maintenance of technological equipment makes it possible to conclude that, after the modernization of equipment in 2018, the repair time was reduced from 350 h to 260 h per year (26%). Since 2019, there is a tendency to increase repair time by 30 h each year.

This paper proposes a new approach to defect detection system design focused on exact damaged areas demonstrated through visual data containing gear wheel images. The main advantage of the system is the capability to detect a wide range of patterns of defects occurring in datasets. The methodology is built on three processes that combine different approaches from unsupervised and supervised methods. The first step is a search for anomalies, which is performed by defining the correct areas on the controlled object by using the autoencoder approach. As a result, the differences between the original and autoencoder-generated images are obtained. These are divided into clusters using the clustering method (DBSCAN). Based on the clusters, the regions of interest are subsequently defined and classified using the pre-trained Xception network classifier. The main result is a system capable of focusing on exact defect areas using the sequence of unsupervised learning (autoencoder)–unsupervised learning (clustering)–supervised learning (classification) methods (U2S-CNN). The outcome with tested samples was 177 detected regions and 205 occurring damaged areas. There were 108 regions detected correctly, and 69 regions were labeled incorrectly. This paper describes a proof of concept for defect detection by highlighting exact defect areas. It can be thus an alternative to using detectors such as YOLO methods, reconstructors, autoencoders, transformers, etc.

In modern conditions, management technology is becoming the main instrument of competition in many areas of production and business. The article focuses on the examinations of issues related to the possibilities of introducing an integrated risk management system in modern high-tech enterprises. The main scientific research methods are bibliographic analysis, formal logic methods, comparative analysis, methods of analogies. Based on the application of the guidelines of systems’ theory, of the theory of self-organization, the theory of risk management, and the theory of reflexive control the expert technology for risk management in a high-tech industrial enterprise was developed. As a result of the study, it was revealed that the use of technologies such as the ETRM will help to increase the feasibility and effectiveness of implementing complex projects such as fast-response QRM production.

Harmful influence of human factor upon industrial processes and while vehicles running control necessitates to search for ways of its minimizing or total elimination. During the trials of ergatic system “locomotive team – brigade” assessment of informative functioning capacity of machinist is done with experimentally-statistic method. According to the research done the total number of signals that influence on locomotive brigade in process of movement makes approximately 20000 for during one trip, and the number of signals indicating abnormal situation reaches 165 for one trip. In time of train movement the period of signal processing lays at such interval – 0.3c < toc < 11c. It was stated during calculations with given assumptions that informative functioning capacity of a locomotive brigade approaches the threshold values, which constitute 2–4 bps. Received model allows to evaluate parameters of informative functioning capacity on a locomotive brigade in contrast to the information which should be obtained by a machinist in process of taking management decisions. This enables to define the ways of a locomotive brigade informative functioning capacity decrease.

This article describes in more detail the issue of using predictive models of NAR neural networks to predict the course of certain quantities, which may indicate a problem with the industrial machines or their major failures. It is very important to find sufficient size of the structure and values of parameters that directly affect the output accuracy of the model. This article presents the way in which it is possible to automatically find the settings of these NAR models so that the required final accuracy metric is achieved. This presented algorithm was tested on simulation data samples collected by using the M5StickC microcontroller device. This collected dataset presented in this article contains accelerometer and gyroscopic data only, but there is a possibility to expand and add some other sensors to this microcontroller, to collect some other relevant data. This M5-StickC microcontroller device can be used for gathering data in the first phase of the machine state analysis without interfering with the mechanical construction and electrical connections of the machine. Testing of proposed algorithm was carried out in MATLAB environment. The article also describes the way in which these predictive NAR neural network models can be implemented directly in control systems, specifically PLCs from the manufacturer SIEMENS without the use of 3rd party analytical platforms. This application can be helpful in the area of predictive maintenance tasks, especially in avoiding critical failures of industrial machines and devices, or some of their specific parts.

In the automotive industry, 3D laser scanning is the most frequently used method to check the geometry of the shape and dimensions of a body in individual stages of production, where the scanned model is compared in a computer model. This procedure identifies excessive deformations around the roof antenna of an SUV. The manufacturer assumed that these deformations occurred during the antenna assembly process. An analysis of possible causes of deformation occurrence was performed based on the comparison of results obtained by numerical and experimental modelling. Experimental measurements using tensometry performed on a painted and unpainted roof structure showed the effect of paint reinforcement on deformations around the antenna. Based on the analysis of results obtained directly under operational load, it was found that the installation of the antenna was carried out on the already deformed roof plate and the installation process itself was not the primary cause of the identified excessive deformations. The presented results of deformation and stress analysis document deeper connections between the measured displacements of 3D scanning and the technological process of component production itself. The use of the achieved results should especially help designers, calculators, technologists as well as production quality controllers.