Efficient Prediction of Thermal History in Wire Arc Additive Manufacturing Through Spatially Varying Functional Models

Abstract

Wire-arc additive manufacturing (WAAM) is a promising technology for fabrication of large-size structural components across various industrial sectors. WAAM offers relatively low equipment and material costs, high production rates, and a large fabrication area. However, high deposition temperatures can introduce significant thermal stresses, leading to substantial distortion and roughness in deposited layers. Precise thermal history prediction is essential for achieving the desired mechanical properties and geometric quality. While finite element methods (FEM) are commonly used, their reliance on solving large PDEs imposes significant computational demands, limiting application to small parts.In this paper, we develop a generalized functional regression model as a surrogate for thermal history prediction during single layer, single track deposition of Stainless Steel 316L in WAAM. Peak energy density, time after deposition, and location are chosen as predictors for the process. The interaction energy density and time after deposition is modeled using tensor-product basis. Thermal history variations between cross-section locations are modeled using Gaussian Process-based varying coefficients. FEM simulations are used to generate thermal history for two geometries at three energy density levels. We studied the effect of track geometry on the FEM results, the accuracy of the proposed surrogate methodology for predictions of the thermal history profile for each geometry, as well as the capability of the results to generalize to heterogeneous geometries.