Temperature-Dependent Multiphysics Modeling of Additive Friction Stir Deposition Using Multi-Task Coupled Physics-Informed Neural Networks
Abstract
Additive friction stir deposition (AFSD) involves coupled thermal and material-flow fields governed by frictional heating, plastic deformation, and tool-imposed boundary conditions. High-fidelity finite-volume methods (FVMs) can resolve these fields but are computationally expensive, while constant-property approximations can distort the predicted thermo-mechanical response. This work develops a steady-state multi-task coupled physics-informed neural network (MCoPINN) to predict three-dimensional temperature and material-flow fields while reconstructing temperature-dependent thermal conductivity, density, and specific heat from sparse property measurements. A theoretical analysis decomposes prediction error into property reconstruction and neural field-solver contributions and characterizes the discrepancy associated with constant-property models. A controlled one-dimensional nonlinear-conduction study evaluates property reconstruction under sparse measurements. The framework is then evaluated for AFSD against a temperature-dependent FVM benchmark and experimental thermocouple measurements. MCoPINN reproduces the reference thermal and material-flow fields and reduces average temperature root-mean-square error from 50.13 K for the constant-property CoPINN to 30.28 K across four traverse speeds. MCoPINN requires approximately 8.5 hours of training per condition, compared with 52 hours for the FVM benchmark. These results demonstrate computationally efficient, physics-coupled prediction with temperature-dependent material properties.
Related
- A multiphysics-coupled DeepONet framework for predicting sodium dendrite growth in wide-temperature sodium-ion capacitors
- Integrated Population Balance and Multiphysics Modeling for Predicting Undesired Agglomeration in Small Molecule Manufacturing
- Additive solid immersion lenses for enhanced collection efficiency of shallow NV centers by pulsed laser deposition and structurization of high-k amorphous oxides
- Geometry and cooling effects on an induction self-heating insert for rapid mold temperature control: Experimental, Multiphysics, and artificial neural network-based analysis
- A Physics-based and Geometry-aware Reduced-order Framework for Accelerated Part-Scale Thermomechanical Simulation of Additive Friction Stir Deposition Process
BibTeX
@misc{gupta2026temperaturedependent,
title = {{Temperature-Dependent Multiphysics Modeling of Additive Friction Stir Deposition Using Multi-Task Coupled Physics-Informed Neural Networks}},
author = {Dhrubajyoti Gupta and Nikhil Gotawala and Raghav Gnanasambandam and Rohit Kannan and Jian Yu and Hang Z. Yu and Zhenyu James Kong},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.04082},
eprint = {2610.04082},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.04082},
}
RIS
TY - UNPB
TI - Temperature-Dependent Multiphysics Modeling of Additive Friction Stir Deposition Using Multi-Task Coupled Physics-Informed Neural Networks
AU - Dhrubajyoti Gupta
AU - Nikhil Gotawala
AU - Raghav Gnanasambandam
AU - Rohit Kannan
AU - Jian Yu
AU - Hang Z. Yu
AU - Zhenyu James Kong
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.04082
UR - https://arxiv.org/abs/2610.04082
AB - Additive friction stir deposition (AFSD) involves coupled thermal and material-flow fields governed by frictional heating, plastic deformation, and tool-imposed boundary conditions. High-fidelity finite-volume methods (FVMs) can resolve these fields but are computationally expensive, while constant-property approximations can distort the predicted thermo-mechanical response. This work develops a steady-state multi-task coupled physics-informed neural network (MCoPINN) to predict three-dimensional temperature and material-flow fields while reconstructing temperature-dependent thermal conductivity, density, and specific heat from sparse property measurements. A theoretical analysis decomposes prediction error into property reconstruction and neural field-solver contributions and characterizes the discrepancy associated with constant-property models. A controlled one-dimensional nonlinear-conduction study evaluates property reconstruction under sparse measurements. The framework is then evaluated for AFSD against a temperature-dependent FVM benchmark and experimental thermocouple measurements. MCoPINN reproduces the reference thermal and material-flow fields and reduces average temperature root-mean-square error from 50.13 K for the constant-property CoPINN to 30.28 K across four traverse speeds. MCoPINN requires approximately 8.5 hours of training per condition, compared with 52 hours for the FVM benchmark. These results demonstrate computationally efficient, physics-coupled prediction with temperature-dependent material properties.
ER -