AI-Driven Process Optimization in Wire Arc Additive Manufacturing of Nickel Alloys Using ANFIS

Wire Arc Additive Manufacturing (WAAM) is gaining significance in recent years as a process for additive manufacturing that produces large and complex components with high deposition rates at economically feasible rates. Nickel-based alloys, praised for their unique features, like mechanical strength, corrosion resistance, and thermal stability, find widespread application in the aerospace, marine, and energy sectors. However, achieving the desired surface finish, dimensional precision, and mechanical performance of WAAM nickel alloy structures is a continuing challenge because of the complex interactions occurring between the process parameters. The study presents an experimental analysis conducted to establish the effect of important process parameters of WAAM-wire feed rate, travel speed, welding current, interpass temperature-on the quality indicators such as surface roughness, dimensional accuracy, and hardness. For the purpose of prediction and process control, the Adaptive Neuro-Fuzzy Inference System (ANFIS) model was developed from the experimental data set. The ANFIS model was able to provide good predictions for the outputs, thereby demonstrating its applicability as a valuable decision-support system in process optimization. The proposed framework constitutes a tangible interface through which manufacturers can obtain performance predictions and optimize process parameters, enhancing reliability and productivity of WAAM for nickel alloy applications.