PSI - Issue 34

Atefeh Rajabi Kafshgar et al. / Procedia Structural Integrity 34 (2021) 71–77 Atefeh Rajabi Kafshgar et al./ Structural Integrity Procedia 00 (2021) 000 – 000

72 2

1. Introduction In the last two decades, there has been appeared a growing trend towards additive manufacturing (AM). Due to the rapid growth of technology, AM can provide many advantages such as increased production speed, the ability to build complex structural models, reduce waste, and reduce warehouses (Majeed et al. (2020), Kleer & Piller (2019)). Moreover, different types of materials and technologies are used in the AM. Therefore, it can be used as a flexible technology in various areas such as medical, automotive, aerospace, defense, education, etc. (Fasel et al. 2020, Ghomi et al. (2020), den Boer et al. (2020), Assante et al. (2020)). The application of these technologies is no longer limited to the production of prototypes and they are used to print the main components of a system or product (Ameri et al. (2021)). Therefore, the mechanical properties of printed materials must be studied to ensure that they have suitable strength and integrity under the applied mechanical loads. In order to optimize the mechanical properties such as strength in the AM components and materials, it seems necessary to investigate the effect of printing parameters. On the other hand, among all different types of materials applied in 3d printing for laboratory and real applications, the most used material is polymer. Also, Fused Deposition Modelling (FDM) is a common additive manufacturing method because of its wide adaptability, simple mechanism, and low cost (Ngo et al. (2018), Ameri et al. (2020)). Hence, this study seeks to improve the mechanical properties of polymer-based printed objects by optimizing the FDM parameters. So far, most of the works in this area are experimental based, and very little attention has been paid to apply the Design of Experiments (DOE) methods to optimize the mechanical properties of the final parts. Also, only few studies have employed meta-heuristic algorithms to investigate 3D printing parameters. For example, Leite et al. (2018) studied the influence of four processing parameters (Filling Density, Extrusion Temperature, Raster Angle and Layer Thickness) on the mechanical properties including the ultimate tensile strength (UTS), yield tensile strength, modulus of elasticity, elongation at break and toughness of PLA 3D printed parts using the ANOVA statistical analysis. Regarding all the combinations of the levels of the factors, twenty-four experiments were done to analyze the results. Taguchi methodology was applied as an optimization tool by Sagias et al. (2018). They tried to improve the mechanical properties of 3D printed ABS material produced by FDM technology. Four factors including layer thickness, printing pattern, print strength, and placement with three levels for each factor were considered in their work and the UTS was measured as the response variable. In the same way, Wankhede, et al. (2020) applied the Taguchi method but for a different case. In their work, Taguchi's L8 Orthogonal Array was employed to analyze the influence of the input experimental parameters (infill density, layer thickness, and support style) on the response parameters (manufacturing time of the part and surface roughness) with samples made of ABS polymer. Moreover, ANOVA was established to understand the significant characteristics of the process variables. The set of input variables has been determined for the individual output response variables. Yang et al. (2019) concentrated on the optimization of the printing parameters to achieve higher tensile strength and lower surface roughness with less build time during the FDM process based on Central Composite Design (CCD) for the tensile specimen. The effects of five process parameters (nozzle diameter, liquefier temperature, extrusion velocity, filling velocity, and layer thickness) on the three outputs of tensile strength, surface roughness and build time were investigated. Response Surface Methodology (RSM) combined with non sorting genetic algorithm (NSGA)-II was developed to optimize the process parameters. Dev and Srivastava (2020) studied multi-objective optimization of FDM process parameters to produce strong and lightweight parts. Two important objectives namely material consumption and compressive strength with respect to selected process parameters (e.g. layer thickness, build orientation, infill patterns with varying densities) were investigated. In this research, the Taguchi approach was used for experimental design and objective functions resulted from the regression model are optimized by using the NSGA-II approach. The Pareto optimal solutions are also obtained and compared with the experimental data. The experimental data used in this study are adapted from a previous study that investigated the tensile properties of 3D printed parts made with different process parameters and this paper attempts to show that the mentioned reference study can be done with fewer experimental tests by the means of DOE and specifically the Taguchi method. Also, by utilizing multi-objective optimization two important mechanical properties (i.e. UTS and toughness) are considered together and Pareto optimal solution set is suggested.