Dynamic modeling of ball screw feed systems: A review and framework from physical mechanisms to a physics-data hybrid approach.
Saved in:
| Title: | Dynamic modeling of ball screw feed systems: A review and framework from physical mechanisms to a physics-data hybrid approach. |
|---|---|
| Authors: | Li, Yongqiang1,2 liyongqiang0215@gmail.com, Bin Mohamed, Saiful Bahri1, Bin Kasim, Mohd Shahir1, Binti Yusof, Siti Nurul Akmal1 |
| Source: | Sound & Vibration. 2026, Vol. 60 Issue 3, p1-28. 28p. |
| Subjects: | Dynamic models, Nonlinear theories, Thermal strain |
| Abstract: | Ball screw feed systems demonstrate that pronounced nonlinearity, time variation, and multiphysics coupling arise under high speed, high acceleration, and long-duration operation, where neither purely physics-based nor purely data-driven models can simultaneously ensure interpretability, deployability, and cross-condition reliability. However, the significant challenges this presents show that a more integrated framework could address these limitations more effectively. Moreover, the findings from existing approaches indicate that a Coupled Mechanism-Model-Data (CMMD) framework could provide a systematic way to connect physical mechanisms with application-oriented modeling through a hierarchical and constrained architecture. In light of these key results, the evidence shows that following a unified logic of "mechanism coupling-model structure-data updating interface" appears to support more coherent analytical development. Research shows friction, stiffness, thermal deformation, and wear affect modal properties. Nevertheless, the important propagation effects indicate that formulating these as identifiable nonlinear and time-varying terms demonstrates that the framework retains critical physical interpretability. Furthermore, the evidence shows that hierarchical physics-based model families appear to differ significantly in representational capability, computational cost, and boundary sensitivity. Given that the findings demonstrate that conditions involving strong nonlinearity and parameter drift present particular analytical difficulty, LPV-NARX (Linear Parameter-Varying Nonlinear AutoRegressive model with eXogenous inputs), sparse identification, neural networks, and three hybrid paradigms--residual compensation, adaptive parameter updating, and physics-constrained learning--indicate that synthesis according to fusion location, updating objects, and failure boundaries could establish a constraint-driven hierarchical selection strategy. Data shows hybrid paradigms yield representative application scenarios. [ABSTRACT FROM AUTHOR] |
| Copyright of Sound & Vibration is the property of Academic Publishing and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
Be the first to leave a comment!
