Search

Machine learning techniques are extensively used to understand and predict complex non-linear phenomena across various applications. Moreover, these techniques minimize the time and costs associated with experimental and numerical analysis. In this work, a deep learning technique, specifically artificial neural networks (ANN), was employed to predict the density/porosity of laser powder-bed fusion (LPBF) additively manufactured (AM) parts by training the ANN model with X-ray computed tomography (CT) images. In addition to the experimental data, synthetic CT data was generated and used to improve the performance of the ANN model. The ANN model was then optimized for the number of hidden layers and neurons. Different errors like mean absolute error (MAE), root mean square error (RMSE), and square of co-relation coefficient (R2) were used as performance metrics to determine the accuracy and effectiveness of the network. The proposed ANN model was validated and showed excellent predictions (R2 = 0.9981, MAE = 1.6944 x 10−5). The framework proposed in this work can be used to speed-up the quality assurance of AM parts by reducing the time required for the analysis of CT data.

An exponential rise in demand for artificial implants emphasizes the need for materials with biocompatibility and mechanical properties comparable to cortical and cancellous bone, to mitigate stress shielding effects. This study investigates the performance of β-Ti alloys (Ti-12Ta-23Nb-24Zr, Ti-32Nb-25Zr) fabricated via Laser Powder Bed Fusion (L-PBF) under optimal process parameters. The process was optimized with laser power (200 W), scanning speed (1000 mm/s), hatch distance (0.065 mm), layer thickness (40 μm), and exposure time (50 μs). Results showed that Ti-12Ta-23Nb-24Zr exhibited the highest compressive strength (864 MPa), superior biocompatibility (RGR = 111.6%), and reduced surface roughness (Ra = 0.462 μm), making it ideal for high-stress applications. Ti-32Nb-25Zr demonstrated good mechanical strength (842 MPa), high ductility, and excellent biocompatibility (RGR = 94.7%), suitable for initial bone integration. Ti-6Al-4V, while exhibiting lower bio-mechanical properties, remains viable for applications with lower mechanical demands. It was found that beta Ti alloys have improved biocompatibility in the order Ti-12Ta-23Nb-24Zr>Ti-32Nb-25Zr>Ti-6Al-4V. The findings underscore the adaptability of optimized L-PBF settings across β-Ti alloys, paving the way for advanced patient-specific orthopedic implants. Future work will focus on advanced surface modifications and long-term in-vivo studies to further enhance performance

The primary concern of the Industry 4.0 is the direct digital manufacturing of customized products on demand at high production speed, high accuracy with functional material property. Although the unique capabilities of existing additive manufacturing technologies make it suitable for direct digital manufacturing, there are numerous limitations which include low printing speed, less accuracy and repeatability, and a limited selection of materials for a particular application. Therefore, a high-speed additive manufacturing approach is proposed in this paper, that is capable of achieving high speed of production, high accuracy, and surface finish, and functional material property. For better understanding, authors describe those additive manufacturing technologies that are capable of achieving the aforementioned characteristics. For validation, samples of various dimensions were 3D printed on a selective laser sintering and a high-speed multijet fusion 3D printer. The results were compared in the context of printing speed, surface roughness (Ra), and hardness of printed parts. Results revealed that the multijet fusion process is significantly faster than its counterpart while sacrificing Ra to some extent but the hardness of printed parts is not changed significantly. The selective laser sintering-printed samples had a 15% lower Ra compared with multijet fusion samples. The results also revealed that the multijet fusion process might be able to print composite/multi-materials; however, more research needs to be done.

Wave springs have better properties than helical springs in terms of load bearing capacity, mass, energy absorption efficiency, and stiffness. Currently, available wave springs have waves that do not merge with each other (non-contact) with uniform shape and dimensions, manufactured by traditional manufacturing processes using customized machines in which forming, twisting, and cutting processes take place simultaneously. This study aims to design and manufacture wave springs with waves that merge with each other by additive manufacturing (AM). Variable-dimension wave springs are complex in design and can only be manufactured by AM owing to its design freedom, but they have much higher performance in load bearing, stiffness, energy absorption, and energy release compared to both uniform-dimension and uniformly shaped wave springs. To accomplish wave springs manufactured by AM, ten different wave springs were designed and manufactured with variable and uniform dimensions, using a HP MultiJet Fusion printer, by keeping the mass and height of each design constant. Compression tests were carried out for up to ten cycles to study and compare the load bearing capacity, energy absorption, stiffness, and energy loss for all designs. The results showed that the stiffness of the spring increased in the contact wave springs, which ultimately led to more energy absorption in these designs. The taper-wave spring was the best among all designs for load bearing, and it had the highest energy loss capacity, which was validated by finite element analysis of each design.