Prediction of bead geometry with consideration of interlayer temperature effect for CMT-based wire-arc additive manufacturing

Abstract

Cold metal transfer (CMT)–based wire-arc additive manufacturing (WAAM) is increasingly popular for the production of large and complex metallic components due to its high deposition rate, low heat input, and excellent material efficiency. The accurate prediction of the bead geometry is of great importance to enhance the stability of the process and its dimensional accuracy. Besides the wire feed speed (WFS) and travel speed (TS), the interlayer temperature is another key factor in determining the bead geometry because of the heat accumulation in the multilayer deposition. In this paper, considering the varying interlayer temperature, WFS, and TS as inputs, an artificial neural network (ANN) is developed to predict the bead width, height, and contact angle; then, by connecting the ANN model with a bead geometric model, a combined model is established to improve the ANN model. Based on experimental test data, with random combinations of input parameters, the combined model is demonstrated to be able to accurately predict the bead geometry (mean error < 5.1%). The general effect of interlayer temperature on the bead geometry was also investigated by experiment.