Improving mechanical properties of 3D printed products through honeycomb structure and deep learning optimization

Abstract

The use of three-dimensional (3D) Printing has rapidly grown due to its various applications in daily life. However, products made with this technology are not widely adopted because their quality is not yet seen as equal to that of products made with traditional materials. To address this issue, a honeycomb structure model was developed to improve the mechanical properties of the items created using 3D printing technology. This model not only saves materials, but also reduces printing time compared to previous models. The honeycomb structure was designed using a theoretical model, and its components were modified to ensure compatibility with the 3D printing extrusion method. Moreover, deep learning was used to optimise the honeycomb structure model’s boundary conditions by generating contour curves through linear interpolation. This process significantly improved the mechanical strength of the honeycomb structure model compared to structures made using 3D printing technology. The theoretical findings were validated through various material tensile tests conducted under different scenarios. As a result, this model can be used in various industries and products that use 3D printing technology, thanks to the critical role that deep learning in improving its mechanical properties.