Enhanced FDM Printing Accuracy in Low-Carbon Production Mode Using RSM-NSGA-II and Entropy Weight TOPSIS Method

Compared to traditional processes, fused deposition modeling 3D printing can manufacture parts of various shapes without the need for additional equipment, moulds, fixtures, or other tools. Its excellent characteristics have been widely applied in many industries. However, balancing product quality with low-carbon production has always been a pressing issue for 3D printing companies to address. To improve the stability of 3D printing in terms of part size accuracy and sustainable development, an orthogonal experimental design method, RSM-NSGA-II, and an entropy weight TOPSIS method were employed to optimise the factors affecting size accuracy and carbon emissions. The layer height, nozzle temperature, filling density, first layer height, and printing pattern were selected as factor variables, and the circular runout tolerance value and carbon emissions of printed parts were set as optimisation objectives. An L18 orthogonal experimental design was established. The influence of process parameters on quality indicators and the optimal combination of process parameters were analysed through range calculation. In addition, the NSGA-II-based optimisation model was constructed using the experimental design method in response surface methodology, and combined with the entropy weight TOPSIS method, to determine the optimal FDM 3D printing process parameter scheme with the best comprehensive performance. The results indicate that the response surface model established in this paper has good adaptability. When the layer height is 0.2 mm, the nozzle temperature is 243 °C, the filling density is 70%, and the first layer height is 0.15 mm, the circular runout tolerance value and carbon emissions are reduced by 64.29% and 53.45% respectively, compared to the original values. This study provides a theoretical basis and technical support for optimising the FDM manufacturing process in low-carbon and environmentally friendly production.