Application of Taguchi Grey-Based ANFIS model for prediction of process parameters in fused deposition modelling of PETG

Fused Deposition Modelling (FDM) is one of the most used Additive Manufacturing technologies in the world due to its application in the fabrication of complex structures with different materials. The present study deals with improving the process control and efficiency of FDM for Polyethylene Terephthalate Glycol (PETG) material through the development of a predictive model using an Adaptive Neuro-Fuzzy Inference System (ANFIS). Systematic analysis of the effect of the most important FDM process factor-infill density, nozzle temperature, and printing speed performance metrics while printing time, dimensional deviation, and surface roughness is performed. The required data set is generated for the conduct of experimental trials based on a Taguchi L27 orthogonal array. An ANFIS Model is, therefore, developed, which integrates neural learning with fuzzy logic to predict printing characteristics by analyzing the interaction between inputs and outputs. The model has been developed such that it is highly capable of predicting results, thereby inducing the best or optimum process variables to enhance FDM performance. Infill density, nozzle temperature, and printing speed follow last in their effects on the performance of FDM. A GRG value of 0.8547 and a R² of 0.9999 confirm quite strong prediction capabilities of the Grey-ANFIS model between those predicted and experimental results, endorsing it as a reliable optimization tool in FDM of PETG material.