Development of Taguchi grey-based hybrid ANFIS prediction model for fused deposition modelling of HIPS

Fused Deposition Modelling (FDM) is often used in Additive Manufacturing (AM), making it popular for producing even the most complex and tailored geometry forms at low costs. With these advantages, it also has limitations in quality and efficiency in the products made out of it, influenced strongly by process parameters, which necessitate the development of predictive tools for ‘control’ of the process. This present work emphasizes HIPS material to enhance FDM performance through the establishment of a predictive model using an Adaptive Neuro-Fuzzy Inference System (ANFIS). The three important input variables are infill density (ID), nozzle temperature (NT), and printing speed (PS). The output responses are printing time, dimensional deviation, and surface quality. The experimental matrix is made by using Taguchi’s L27 orthogonal array, and therefore, the multiple performance indices from the different responses are derived using Grey Relational Analysis (GRA). These Grey Relational Coefficient (GRC) values obtained from that analysis will then be used as an input variable for training and testing of the ANFIS model. The model evolved from one that had shown good performance in prediction and predicted output responses very well. The model also gives the optimum parameter setting of 25% infill density, nozzle temperature of 240 °C, and printing speed of 65 mm/s for better and improved multiple performance. The findings indicate that the proposed ANFIS-based approach undoubtedly emerges as a strong and effective tool in improving productivity and dimensional precision, as well as overall quality in FDM of HIPS material.