A Transfer Learning-Based Multi-Fidelity Point-Cloud Neural Network Approach for Melt Pool Modeling in Additive Manufacturing
ASCE-ASME Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering, 8(1), 011104
*Corresponding author
Cite
@article{10.1115/1.4051749,
author = {Huang, Xufeng and Xie, Tingli and Wang, Zhuo and Chen, Lei and Zhou, Qi and Hu, Zhen},
title = "{A Transfer Learning-Based Multi-Fidelity Point-Cloud Neural Network Approach for Melt Pool Modeling in Additive Manufacturing}",
journal = {ASCE-ASME J Risk and Uncert in Engrg Sys Part B Mech Engrg},
volume = {8},
number = {1},
year = {2022},
month = {08},
abstract = "{Melt pool modeling is critical for model-based uncertainty quantification (UQ) and quality control in metallic additive manufacturing (AM). Finite element (FE) simulation for thermal modeling in metal AM, however, is tedious and time-consuming. This paper presents a multifidelity point-cloud neural network method (MF-PointNN) for surrogate modeling of melt pool based on FE simulation data. It merges the feature representations of the low-fidelity (LF) analytical model and high-fidelity (HF) FE simulation data through the theory of transfer learning (TL). A basic PointNN is first trained using LF data to construct a correlation between the inputs and thermal field of analytical models. Then, the basic PointNN is updated and fine-tuned using the small size of HF data to build the MF-PointNN. The trained MF-PointNN allows for efficient mapping from input variables and spatial positions to thermal histories, and thereby efficiently predicts the three-dimensional melt pool. Results of melt pool modeling of electron beam additive manufacturing (EBAM) of Ti-6Al-4V under uncertainty demonstrate the efficacy of the proposed approach.}",
issn = {2332-9017},
doi = {10.1115/1.4051749},
url = {https://doi.org/10.1115/1.4051749},
note = {011104},
eprint = {https://asmedigitalcollection.asme.org/risk/article-pdf/8/1/011104/6736149/risk\_008\_01\_011104.pdf},
}
Abstract
Melt pool modeling is critical for model-based uncertainty quantification (UQ) and quality control in metallic Additive Manufacturing (AM). Finite element (FE) simulation for thermal modeling in metal AM, however, is tedious and time-consuming. This paper presents a multi-fidelity point-cloud neural network method (MF-PointNN) for surrogate modeling of melt pool based on FE simulation data. It merges the feature representations of low-fidelity (LF) analytical model and high-fidelity (HF) FE simulation data through the theory of transfer learning (TL). A basic PointNN is firstly trained using LF data to construct correlation between the inputs and thermal field of analytical models. Then, the basic PointNN is updated and fine-tuned using the small size of HF data to build the MF-PointNN. The trained MF-PointNN allows for efficient mapping from input variables and spatial positions to thermal histories, and thereby efficiently predict the three-dimensional melt pool. Results of melt pool modeling of electron beam additive manufacturing (EBAM) of Ti-6Al-4V under uncertainty demonstrate the efficacy of the proposed approach.