Geometry-Aware Latent Representation Learning for Drag Coefficient Prediction of Irregular Particles
Abstract
Predicting the drag coefficient Cd of irregular particles requires representing their complex three-dimensional geometry in a form suitable for predictive models. Empirical drag correlations and many data-driven models represent particle geometry through a small set of hand-crafted shape descriptors. Geometric differences omitted by these descriptors are unavailable to the predictor. We propose a variational autoencoder-multilayer perceptron (VAE-MLP) framework that learns a latent representation directly from voxelized particle geometry. Geometry reconstruction and drag coefficient prediction jointly guide the learning of this representation. We compare VAE-MLP with a descriptor-based MLP (Descriptor-MLP) and established empirical correlations using CFD data for synthetic particle geometries at Reynolds numbers of 10--200. When the evaluated geometric variations are represented in the training data, the two learned models achieve comparable accuracy for regular ellipsoids, while VAE-MLP yields lower errors for irregular particles. When high-order surface perturbations are excluded from training, VAE-MLP achieves lower overall mean absolute relative error in all five training runs. At the strongest tested perturbation, the mean errors across these runs are 4.97% for VAE-MLP and 9.88% for Descriptor-MLP. These results support learning geometric representations directly from particle geometry to improve drag prediction for the irregular geometries and unseen surface perturbations examined here.
Related
- Interacting particles in random irregular media
- PAR2COX: Survival-Informed Tensor Decomposition for Phenotyping and Risk Prediction from Irregular Longitudinal Data
- Boundary Geometry Determines Optimal Shallow ReLU Approximation under Weighted Coefficient Constraints
- Rethinking Representation Geometry in Supervised Contrastive Learning under Long-Tailed Data
- What Converges in the Platonic Representation Hypothesis? Structure over Geometry
BibTeX
@misc{li2026geometryaware,
title = {{Geometry-Aware Latent Representation Learning for Drag Coefficient Prediction of Irregular Particles}},
author = {Zihao Li and Shota Deguchi and Mitsuteru Asai and Shunhua Chen and Naoto Mitsume},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.23087},
eprint = {2610.23087},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.23087},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Geometry-Aware Latent Representation Learning for Drag Coefficient Prediction of Irregular Particles
AU - Zihao Li
AU - Shota Deguchi
AU - Mitsuteru Asai
AU - Shunhua Chen
AU - Naoto Mitsume
PY - 2026
DA - 2026/10/16
PB - arXiv
DO - 10.48550/arxiv.2610.23087
UR - https://arxiv.org/abs/2610.23087
AB - Predicting the drag coefficient C_d of irregular particles requires representing their complex three-dimensional geometry in a form suitable for predictive models. Empirical drag correlations and many data-driven models represent particle geometry through a small set of hand-crafted shape descriptors. Geometric differences omitted by these descriptors are unavailable to the predictor. We propose a variational autoencoder-multilayer perceptron (VAE-MLP) framework that learns a latent representation directly from voxelized particle geometry. Geometry reconstruction and drag coefficient prediction jointly guide the learning of this representation. We compare VAE-MLP with a descriptor-based MLP (Descriptor-MLP) and established empirical correlations using CFD data for synthetic particle geometries at Reynolds numbers of 10--200. When the evaluated geometric variations are represented in the training data, the two learned models achieve comparable accuracy for regular ellipsoids, while VAE-MLP yields lower errors for irregular particles. When high-order surface perturbations are excluded from training, VAE-MLP achieves lower overall mean absolute relative error in all five training runs. At the strongest tested perturbation, the mean errors across these runs are 4.97% for VAE-MLP and 9.88% for Descriptor-MLP. These results support learning geometric representations directly from particle geometry to improve drag prediction for the irregular geometries and unseen surface perturbations examined here.
ER -