Multi-criteria optimization of 3D-printed PETG-carbon fiber composite parts using entropy-TOPSIS method

Abstract

This study outlines the investigation of additively manufactured parts based on polyethylene terephthalate glycol reinforced with carbon fiber (PETG-CF). A total of 27 experimental runs was designed, selecting 3 distinct levels of layer height (LH), infill density (ID), and raster angle (RA). The responses of the study were tensile strength (TS), flexural strength (FS), and compressive strength (CS). The entropy-based technique for order of preference by similarity to ideal solution (TOPSIS) method was applied to simultaneously optimize the multiple responses. Additionally, analysis of variance (ANOVA) was used to comprehend the statistical significance of the input parameters. The highest values of TS, CS, and FS were 40.80 MPa, 51.76 MPa, and 61.00 MPa, respectively. The ANOVA revealed that the statistical models for TS, FS, and CS fitted well with the data, as evident from their high R-sq values of 0.9090, 0.9683, and 0.9820, respectively. The TOPSIS determined the optimal values of TS, CS, and FS to be 40.80 MPa, 51.76 MPa, and 53.10 MPa for an LH of 0.1 mm, an ID of 100%, and an RA of 0^°. The microscopic analysis indicated ductile failure of the material, characterized by void coalescence and both partial and total matrix separation during tensile loading.