Application of Box-Behnken, Ann, and Anfis Techniques for Identification of the Optimum Processing Parameters for FDM 3d Printing Parts

Abstract

Fused Deposition Modeling, among the various 3D printing approaches, is becoming more and more popular because of its capacity to produce complicated parts quickly. The tensile strength of parts printed with polylactic acid (PLA) showed a significant variation of many factors such as printing speed, printing temperature, printing angle and infill pattern. This study presented an experimental investigation of collecting data with four input factors namely printing speed, printing temperature, printing angle and infill pattern with the tensile strength response. The research methodology of the RSM Box-Behnken DOE method, ANN (Artificial neural network), and ANFIS (Adaptive neuro-fuzzy inference systems) has been used to determine the optimum process 3D printing parameters. The obtained results based on RSM, ANN and ANFIS methods are used to predict the tensile strength of 3D printed FDM details. The best tensile value is 7,03303 MPa corresponding to print speed of 30,0003 mm/s, printing temperature of 211,594℃, printing angle of 90° with Honeycomb” infill printing pattern. Moreover, the results also highlighted that ANFIS is potential approach for forecasting the tensile strength of 3D printing parts more competitively.

Research Into the Accuracy of Holes in 3d Printing Using Taguchi Method

“…Of scientific interest is t accuracy of 3D printed details and possibility for industrial use in various fields of technology. [1] - [9] Parida used the experimental design proposed by Taguchi based on the L27 orthogonal array. It was found the optimal combination of different parameters in this research.…”