Physics-informed machine learning surrogate for scalable simulation of thermal histories during wire-arc directed energy deposition

TL;DR

This paper develops a physics-informed neural network surrogate model to efficiently simulate thermal histories in wire-arc directed energy deposition, significantly reducing computational costs while maintaining accuracy for large-scale additive manufacturing.

Contribution

It demonstrates the scalability of PINNs for large-scale DED thermal simulations by optimizing collocation point sampling, enabling faster and resource-efficient predictions.

Findings

PINNs reduce simulation time by up to 98.6%.

The model maintains high accuracy with fewer training points.

Super-resolution capabilities enhance detailed thermal predictions.

Abstract

Wire-arc directed energy deposition (DED) has emerged as a promising additive manufacturing (AM) technology for large-scale structural engineering applications. However, the complex thermal dynamics inherent to the process present challenges in ensuring structural integrity and mechanical properties of fabricated thick walls and plates. While finite element method (FEM) simulations have been conventionally employed to predict thermal history during deposition, their computational demand remains prohibitively high for actual large-scale applications. Given the necessity of multiple repetitive simulations for heat management and the determination of an optimal printing strategy, FEM simulation quickly becomes entirely infeasible. Instead, advancements have been made in using trained neural networks as surrogate models for rapid prediction. However, traditional data-driven approaches necessitate large amounts of relevant and verifiable external data, during the training and validation of the neural network. Regarding large-scale wire-arc DED, none of these data sources are readily available in quantities sufficient for an accurate surrogate. The introduction of physics-informed neural networks (PINNs) has opened up an alternative simulation strategy by leveraging the existing physical knowledge of the phenomena with advanced machine learning methods. Despite their theoretical advantages, PINNs have seen limited application in the context of large-scale wire-arc DED for structural engineering. This study investigates the scalability of PINNs, focusing on efficient collocation points sampling, a critical factor controlling both the training time and model performance. Results show PINNs can reduce computational time and effort by up to 98.6%, while maintaining the desired accuracy and offering "super-resolution". Future directions for enhancing PINN performance in metal AM are discussed.