Development of ANFIS Predictive Model for Additive Manufacturing (Fusion Deposition Modeling) of PETG Material for automotive components

Additive Manufacturing (AM), specifically Fusion Deposition Modeling (FDM), has become widely popular due to its capacity to produce intricate shapes using different materials. The objective of this study is to improve the efficiency and accuracy of the FDM process for Polyethylene Terephthalate Glycol (PETG) material, which is used for producing automotive components that require impact resistance, dimensional stability, and good surface finish. This will be achieved by creating a predictive model using an Adaptive Neuro-Fuzzy Inference System (ANFIS). The study examines the impact of FDM parameters, such as layer height, nozzle temperature, and printing speed, on important printing attributes like dimensional accuracy, surface quality, printing speed and dimensional deviation. The necessary information is obtained from experimental trials of FDM printing, which were conducted using various combinations of parameters. The experimental trials have been planned by Taguchi’s approach. The ANFIS predictive model is built by utilizing the gathered dataset, harnessing the synergistic learning capabilities of neural networks and the interpretability of fuzzy logic systems. The model is trained and optimized to precisely forecast printing characteristics by analyzing input parameters, offering valuable insights into the intricate relationships within the FDM process. The ANFIS predictive model's performance is assessed using statistical analysis and compared to experimental results. The model that was created shows its ability to accurately forecast printing characteristics and capture complex process dynamics. The proposed Adaptive Neuro-Fuzzy Inference System (ANFIS) predictive model provides a methodical approach to optimize Fused Deposition Modeling (FDM) parameters. This allows manufacturers to improve productivity and quality in additive manufacturing operations using PETG material. This research enhances the comprehension of FDM processes and offers a useful tool for optimizing the process in different engineering and industrial applications.