EUVIMEDEUVIMED entdecken
Crossref

A two-branch physics-informed neural network for small-sample sedimentation prediction of magnetorheological fluids

Wei Zhou, Tianle Cui, Decai Li, Xin Qi, Chao Ma, Xueqi Wang, Qiusheng Shi, Changrong Liao, Lei Xie

Smart Materials and Structures · 02.11.2026

DOI: 10.1088/1361-665x/aebcbb ↗

Kurz gefasst

Fachartikel in der Zeitschrift Smart Materials and Structures, online erschienen am 02.11.2026. Verfasst von Wei Zhou und 8 weiteren Autor:innen. Gefördert von: Study on the Development of China's Modern Science and Technology Museum System.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Sedimentation tests of magnetorheological fluids (MRFs) under natural gravity are time-consuming, limiting the experimental coverage of particle concentrations and formulations. This study proposes a moving-boundary data-interpolated physics-informed network (MB-DIP) to jointly predict concentration-field evolution and mudline motion from a limited number of measured source samples and the known initial state of a target sample. The two-branch architecture combines a physical mudline prior with a moving-boundary-aligned concentration-field prior. Conservation-constrained field refinement and reliability-guided interface consistency refinement subsequently coordinate local particle transport, overall particle content, and interface motion. For baseline MRFs, samples with initial particle volume fractions of 7.5, 10, 15, and 32 vol% were used to predict sedimentation at 20 and 26 vol%. The mean absolute errors of the concentration field and mudline position were 2.04% and 1.96 mm, respectively, and the mean absolute percentage error of particle content was 2.83%. Ablation experiments demonstrated the contribution of physics-consistency refinement, while six-fold leave-one-source-concentration-out validation examined prediction performance under varying source-data coverage. Further validation used SiO₂-modified MRFs containing 28 vol% carbonyl iron particles and 0.25–1 wt% silica. MB-DIP captured changes in sedimentation behavior with silica content, whereas the theoretical predictions remained unchanged. Concentration-field and mudline mean absolute errors ranged from 0.620% to 1.872% and from 0.306 to 1.081 mm, respectively. These results support MB-DIP as a method for predicting MRF sedimentation under limited-data conditions, with potential to reduce the experimental effort required for formulation evaluation.

Förderung

  • Study on the Development of China's Modern Science and Technology Museum System

Zitieren

APA

Zhou, W., Cui, T., Li, D., Qi, X., Ma, C., Wang, X., Shi, Q., Liao, C., & Xie, L. (2026). A two-branch physics-informed neural network for small-sample sedimentation prediction of magnetorheological fluids. Smart Materials and Structures. https://doi.org/10.1088/1361-665x/aebcbb

Vancouver

Zhou W, Cui T, Li D, Qi X, Ma C, Wang X, et al. A two-branch physics-informed neural network for small-sample sedimentation prediction of magnetorheological fluids. Smart Materials and Structures. 2026. doi:10.1088/1361-665x/aebcbb

BibTeX
@article{Zhou2026,
  author = {Zhou, Wei and Cui, Tianle and Li, Decai and Qi, Xin and Ma, Chao and Wang, Xueqi and Shi, Qiusheng and Liao, Changrong and Xie, Lei},
  title = {A two-branch physics-informed neural network for small-sample sedimentation prediction of magnetorheological fluids},
  journal = {Smart Materials and Structures},
  year = {2026},
  doi = {10.1088/1361-665x/aebcbb}
}
Kostenlos starten →