Tomasz Ryba

This paper presents the experimental results of a Machine Learning application for the health monitoring of a conveyor belt. The real-time analysis of the rubber belt condition is a crucial issue in achieving safety and avoiding critical failures and related expenses. The measuring system based on strain gauges was applied to identify the actual state of the belt. Using the Classification Lerner application from MATLAB platform, 22 algorithms were tested, and using the Diagnostic Feature Designer application, the analysis was performed. Three tested ML algorithms were able to classify the states of the conveyor belt with preset damages correctly, exhibiting 100% prediction accuracy. The k-nearest neighbors (KNN) classifiers and neural networks failed to achieve that level of accuracy.

The repeatability of the material properties is required to ensure the proper performance of the engineered systems that are constructed using these materials. In this paper, an analysis of the sintered ceria-stabilized zirconia is presented. This material exhibited high mechanical properties, due to the mechanism of strengthening via phase transition. The reproducibility was assessed for the material made out of a starting powder produced by fluoride salt precipitation. To fabricate specimens, a novel electroconsolidation method was used, ensuring a high heating rate, relatively low sintering temperatures, and short holding time. Weibull analysis was performed considering the bending strength of specimens and their microhardness. The obtained values of both shape parameter m and scale parameter σ₀ indicated that the ZrO₂ stabilized with 5 wt.% CeO₂ samples exhibited low variability of strength and hardness. The experimental evidence and statistical analysis reveal an influence of the m-phase, which has lower symmetry and therefore its addition makes ceramic weaker and softer. Furthermore, its progressive replacement by the t-phase, which has higher symmetry, makes ceramic both harder and stronger. Reducing the mol% increases the risk of the appearance of the highest addition of the monoclinic phase; increasing it is unfavorable from the point of view of the sintering process. Statistical and manufacturing evidence suggests that the choice of 5%/mol is optimal.

This paper presents the results of investigations on the accuracy of contact point identification during coordinate measurement, which is crucial in the context of the Industry 4.0 concept. In particular, the effects of swivel length and probe declination angle during measurement were analyzed. In the experiments, deviations from the expected coordinates (0,0,0) of the contact point were analyzed for different rotational angles of the probing head. It was found that the recommended vertical positioning of the stylus at an angle of A = 0° might have introduced some insignificant errors. Increasing angle A up to 15° generated additional errors of negligible values in comparison with the measurement accuracy of the CMM. However, an increase in angle A up to 90° introduced additional errors as high as 10 μm. This contact point identification error will have a certain effect on the best fitting element and subsequent calculations and on the respective measurement results.

This paper is devoted to the real-time monitoring of close transportation devices, namely, belt conveyors. It presents a novel measurement system based on the linear strain gauges placed on the tail pulley surface. These gauges enable the monitoring and continuous collection and processing of data related to the process. An initial assessment of the machine learning application to the load identification was made. Among the tested algorithms that utilized machine learning, some exhibited a classification accuracy as high as 100% when identifying the load placed on the moving belt. Similarly, identification of the preset damage was possible using machine learning algorithms, demonstrating the feasibility of the system for fault diagnosis and predictive maintenance.

This paper presents the results of a metrological analysis of the additively manufactured (AM) copies of a complex geometrical object, namely the fossil skull of Madygenerpeton pustulatum. This fossil represents the unique remains of an extinct “reptiliomorph amphibian” of high importance for palaeontological science. For this research, the surface was scanned and twelve different copies were 3D-printed using various devices, materials, and AM techniques. The same digitized model was used as a reference to compare with the surfaces obtained by Mitutoyo Coordinate Measuring Machine (CMM) CRYSTA-Apex S 9166 for each copy. The fidelity of the copies was assessed through statistical analysis of the distances between compared surfaces. The methodology provided a good background for the choice of the most accurate copies and the elimination of the less accurate ones. The proposed approach can be applied to any object of complex geometry when reproduction accuracy is to be assessed.