Optimizing FDM 3D printing parameters for improved tensile strength using the Takagi-Sugeno fuzzy neural network

3D printing is a popular technology for fabricating three-dimensional objects, and it is crucial to select appropriate printing parameters to enhance production quality, reduce costs, and purposefully control the performance of printed products. This study aimed to establish a predictive model for the tensile strength of different FDM 3D printing parameter combinations with polylactic acid (PLA) materials using a Takagi-Sugeno (T-S) fuzzy neural network model. The deposition layer thickness (d), printing temperature (T), printing speed (V), and filling density (rho) were selected as influential factors on the tensile strength. Based on the prediction results of the T-S fuzzy neural network model, three sets (the 4th, 32nd, and 58th) of predicted parameters were chosen for experimental validation. The results showed that the deposition layer thickness affects the morphology of the PLA fibers, and increasing the filling density reduces the air gaps, ultimately impacting the tensile strength of the test specimen. Furthermore, the relative error between the predicted and experimental test values of the tensile strength ranged from 5 % to 10 %. Therefore, the T-S fuzzy neural network model holds great potential for predicting the tensile strength of FDM 3D printed tests.