Research On Micro-vibration Measurement Technology Based On Machine Vision

Vibration is an important indicator of mechanical equipment during operational conditions.Monitoring and analyzing it is conducive to ensuring the normal operation of the equipment and detecting potential hazards.In recent years,with the continuous development of image processing technology and the reduction of hardware costs,machine vision-based vibration measurement has become a research hot spot in the field of vibration measurement.However,in recent researches,there are still some shortcomings exist.Firstly,in machine condition monitoring,micro-vibration measurement are still in great demand in terms of researches and real engineering practice.Secondly,most of the researches are focused on measuring at a certain point of the structure,and the advantages of multiple points synchronous measurement of machine vision are not fully utilized,and it is difficult to reflect the overall vibration of the structure.In this work,emphasis are put on the measurement and visualization technology of micro-vibration,and the key contributions and research content are as follows,(1)Two machine vision-based vibration measurement methods are introduced,namely optical flow-based vibration measurement and phase-based vibration measurement,and a video capture system is built using industrial cameras,light sources,and other equipment.Micro-vibration measurements on a cantilever beam testrig are conducted,and the experimental results show that both methods can effectively measure micro-vibrations.(2)A machine vision-based bearing seat holo-spectrum analysis method is proposed to solve the problems of cumbersome sensor installation and the need for at least two sensors to synchronize measurement in traditional holo-spectrum measurement.The vibration in the horizontal and vertical directions was simultaneously obtained using the phase-based vibration measurement method,which simplified the holo-spectrum measurement process.Three holo-spectrum analysis with different rotor conditions from the machinery fault simulator were conducted.The holo-spectrum obtained through sensor measurement data and machine vision are compared,and the differences between the two methods are analyzed.(3)To address the problem of difficulty in directly observing the overall vibration of a structure,micro-vibration visualization technology is studied,and two video-based amplification methods are introduced and their amplification effects are compared and discussed.Amplification experiments are conducted through micro-vibrations measured from a cantilever beam and a bearing seat respectively,and the experimental results show that the video amplification algorithm can effectively visualize micro-vibrations in the selected frequency band,which can be used for structure modal vibration mode detection and determination of system anisotropy.(4)To address the problem that two-dimensional vibration measurement cannot fully reflect the spatial vibration characteristics,the optical flow method is combined with binocular vision theory to realize three-dimensional vibration measurement based on machine vision.The feasibility and accuracy of this method were verified through microvibration measurement experiments on the cantilever beam testrig.Finally,this method was applied to the medical device spatial motion evaluation and successfully measured the end trajectory of an electric stapler.