EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics-Informed Symbolic Regression Ensemble (PISRE) for surface roughness prediction in Ti-6Al-4V dry turning

Natália Vilas Boas Pappi Maciel, Alex Fernandes de Souza, Paulo Henrique da Silva Campos, Antônio Carlos Zambroni de Souza, Filipe Alves Neto Verri, Pedro Paulo Balestrassi

The International Journal of Advanced Manufacturing Technology · 2026-05 · 144 (3-4) · 2253-2277

DOI: 10.1007/s00170-026-17994-x ↗

Kurz gefasst

Fachartikel in der Zeitschrift The International Journal of Advanced Manufacturing Technology, online erschienen am 10.04.2026. Verfasst von Natália Vilas Boas Pappi Maciel und 5 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Universidade Federal De Itajubá.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Predicting surface roughness in titanium alloy machining remains challenging due to the nonlinear interactions among process parameters and the small datasets inherent to experimental campaigns. Physics-Informed Symbolic Regression Ensemble (PISRE) is a novel physics-informed ensemble model for Ti-6Al-4V roughness \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{Ra}$$\end{document} that integrates a physics layer with four log-optimized symbolic branches and an uncertainty-weighted mechanism for dynamic branch confidence. Experiments were conducted following a Central Composite Design yielding 19 runs spanning cutting velocity \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{V}_{\varvec{c}} \varvec{=} \varvec{104}$$\end{document}–\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{256}$$\end{document} m/min, feed rate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{f = 0.035}$$\end{document}–\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{0.385}$$\end{document} mm/rev, and depth of cut \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{a}_{\varvec{p}} \varvec{=} \varvec{0.03}$$\end{document}–\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{0.37}$$\end{document} mm. Leave-One-Out cross-validation showed that PISRE achieved RMSE = 0.443 µm and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{R}^{\varvec{2}} \varvec{=} \varvec{0.915}$$\end{document}, outperforming XGBoost (RMSE = 0.657, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{R}^{\varvec{2}} \varvec{=} \varvec{0.814}$$\end{document}) and Random Forest (RMSE = 0.716, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{R}^{\varvec{2}} \varvec{=} \varvec{0.778}$$\end{document}) by 32.6% and 38.1% in RMSE respectively. The optimised feed rate exponent (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{b} \varvec{\approx } \varvec{1.18}$$\end{document}) is physically consistent with Ti-6Al-4V turning literature, and Spearman correlation analysis confirmed feed rate as the sole significant predictor of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{Ra}$$\end{document} within the tested window (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{r}_{\varvec{s}} \varvec{=} \varvec{+0.937}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{p} \varvec{

Förderung

  • Universidade Federal De Itajubá

Zitieren

APA

Maciel, N. V. B. P., Souza, A. F. D., Campos, P. H. D. S., Souza, A. C. Z. D., Verri, F. A. N., & Balestrassi, P. P. (2026). Physics-Informed Symbolic Regression Ensemble (PISRE) for surface roughness prediction in Ti-6Al-4V dry turning. The International Journal of Advanced Manufacturing Technology, 144(3-4), 2253-2277. https://doi.org/10.1007/s00170-026-17994-x

Vancouver

Maciel NVBP, Souza AFD, Campos PHDS, Souza ACZD, Verri FAN, Balestrassi PP. Physics-Informed Symbolic Regression Ensemble (PISRE) for surface roughness prediction in Ti-6Al-4V dry turning. The International Journal of Advanced Manufacturing Technology. 2026;144(3-4):2253-2277. doi:10.1007/s00170-026-17994-x

BibTeX
@article{Maciel2026,
  author = {Maciel, Natália Vilas Boas Pappi and Souza, Alex Fernandes de and Campos, Paulo Henrique da Silva and Souza, Antônio Carlos Zambroni de and Verri, Filipe Alves Neto and Balestrassi, Pedro Paulo},
  title = {Physics-Informed Symbolic Regression Ensemble (PISRE) for surface roughness prediction in Ti-6Al-4V dry turning},
  journal = {The International Journal of Advanced Manufacturing Technology},
  year = {2026},
  volume = {144},
  number = {3-4},
  pages = {2253--2277},
  doi = {10.1007/s00170-026-17994-x}
}
Kostenlos starten →