Rotating shaft radial vibration measurement system and method based on linear array vision and projection fringes

Abstract

The present invention relates to a system and method for measuring the radial vibration of a rotating shaft based on linear array vision and projected stripes. The system comprises a projection lamp, a linear array camera, and an image processing module. The projection lamp projects variable-density stripes onto the surface of the rotating shaft counterweight plate to sense the radial vibration information of the rotating shaft; the linear array camera captures the modulated stripe image and transmits it to the image processing module; the image processing module extracts the vibration information of the rotating shaft from the stripe features to measure the displacement of the radial vibration of the rotating shaft. The present invention can measure high-frequency, multi-dimensional vibration of the rotating shaft, has a simple system structure, a large measurement frequency range, and a small amount of collected data, while achieving both measurement efficiency and accuracy, providing a new method for measuring the radial vibration of the rotating shaft of rotating machinery.

Description

Technical Field

[0001] The present invention relates to the technical field of vision-based vibration measurement, and in particular to a system and method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes. Background Art

[0002] Due to the inevitable errors in manufacturing and assembly, rotating machinery is often accompanied by vibration during operation. Vibration signals contain a wealth of information about the machinery's operation. Therefore, vibration measurement plays a crucial role in monitoring the condition of rotating machinery. Diagnosing the operating status of machinery using vibration signals is an important means of ensuring structural safety. The vibration characteristics of rotating structures can reflect their operating status and provide early warning of potential failure risks. Therefore, accurate vibration measurement technology is of great significance for the health monitoring of rotating machinery.

[0003] With the development of computer science and imaging technology, machine vision-based methods have received widespread attention in the field of vibration measurement. However, current visual measurement methods generally suffer from complex processing procedures, low measurement efficiency, and limited vibration frequency. Summary of the Invention

[0004] The purpose of the present invention is to provide a system and method for measuring the radial vibration of a rotating shaft based on linear array vision and projected fringes. The system has a simple structure, a large measurement frequency range, a small amount of collected data, and has both measurement efficiency and accuracy, providing a new method for measuring the radial vibration of a rotating shaft of a rotating machine.

[0005] To achieve the above objectives, the technical solution of the present invention is: a radial vibration measurement system for a rotating shaft based on linear array vision and projected stripes, comprising a projection lamp, a linear array camera, and an image processing module; the projection lamp generates a stripe pattern through a prefabricated light sheet and projects it onto the surface of the counterweight plate of the rotating shaft; the linear array camera captures the stripe intensity signal modulated by the curved surface, splices the stripe intensity signals of each frame into a two-dimensional stripe image in the order of acquisition time, and transmits it to the image processing module; the image processing module extracts the vibration information of the rotating shaft from the stripe change characteristics based on the information transmitted by the linear array camera, so as to realize the displacement measurement of the radial vibration of the rotating shaft.

[0006] The present invention also provides a method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes using the above-mentioned system, comprising the following steps:

[0007] Step S1: Perform imaging processing on the surface of the counterweight plate to improve the projection quality of the stripes; adjust the position and projection direction of the projection lamp to project the stripes horizontally to the measurement position; adjust the focal length of the projection lamp lens so that the projected stripes are clearly and completely displayed on the surface of the counterweight plate;

[0008] Step S2: Set the imaging range of the line scan camera to the position to be measured on the rotating shaft and secure the camera; set important camera parameters including exposure time and acquisition line rate according to the requirements of the measurement scenario; adjust the camera's focal length, aperture, and magnification so that the stripes are clearly imaged at the center of the line scan sensor of the line scan camera;

[0009] Step S3: transmitting the fringe intensity signal collected by the linear array camera to the image processing module, and using the image processing algorithm to recover the radial vibration information of the shaft to be measured from the fringe feature information.

[0010] In one embodiment of the present invention, step S3 is specifically as follows:

[0011] Step S31: Using the first frame of fringe intensity signals as a reference signal, the reference signal is linearly scaled within a fixed signal length using a resampling algorithm to simulate the fringe intensity signal collected when the shaft is linearly displaced in the horizontal direction. The scaled fringe signals of each row are then stacked to form a mother wavelet fringe image.

[0012] Step S32: performing a correlation coefficient calculation on the first frame fringe intensity signal collected by the linear array camera and the mother wave fringe to obtain a correlation coefficient curve of the first frame fringe;

[0013] Step S33: Processing the peak coordinates of the correlation coefficient curve using a peak search algorithm to improve the positioning accuracy of the peak coordinates from the pixel level to the sub-pixel level, and then calculating the relative position of the rotation axis in the horizontal direction based on the mathematical relationship between the coordinates and the position of the rotation axis;

[0014] Step S34: repeat step S32 to calculate the correlation coefficient curve between the frame-by-frame fringe intensity curve and the mother wavelet fringe;

[0015] Step S35: Repeat step S33, and process the peak coordinates of the correlation coefficient curve of each frame using a peak search algorithm to obtain the relative spatial position of the rotating shaft at each moment in the horizontal direction, so as to restore the vibration displacement curve of the rotating shaft in the horizontal direction;

[0016] Step S36: Select the first frame of fringe intensity signal collected as the reference signal, calculate the autocorrelation curve of the reference signal through autocorrelation operation, and then calculate the cross-correlation curve between the reference signal and the fringe intensity of the remaining frames;

[0017] Step S37: Use a peak search algorithm to accurately locate the peak coordinates of each related curve, and finally convert the change of the peak coordinates into a vibration displacement curve of the shaft along the vertical direction according to a mathematical relationship.

[0018] In one embodiment of the present invention, the mother wave fringes are generated based on a resampling algorithm. The first frame fringe signal is selected as a reference and the fringe magnification factor is set. First, the length of the fringes is amplified to a specified multiple through an upsampling algorithm to obtain multiple groups of amplified fringes. The amplified fringes are processed to simulate the fringes recorded under linear displacement in actual measurement. Finally, the processed fringes are restored to their original length through a downsampling algorithm for subsequent correlation coefficient calculation.

[0019] In one embodiment of the present invention, the projection pattern is a variable density stripe pattern with a linear frequency change within a unit length, and the direction of change of the stripe intensity is perpendicular to the axis of the rotating shaft.

[0020] In one embodiment of the present invention, the mathematical relationship for the displacement of the measured rotating shaft in the horizontal direction is:

[0021]

[0022] Where ΔX(n) is the horizontal displacement of the rotation axis of the nth frame, ΔZ(n) is the axial displacement of the rotation axis of the nth frame, θ is the angle between the projection axis and the rotation axis, M x is the true displacement scale factor in the horizontal direction, and is the pixel coordinate of the peak value of the correlation coefficient curve of the n1th and n2th frames, and Δp is the pixel coordinate difference between the peak value coordinates of the correlation coefficient curve of the n1th frame and the n2th frame.

[0023] In one embodiment of the present invention, the mathematical relationship for the displacement of the measured rotating shaft along the vertical direction is:

[0024] ΔY(n)=M y Δp'=M y (P y(n) -P y(1) )

[0025] Where ΔY(n) is the vertical displacement of the axis of rotation in the nth frame, M y is the true displacement proportional factor in the vertical direction, P y(n) is the peak coordinate of the cross-correlation curve between the n-th frame fringe and the reference fringe, P y(1) is the peak coordinate of the autocorrelation curve of the reference fringe, and Δp' is the pixel distance between the peak coordinates of the autocorrelation curve and the cross-correlation curve of the fringe of the nth frame.

[0026] In one embodiment of the present invention, the mathematical relationship of the peak search algorithm is:

[0027]

[0028] Where P is the peak coordinate obtained by the algorithm, G iis the peak energy of the processed signal, i represents G i The order of G i+j is the energy representing the amplitude of the (i+j)th point in the processed signal, and k represents the number of points involved in the search process.

[0029] Compared with the existing technology, the present invention has the following beneficial effects: the system structure of the present invention is simple, the measurement frequency range is large, the amount of collected data is small, and it has both measurement efficiency and accuracy, providing a new method for radial vibration measurement of rotating machinery shafts. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a schematic diagram of the device structure of an embodiment of the present invention;

[0031] Figure 2 This is a schematic diagram of the principle of calculating the horizontal displacement of a structure according to an embodiment of the present invention. (a) is the fringe intensity signal of each frame collected by the linear array camera; (b) is the mother wave fringe generated from the reference fringe signal; (c) is the correlation coefficient curve of the fringe intensity and the mother wave fringe shown by the red and blue dashed lines in (a); (d) is the correlation coefficient curve set composed of the correlation coefficient curves of each frame; (e) is the displacement curve measured along the horizontal direction of the rotating shaft;

[0032] Figure 3 : is a schematic diagram of the principle of calculating the vertical displacement of a structure according to an embodiment of the present invention; wherein (a) is the fringe intensity signal of each frame collected by the linear array camera; (b) is a set of correlation curves between the reference fringe intensity and the fringe intensity of the remaining frames; (c) is a spatial three-dimensional graph of the correlation curve set in Figure (b); (d) is the cross-correlation curve at the red dotted line and the blue dotted line in Figure (b); and (e) is the displacement curve measured along the vertical direction of the rotating shaft.

[0033] In the figure, 1-projection lamp, 2-rotating shaft, 3-line array camera, 4-computer, 5-counterweight plate. DETAILED DESCRIPTION

[0034] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0035] Please refer to Figure 1 The present invention provides a system and method for measuring the radial vibration of a rotating shaft based on linear array vision and projected fringes. The system comprises 1, a projection lamp 2, a rotating shaft 3, a linear array camera 4, a computer 5, and a counterweight plate. The projection lamp 1 projects a fringe pattern onto the surface of the counterweight plate 5. The linear array camera 3 continuously images and captures the fringes on the surface of the counterweight plate 5, transmitting the images to the computer 4 for storage and processing. An image processing module within the computer processes the captured fringe intensity signals to obtain information on the radial vibration of the rotating shaft.

[0036] In this embodiment, reference Figure 2, is the principle diagram for calculating the horizontal displacement of the structure. Figure 2 As shown in the figure, when the measured shaft has a horizontal displacement of Δx, the peak coordinates of the correlation coefficient curve in Figure (c) will change. The peak coordinates are then accurately located using the peak search algorithm. The vibration signal of shaft 2 along the horizontal direction can be obtained from the changes in the peak coordinates of the correlation coefficient curve of each frame of the fringe signal. The mathematical formula for calculating the horizontal displacement of shaft 2 is:

[0037]

[0038] Where ΔX(n) is the horizontal displacement of the rotation axis of the nth frame, ΔZ(n) is the axial displacement of the rotation axis of the nth frame, θ is the angle between the projection axis and the rotation axis, M x is the true displacement scale factor in the horizontal direction, and is the pixel coordinate of the peak value of the correlation coefficient curve of the n1th and n2th frames, and Δp is the pixel coordinate difference between the peak value coordinates of the correlation coefficient curve of the n1th frame and the n2th frame.

[0039] In this embodiment, reference Figure 3 , is the principle diagram for calculating the vertical displacement of the shaft. Figure 3 As shown in Figure 2, when the measured shaft has a vertical displacement of Δy, the peak coordinates of the cross-correlation curve shown in Figure (d) will change. First, the peak coordinates of each frame are extracted, and then the peak coordinates are corrected using the peak search algorithm to obtain the vibration signal of shaft 2 in the vertical direction. The mathematical formula for the displacement of shaft 2 in the vertical direction is:

[0040] ΔY(n)=M y Δp'=M y (P y(n) -P y(1) )

[0041] Where ΔY(n) is the vertical displacement of the axis of rotation in the nth frame, M y is the true displacement proportional factor in the vertical direction, P y(n) is the peak coordinate of the cross-correlation curve between the n-th frame fringe and the reference fringe, P y(1) is the peak coordinate of the autocorrelation curve of the reference fringe, and Δp' is the pixel distance between the peak coordinates of the autocorrelation curve and the cross-correlation curve of the fringe of the nth frame.

[0042] The above are preferred embodiments of the present invention. Any changes made according to the technical solution of the present invention, as long as the resulting functions and effects do not exceed the scope of the technical solution of the present invention, shall fall within the scope of protection of the present invention.

Claims

1. A method for measuring radial vibration of a rotating shaft based on linear array vision and projected fringes, characterized in that: The steps include: Step S1: Perform imaging processing on the surface of the counterweight plate to improve the projection quality of the stripes; adjust the position and projection direction of the projection lamp to project the stripes horizontally to the measurement position; adjust the focal length of the projection lamp lens so that the projected stripes are clearly and completely displayed on the surface of the counterweight plate; Step S2: setting the imaging range of the linear array camera to the position to be measured on the rotating shaft and fixing the camera; According to the needs of the measurement environment, set the important parameters of the camera, including exposure time and acquisition line frequency; adjust the camera's focal length, aperture, and magnification so that the stripes are clearly imaged in the center of the linear array sensor of the linear array camera; Step S3: transmitting the fringe intensity signal collected by the linear array camera to the image processing module, and using the image processing algorithm to recover the radial vibration information of the shaft to be measured from the fringe feature information; specifically: Step S31: Using the first frame of fringe intensity signals as a reference signal, the reference signal is linearly scaled within a fixed signal length using a resampling algorithm to simulate the fringe intensity signal collected when the shaft is linearly displaced in the horizontal direction. The scaled fringe signals of each row are then stacked to form a mother wavelet fringe image. Step S32: performing a correlation coefficient calculation on the first frame fringe intensity signal collected by the linear array camera and the mother wave fringe to obtain a correlation coefficient curve of the first frame fringe; Step S33: Processing the peak coordinates of the correlation coefficient curve using a peak search algorithm to improve the positioning accuracy of the peak coordinates from the pixel level to the sub-pixel level, and then calculating the relative position of the rotation axis in the horizontal direction based on the mathematical relationship between the coordinates and the position of the rotation axis; Step S34: repeat step S32 to calculate the correlation coefficient curve between the frame-by-frame fringe intensity curve and the mother wavelet fringe; Step S35: Repeat step S33, and process the peak coordinates of the correlation coefficient curve of each frame using a peak search algorithm to obtain the relative spatial position of the rotating shaft at each moment in the horizontal direction, so as to restore the vibration displacement curve of the rotating shaft in the horizontal direction; Step S36: Select the first frame of fringe intensity signal collected as the reference signal, calculate the autocorrelation curve of the reference signal through autocorrelation operation, and then calculate the cross-correlation curve between the reference signal and the fringe intensity of the remaining frames; Step S37: Use a peak search algorithm to accurately locate the peak coordinates of each related curve, and finally convert the change of the peak coordinates into a vibration displacement curve of the shaft along the vertical direction according to a mathematical relationship.

2. The method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes according to claim 1, characterized in that: The mother wave fringes are generated based on a resampling algorithm. The first frame of the fringes signal is selected as a reference and the fringes magnification factor is set. First, the length of the fringes is amplified to a specified multiple through an upsampling algorithm to obtain multiple groups of amplified fringes. The amplified fringes are processed to simulate the fringes recorded under linear displacement in actual measurement. Finally, the processed fringes are restored to their original length through a downsampling algorithm for subsequent correlation coefficient calculation.

3. The method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes according to claim 1, characterized in that: The projection pattern is a variable density fringe pattern with linear frequency variation within a unit length, and the direction of variation of the fringe intensity is perpendicular to the axis of the rotating shaft.

4. The method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes according to claim 1, characterized in that: The mathematical relationship of the displacement of the measured shaft in the horizontal direction is: Where ΔX(n) is the horizontal displacement of the rotation axis of the nth frame, ΔZ(n) is the axial displacement of the rotation axis of the nth frame, θ is the angle between the projection axis and the rotation axis, M x is the true displacement scale factor in the horizontal direction, and is the pixel coordinate of the peak value of the correlation coefficient curve of the n1th and n2th frames, and Δp is the pixel coordinate difference between the peak value coordinates of the correlation coefficient curve of the n1th frame and the n2th frame.

5. The method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes according to claim 1, characterized in that: The mathematical relationship of the displacement of the measured shaft in the vertical direction is: ΔY(n)=M y Δp'=M y (P y(n) -P y(1) ) Where ΔY(n) is the vertical displacement of the axis of rotation in the nth frame, M y is the true displacement proportional factor in the vertical direction, P y(n) is the peak coordinate of the cross-correlation curve between the n-th frame fringe and the reference fringe, P y(1) is the peak coordinate of the autocorrelation curve of the reference fringe, and Δp' is the pixel distance between the peak coordinates of the autocorrelation curve and the cross-correlation curve of the fringe of the nth frame.

6. The method for measuring radial vibration of a rotating shaft based on linear array vision and projection fringes according to claim 1, characterized in that: The mathematical relationship of the peak search algorithm is: Where P is the peak coordinate obtained by the algorithm, G i is the peak energy of the processed signal, i represents G i The order of G i+j is the energy representing the amplitude of the i+jth point in the processed signal, and k represents the number of points involved in the search process.

7. A system for measuring the radial vibration of a rotating shaft based on linear array vision and projection fringes, which implements the method for measuring the radial vibration of a rotating shaft based on linear array vision and projection fringes as claimed in claim 1, characterized in that: The system includes a projection lamp, a linear array camera, and an image processing module. The projection lamp generates a fringe pattern through a prefabricated light sheet and projects it onto the counterweight plate surface of the rotating shaft. The linear array camera captures the fringe intensity signal modulated by the curved surface, splices each frame of the fringe intensity signal into a two-dimensional fringe image in the order of acquisition time, and transmits it to the image processing module. The image processing module extracts the vibration information of the rotating shaft from the fringe change characteristics based on the information transmitted by the linear array camera, thereby realizing the displacement measurement of the radial vibration of the rotating shaft.

Images