Visualization of Frequency-Varying Mechanical Vibrations via Automatic Resonance Tuning
IEEE Sensors Journal · 15.05.2026 · 26 (10) · 16286-16296
DOI: 10.1109/jsen.2026.3680944 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 15.05.2026. Verfasst von T. Yoshida und 4 weiteren Autor:innen.
Abstract (Originalsprache)
On-site fault detection in rotating machinery is essential for preventive maintenance. This article presents a user-friendly diagnostic system that enables visual inspection of vibrations and evaluates machine severity based on vibration magnitude. The proposed method is based on the mechanical amplification and visualization of microvibrations, which are imperceptible to the naked eye. Using mechanical resonance together with an optical lever effect, minute vibrations are converted into the macroscopic amplitude of the trajectory generated by the reciprocating motion of a reflected laser beam. Furthermore, a mechanism employing a sliding mass attached to a leaf spring is introduced, which autonomously adjusts its position according to the external vibration frequency, tuning the system to a resonant state and extending the range of vibration frequencies that can be macroscopically visualized. Consequently, the vibration condition can be intuitively assessed on-site by observing the laser trajectory projected onto surrounding surfaces. A hardened steel leaf spring (dimensions: 5 $\times$ 0.3 $\times$ 50 mm; Young’s modulus: 100 MPa) with a 1.8 g sliding mass exhibited self-regulated behavior, in which the sliding mass showed perturbations near resonance or remained stationary depending on the vibration conditions under external rotational vibration $\theta(t)=\theta_{\mathrm{e}} \sin (2 \pi f t) \text {, where } \theta_{\mathrm{e}} \geq 1.4^{\circ} \text { and } f=45-55 \mathrm{~Hz}$ . This phenomenon is referred to as automatic quasi-resonance. A nonlinear model was developed to represent these experimentally observed behaviors. Measurable parameters were taken directly from the experiment, whereas friction-related parameters were optimized using a genetic algorithm (GA). Simulation and experimental results both confirmed perturbations on either side of the resonance point, with noise appearing only in the experimental data.
Zitieren
APA
Yoshida, T., Yamakawa, T., Kurihara, Y., Kobayashi, K., & Watanabe, K. (2026). Visualization of Frequency-Varying Mechanical Vibrations via Automatic Resonance Tuning. IEEE Sensors Journal, 26(10), 16286-16296. https://doi.org/10.1109/jsen.2026.3680944
Vancouver
Yoshida T, Yamakawa T, Kurihara Y, Kobayashi K, Watanabe K. Visualization of Frequency-Varying Mechanical Vibrations via Automatic Resonance Tuning. IEEE Sensors Journal. 2026;26(10):16286-16296. doi:10.1109/jsen.2026.3680944
BibTeX
@article{Yoshida2026,
author = {Yoshida, T. and Yamakawa, T. and Kurihara, Y. and Kobayashi, K. and Watanabe, K.},
title = {Visualization of Frequency-Varying Mechanical Vibrations via Automatic Resonance Tuning},
journal = {IEEE Sensors Journal},
year = {2026},
volume = {26},
number = {10},
pages = {16286--16296},
doi = {10.1109/jsen.2026.3680944}
}
EUVIMED