The effective utilisation of equipment is essential for achieving the operational goals within production sectors, particularly in industries involving heavy machinery. Throughout its lifecycle, equipment is exposed to dynamic loads and harsh operational environments, leading to potential failures that may significantly shorten their service life. Therefore, evaluating equipment reliability is crucial for mitigating production losses and ensuring continuous operations. This study presents a comprehensive reliability analysis of underground mining machinery, with a focus on Load-Haul-Dump (LHD) systems, which are key to material handling in mining operations. Reliability assessments are performed using methodologies based on the series configuration of repairable systems. The reliability of each LHD system is quantitatively evaluated by employing a feed-forward back-propagation artificial neural network (ANN) model implemented in MATLAB. This model is designed to predict the optimal responses of each LHD machine under varying operational conditions. The results obtained from the ANN model are compared with the calculated reliability values, demonstrating a high degree of correlation between the predicted and observed outcomes. This strong alignment underscores the potential of ANN-based models in accurately forecasting system reliability. Based on the analysis, recommendations are made to identify the most critical components contributing to the system's unreliability, thereby enabling targeted corrective actions. The findings provide valuable insights for engineers seeking to enhance the performance and operational efficiency of mining machinery through more informed maintenance and operational strategies.

In response to the complex characteristics of gearbox vibration signals, including high frequency, high dimensionality, non-stationarity, non-linearity, and noise interference, this paper proposes a data processing method based on improved compressed sensing. First, the K-means Singular Value Decomposition (K-SVD) dictionary is used for sparse representation, ensuring good sparsity in the frequency domain. Next, a random convolution kernel measurement matrix is employed in place of the traditional Gaussian random matrix, satisfying the equidistant constraint while enhancing both computational and hardware implementation efficiency. Finally, the Generative Flow (GLOW) model is introduced, incorporating the measurement matrix, dictionary matrix, and sparse coefficient matrix into a unified optimization framework for joint solving. Through reversible mapping and probabilistic distribution modeling, the method effectively addresses noise interference and the challenges posed by complex signal distributions. Experimental results show that, compared with traditional compressed sensing methods, the proposed method offers superior signal reconstruction quality and better noise robustness.

Precision steel is a critical material in modern engineering, particularly in precision mechanics and high-performance construction. In this study, a mathematical model is presented to simulate the vacuum degassing (VD) process employed to reduce the hydrogen content in steel produced via the basic oxygen furnace (BOF) process. The steel, which is subsequently used for ingot casting, requires a significant reduction in hydrogen levels— from 7 ppm to below 1.5 ppm—to meet the stringent quality requirements for high-precision applications. This reduction is achieved through the VD process in combination with argon bottom stirring. The model, developed in collaboration with an industrial project in Bosnia and Herzegovina, is designed to predict the necessary degassing time and the temperature variation during the process. The model accounts for the operational parameters specified by the project sponsor and the constraints of the process. Results indicate that the hydrogen content can be reduced within 8.39 minutes under optimal conditions. Furthermore, for a molten steel starting temperature of 1670℃, the final temperature after degassing is predicted to be 1637℃. The applicability of the model has been validated through practical implementation in a new industrial facility, constructed based on the model’s predictions. This study demonstrates the broader utility of the model in designing and optimizing VD processes for precision steel production, offering significant potential for enhancing steel quality and process efficiency in similar industrial settings.

Incremental sheet metal forming (ISMF) is a promising manufacturing technique that has gained significant attention due to its ability to produce complex geometries and high-quality products, particularly for small-scale production and rapid prototyping. The integration of industrial robots into the ISMF process, referred to as roboforming, has enabled advancements in this field. However, the inherent limitations of industrial robots—particularly the reduced rigidity of robotic arms with rotary joints—can lead to dimensional inaccuracies and deviations in the final product. These limitations are primarily due to the lack of precise force control during the forming process. To address these challenges, this study introduces a novel approach to roboforming that incorporates force control alongside the position control of the industrial robot. The contact force between the tool and the workpiece is considered as an additional variable in the control loop, with the objective of improving dimensional accuracy and the overall quality of the formed product. A regression analysis was conducted to determine the mean process force required for conical geometries, with the starting radius, infeed depth, wall angle, and supporting angle serving as input variables. Experimental validation revealed that force-controlled incremental forming with a constant contact force is unfeasible, as the pressure force is highly dependent on the current radius of the workpiece and varies during the forming process. Therefore, a new control strategy is proposed, which involves the dynamic adjustment of the contact force, using the variable pressure force as an input parameter. This approach is expected to significantly enhance the precision and reliability of robot-assisted ISMF, offering a pathway for overcoming current limitations in industrial applications.