Cirp Annals-Manufacturing Technology最新文献

Cutting and abrasive processes affect the surface layer state of the components treated. This determines their performance in service. An adjustment of the surface layer properties would allow for enhanced performance. This paper introduces the influences of named processes on the surface layer state and their systematics. Models and sensor concepts for surface conditioning are described and combined to soft sensors which are the basis for active control within the processes. A validation study and actual applications of the conditioning concept are shown, allowing for further technological and scientific understanding of surface conditioning and its contribution to material and energy efficiency.

The absence of effective quality prediction methods for functional composite materials (FCMs) produced by laser additive remanufacturing (LARM) hampers their application due to the complex cross-scale defects, including surface cracks and thermal damage to the internal reinforcement phase. This paper presents a multi-task deep learning-empowered digital twin for predicting visible and invisible defects in the fabricating process of FCMs. The dimensions of FCM trajectory, thermal damage to the reinforcement phase, and forming cracks were predicted via a parallel multi-task deep learning model. The dynamic visualization of the digital twin is realized through cross-sectional modeling and provides an intuitive and effective perception for monitoring the process.

Laser powder blown directed-energy deposition (DED) enables the flexible fabrication of functionally gradient structures. In the region where multiple feedstocks are blended, it is natural to apply the ‘rule of mixtures’. Here, we show that by blending PH48S and 316L-Si steel powders using DED, the resulting material possesses lower yield strength but increased hardening and a secondary hardening regime not observed in the original alloys, invalidating direct interpolation between feedstock properties. Microstructural analysis of the mixed material revealed a solidification-driven multi-phase hierarchal microstructure. This result enables greater accuracy and increased design space in the co-design of materials and manufacturing processes.

Recent advancements in large language models (LLMs) demonstrate great potential in supporting engineering design, especially conceptual design. Prompt engineering plays an important role in facilitating designer-LLM collaboration in conceptual design. This paper proposes a new classification scheme that categorizes design-specific prompts into multiple classes. It also introduces different patterns for synthesizing design prompts, being grounded in the theoretical foundations of prompt engineering and domain-specific design methodology. A design experiment, utilizing ChatGPT, was conducted to investigate the impacts of different syntheses of design prompts on the effectiveness of LLM in concept generation, as measured by the metrics of novelty and diversity.

Autonomous robots that understand human instructions can significantly enhance the efficiency in human-robot assembly operations where robotic support is needed to handle unknown objects and/or provide on-demand assistance. This paper introduces a vision AI-based method for human-robot collaborative (HRC) assembly, enabled by a large language model (LLM). Upon 3D object reconstruction and pose establishment through neural object field modelling, a visual servoing-based mobile robotic system performs object manipulation and navigation guidance to a mobile robot. The LLM model provides text-based logic reasoning and high-level control command generation for natural human-robot interactions. The effectiveness of the presented method is experimentally demonstrated.

The effects of thermal history on workpiece hardness and cracking in the cladding of M2 high speed steel were investigated through the combination of numerical and experimental techniques. A simple thermal model was developed to simulate the workpiece temperature in the deposition. The obtained results showed that the workpiece temperature should be kept above the martensitic transformation point by continuously applying heat to avoid cracking and temper softening. The simplified temperature simulation in this study is effective for the deposition path strategy because it was able to predict where the temper softening would occur.

This paper aims to investigate the scientific integration of data science with product design towards data-driven design (D3). Data science has potential to facilitate design decision-making through insight extraction, predictive analytics, and automatic decisions. A systematic scoping review is conduced to converge various D3 applications in four dimensions: the design dimension about design operations, the data dimension about popular data sources and common data-related challenges, the method dimension about the methodological foundations, and the social/ethical dimension about social/ethical considerations and implications. Based on the state-of-the-art, this paper also highlights potential future research avenues in this dynamic field.

This research investigates the multi-material laser powder bed fusion (MM-LPBF) printing of CuZr/316 L stainless steel (SS) within the same layer using four printing strategies. Printing CuZr first exhibits better quality. Conversely, printing 316 L SS first led to notable cracks due to elevated thermal stress and resulted in a larger diffusion zone due to material convection caused by the Marangoni effect. Material overlapping is effective in reducing the formation of cracks and creating a larger diffusion zone compared to non-overlapping. A topology-optimized liquid cooling heat sink with CuZr core and 316 L SS protective layer was printed to demonstrate electronic cooling applications.

An adaptive laser power control strategy for Selective Laser Melting (SLM) has been developed using data from a co-axial photodiode monitoring system with 200 KHz temporal resolution. A supervised machine learning based algorithm outputs variable laser power along the scanning path based on mechanistic features. The approach was implemented on a commercial machine and demonstrated an average 12 % reduction in porosity size and 65 % reduction in the standard deviation of porosity size measured by X-Ray Computed Tomography (CT) compared to parts built with constant laser power. This approach is scalable and its precalculated nature is compatible with regulatory concerns.

Optical waveguides are crucial in the development of integrated optical circuits. This study examines their formation in dielectric materials using ultrashort pulse laser direct writing. Emphasizing the significance of the processing parameters for writing low-loss waveguides, we explored the underlying mechanisms through in situ observations of internal modifications and a detailed analysis of electron density and refractive index changes. Employing a combination of pump-probe imaging and Mach-Zehnder interferometry, we reveal the dynamics of laser-induced modifications in quartz, highlighting the influence of spatial overlap and irradiation spacing. Our research offers insights into the micro-level phenomena that affect the overall waveguide performance, thereby providing enhanced fabrication methods.