Weighted Unsupervised Domain Adaptation Considering Geometry Features and Engineering Performance of 3D Design Data

TL;DR

This paper introduces a bi-weighted unsupervised domain adaptation method that leverages geometry features and engineering performance data to improve deep learning predictions on 3D design data, reducing reliance on costly simulations.

Contribution

It proposes a novel bi-weighting strategy for domain adaptation that considers geometry and performance features, enhancing prediction accuracy for 3D engineering data in new domains.

Findings

Reduces prediction errors in 3D design data across domains.

Efficiently replaces finite element analysis in specific applications.

Improves robustness of deep learning models for engineering performance prediction.

Abstract

The product design process in manufacturing involves iterative design modeling and analysis to achieve the target engineering performance, but such an iterative process is time consuming and computationally expensive. Recently, deep learning-based engineering performance prediction models have been proposed to accelerate design optimization. However, they only guarantee predictions on training data and may be inaccurate when applied to new domain data. In particular, 3D design data have complex features, which means domains with various distributions exist. Thus, the utilization of deep learning has limitations due to the heavy data collection and training burdens. We propose a bi-weighted unsupervised domain adaptation approach that considers the geometry features and engineering performance of 3D design data. It is specialized for deep learning-based engineering performance predictions. Domain-invariant features can be extracted through an adversarial training strategy by using hypothesis discrepancy, and a multi-output regression task can be performed with the extracted features to predict the engineering performance. In particular, we present a source instance weighting method suitable for 3D design data to avoid negative transfers. The developed bi-weighting strategy based on the geometry features and engineering performance of engineering structures is incorporated into the training process. The proposed model is tested on a wheel impact analysis problem to predict the magnitude of the maximum von Mises stress and the corresponding location of 3D road wheels. This mechanism can reduce the target risk for unlabeled target domains on the basis of weighted multi-source domain knowledge and can efficiently replace conventional finite element analysis.