Metallic wing covers are defined here as wing skins with mechanically attached or integrally machined stringers; an integrally stiffened panel is the monolithic case. This structured narrative review examines upper and lower metallic wing covers as manufacturing objects in civil transport, business aircraft, and selected military fixed-wing programmes. Public evidence, spanning shot-peening process origins in the 1920s–1930s and principal aircraft-programme evidence from the 1950s onward, was synthesised from peer-reviewed papers, SAE Technical Papers, granted patents, published patent applications, trade literature, government reports, and supplier disclosures using explicit source weighting rather than statistical aggregation. Evidence was graded by source strength, and granted patents and published patent applications were treated as capability evidence rather than proof of serial production unless independently corroborated. The synthesis shows that forming-route selection is governed by structural scale, cover role, curvature class, alloy and temper, inherited stock state, panel architecture, and compensation or validation capability. Upper/lower cover divergence is directly evidenced for selected Airbus, Gulfstream, and B-1B cases; where cover-separated routes are not publicly disclosed, it is treated as an inference rather than a production fact. The strongest public evidence supports peen and particle-impact routes for directional or inflected lower covers, creep age forming for large smooth heat-treatable covers, and specialised laser or hybrid routes where independently corroborated.
Wear of the engine cylinder liner can significantly affect its performance and service life. The analysis of the oil lubricating the engine can be used to determine the amount of wear. Traditional detection methods are labor-intensive and require high technical expertise. This study proposes a new method for engine cylinder liner wear detection based on deep learning. By introducing multi-feature fusion technology, the method upsamples deep feature maps and adds shallow feature information to construct a multi-scale feature pyramid structure, effectively improving detection accuracy. It was observed that the Faster R-CNN algorithm with multi-feature fusion had the ability to detect different forms of wear particles in the image of lubricating oil. The results were compared against the initial Faster R-CNN model, and AP50 accuracy rose by 1.8%, while AR1 recall increased by 4.5%. Through continuous optimization of model parameters, this paper obtained the optimal model configuration. Distinct from conventional single-scale feature extraction wear particle detection schemes, the multi-layer cross-scale feature fusion module proposed in this paper strengthens shallow edge texture and deep semantic feature coupling and achieves targeted performance improvement for small, irregular lubricant wear debris without redesigning the two-stage detection paradigm.
The present strategy for designing ceramic mould materials (CMMs) is primarily based on modifying the preparation process and material components through experimental data to achieve improved material performance. However, this approach necessitates a significant investment in labour costs and time. The study proposes an immunogenetic algorithm to optimise the design of CMMs, alongside a hybrid algorithm that incorporates both the immunogenetic algorithm and a backpropagation neural network. The study's findings indicate that the immunogenetic algorithm requires fewer iterations during the process of optimising fracture toughness and hardness. Specifically, it only needs 9 and 10 iterations, respectively, compared with other mainstream algorithms. The sintering process design of CMMs based on a hybrid algorithm was optimised, resulting in errors in fracture toughness, flexural strength, and hardness of 5.75%, 4.03%, and 0.23%, respectively. The corresponding sintering temperatures and holding times were 1455.6 °C and 12.7 min, 1454.2 °C and 11.0 min, and 1490.8 °C and 24.1 min. The fracture toughness, flexural strength, and hardness all had optimal values of 6.94 MPa·m 1/2 , 924.96 MPa, and 15.20 GPa, respectively. The corresponding mass percentages of ZrO 2 and WC were 12.30% and 7.0%, 13.76% and 19.75%, and 7.75% and 2.23%, respectively. In summary, our proposed method is able to effectively reduce the cycle time for optimal CMM design. Furthermore, it provides a valuable point of reference for future CMM designs.
There is a high need for parts with well-balanced mechanical properties in industries. This is achieved through the use of various heat treatment processes. The adjustment of heat treatment parameters can lead to a diversity of microstructures. The heat treatment parameters of cooling, such as furnace, air, oil, and water cooling, as well as the soaking time, play important roles in determining the characteristics of microstructures obtained. The importance of these parameters is underscored by the fact that the mechanical properties of metals and their alloys, such as strength and ductility, are dependent on their microstructures. This paper presents a review of the effects of annealing, normalizing, high-temperature annealing, and solution and aging heat treatments on the microstructures and mechanical properties of Ti6Al4V, high-strength steels, and aluminum alloys. It is clear from this study that annealing coarsens the microstructural features, reduces hardness and strength, and increases ductility and machinability of all three different alloys considered here, while normalizing refines their microstructures and improves strength, while ductility is reduced. Evidently, heat treatment processes play an important role in the microstructure and mechanical properties of the three alloys. Therefore, structural parts with well-balanced mechanical properties needed in various industries can be achieved through the application of these heat treatment processes.
This study presents a comparative evaluation of unsupervised anomaly detection techniques using the PyOD Python library, applied to three representative industrial scenarios: Intrusion Detection, Fault Detection, and Predictive Maintenance. Three public datasets, each corresponding to one of these domains, are used to assess the generalizability and robustness of the methods. The methodology includes data normalization, feature selection, and class balancing to address the specific challenges of each dataset, such as class imbalance and heterogeneous feature spaces. Performance is assessed using metrics such as Precision, Recall, Specificity, F1-score, AUC-PR, AUC-ROC, Balanced Accuracy, and Execution Time (training and testing). The findings offer valuable guidance for selecting robust anomaly detection models in industrial applications beyond cybersecurity, including those related to reliability and maintenance. The code and results will be publicly available on GitHub https://github.com/hvelesaca/PyOD_OPC-UA to support future research.
