In this study, a new lattice structure ARCH is proposed, ARCH lattice structure and traditional lattice structures BCT were fabricated from 316 L stainless steel by SLM. Mechanical properties and deformation behavior of the lattice structures were experimentally investigated. The results show that the ARCH lattice structures have better mechanical property and energy absorption capability than BCT lattice structures. Under the same relative density, the compressive strength and elastic modulus of the ARCH lattice structure is 121.27% and 60.48% higher than the BCT lattice structure, and the ARCH lattice structures have better energy absorption properties than the BCT lattice structures.

In recent years, selective laser melting (SLM) has been widely used in aerospace, automobile, biomedicine and other fields. However, there still remain many challenges to obtain consistent parts at the different positions on the base plate, which could be harmful to the industrial mass-production. In SLM process, the process by-products that flow with the shielding gas may influence the microstructure and tensile properties of the parts placed on different positions of the base plate. In this study, the velocity field of the shielding gas with different shielding gas volume flows was simulated. The tensile properties of the samples fabricated with different shielding gas volume flow were experimentally studied. The results show that the shielding gas volume flow has a strong influence on the sample consistency, and proper increase in shielding gas volume flows can be beneficial to consistency and tensile strength.

With the deepening adoption of whole-lifecycle management paradigms in aviation equipment, accurate prediction of remaining useful life has emerged as a pivotal technical enabler for ensuring flight safety and optimizing maintenance resource allocation. This study systematically addresses limitations in existing data-driven remaining useful life prediction methodologies through comprehensive optimizations spanning data preprocessing protocols and model architectural enhancements. To mitigate the local optimum entrapment inherent in deep learning hyperparameter optimization, an improved Gray Wolf Optimizer incorporating dynamic perturbation factors is proposed. This algorithm is subsequently deployed to optimize hyperparameters within the redesigned predictive architecture. Comparative analyses reveal that the proposed framework achieves superior prediction accuracy compared to mainstream optimization-driven models and state-of-the-art approaches. The results substantiate the capability of dynamic perturbation strategies to enhance both hyperparameter optimization quality and prediction stability, ultimately delivering an efficient solution for aero-engine remaining useful life estimation. The experiments on the C-MAPSS Dataset verify the effectiveness of these improvements.

AI-integrated robotics in Industry 5.0 demands advanced manufacturing systems capable of autonomously interpreting complex geometries and dynamically adjusting machining strategies in real time—particularly when dealing with aerospace components featuring large-curvature surfaces. Large-curvature aerospace components present significant challenges for precision drilling due to surface-normal deviations caused by curvature, roughness, and thin-wall deformation. This study presents a robotic drilling system that integrates adaptive PCA-based surface normal estimation with in-process pre-drilling correction and post-drilling verification. This system integrates a 660 nm wavelength linear laser projector and a 1.3-megapixel industrial camera arranged at a fixed 30° angle, which project and capture structured-light fringes. Based on triangulation, high-resolution point clouds are reconstructed for precise surface analysis. By adaptively selecting localized point-cloud regions during machining, the proposed algorithm converts raw measurements into precise normal vectors, thereby achieving an accurate solution of the normal direction of the surface of large curvature parts. Experimental validation on a 400 mm-diameter cylinder shows that using point clouds within a 100 mm radius yields deviations within an acceptable range of theoretical normals, demonstrating both high precision and reliability. Moreover, experiments on cylindrical aerospace-grade specimens demonstrate normal direction accuracy ≤ 0.2° and hole position error ≤ 0.25 mm, maintained across varying curvature radii and roughness levels. The research will make up for the shortcomings of existing manual drilling methods, improve the accuracy of hole-making positions, and meet the high fatigue service needs of aerospace and other industries. This system is significant in promoting the development of industrial automation and improving the productivity of enterprises by improving drilling precision and repeatability, enabling reliable assembly of high-curvature aerospace structures within stringent tolerance requirements.