Forward and Backward Modelling of Wire and Arc Additive Manufacturing Process using Multiple Adaptive Neuro-Fuzzy Inference System

Abstract

Wire and arc additive manufacturing (WAAM) process is regarded as one of the most advanced additive manufacturing processes. The process parameters play an extremely influential role to produce the desired dimensional accuracy, surface finish, and overall process stability. Hence, accurate determination of a suitable combination of the process parameters stands extremely crucial. In this paper, adaptive neuro-fuzzy inference system-based models have been developed in order to achieve a bi-directional predictive capability for a set of 3 inputs and 12 responses. The models have been trained and tested with the additional data generated from statistical regression applied on experimental data. The developed model has been compared with artificial neural network-based model, a technique that had been widely employed in the past literature. The R-squared values of the training samples and mean absolute percentage errors of the test samples for each response have been found to be quite satisfactory. With this approach both forward and backward mappings have been successfully achieved.Keywordswire and arc additive manufacturingcold metal transferregressionforward and backward modellingadaptive neuro-fuzzy inference system