Learning to See the Vibration: A Neural Network for Vibration Frequency Prediction.

Jiantao Liu,Xiaoxiang Yang

doi:10.3390/s18082530

Abstract

Vibration measurement serves as the basis for various engineering practices such as natural frequency or resonant frequency estimation. As image acquisition devices become cheaper and faster, vibration measurement and frequency estimation through image sequence analysis continue to receive increasing attention. In the conventional photogrammetry and optical methods of frequency measurement, vibration signals are first extracted before implementing the vibration frequency analysis algorithm. In this work, we demonstrate that frequency prediction can be achieved using a single feed-forward convolutional neural network. The proposed method is verified using a vibration signal generator and excitation system, and the result compared with that of an industrial contact vibrometer in a real application. Our experimental results demonstrate that the proposed method can achieve acceptable prediction accuracy even in unfavorable field conditions.

Highlights

Structural vibration measurement is widely used in civil engineering, mechanical engineering, and other engineering practices
The frequency components can be estimated from the measured vibration signal; the resonant frequency or the natural frequency can be determined from the signal
We do not explicitly analyze and extract vibration signals from the image sequence; rather, we propose a convolutional neural network (CNN) trained with generated artificial signals, and utilize the learned features to discriminate the pixel brightness variation in the time domain for vibration frequency prediction

Summary

Introduction

Structural vibration measurement is widely used in civil engineering, mechanical engineering, and other engineering practices. The frequency components can be estimated from the measured vibration signal; the resonant frequency or the natural frequency can be determined from the signal. The estimation of the resonant frequency or natural frequency of a mechanical system is of great importance in many engineering applications. The resonant frequency and natural frequency can be used in monitoring the mechanical behavior of important structural components, detecting variations in the mechanical properties, design optimization, avoiding the occurrence of resonance disasters, and so forth. Vibration measurement methods can be divided into two categories: contact measurement and non-contact measurement. Non-contact measurement utilizes certain types of electromagnetic radiation to transmit information, such as laser Doppler vibrometer (LDV) [1,2,3]

Methods

Discussion

Conclusion