Laser Doppler vibrometer signal optimization method for torsional vibration measurement of transmission shaft system

Abstract

The invention provides a laser Doppler vibrometer signal optimization method for torsional vibration measurement of a transmission shaft system, and the method comprises the steps: estimating the limit rotation speed of a measured shaft based on the measured shaft in the transmission shaft system, and setting the range of a laser Doppler vibrometer; adjusting a laser Doppler vibration meter and determining a to-be-traversed offset point based on the measured shaft: collecting a vibration signal of the measured shaft based on the to-be-traversed offset point, and calculating an improved signal-to-noise ratio index; determining an optimal incident bias by means of interpolation based on the improved signal-to-noise ratio index; and acquiring a smooth torsional vibration signal for torsional vibration measurement of the transmission shaft system based on the optimal incident bias measurement vibration signal. The method is based on the Vold-Kalman filtering algorithm, the EEMD algorithm and the median filtering algorithm, the operation rate is high, the accuracy rate is high, and real-time rapid measurement of the torsional vibration of the shaft system can be assisted.

Description

Technical Field

[0001] The invention belongs to the technical field of non-contact torsional vibration measurement, in particular to a laser Doppler vibrometer signal optimization method for torsional vibration measurement of a transmission shaft system. Background Art

[0002] The transmission shaft system is generally composed of rotating parts such as shafts, bearings and gears, and is the core component of the power unit. Due to errors in the manufacturing and assembly of transmission components, the input and output torques of the shaft system also vary within a certain range. Under the action of internal and external excitation sources, the normal torsional vibration of the shaft system is usually also called torsional vibration, which is manifested as the rotation speed of each rotating component in the system periodically fluctuating around the nominal value. When subjected to abnormal excitation such as component degradation, the torsional vibration response of the system changes significantly, such as the appearance of periodic pulses and a significant increase in the amplitude of a specific period. Measuring and identifying abnormal torsional vibration responses can predict transmission shaft system failures and ensure reliable operation of the system. This is of great significance for large-scale equipment that has been in service for a long time, such as the main transmission chain of wind power generation and the reducer of nuclear power cooling water circulation pumps.

[0003] The most widely used torsional vibration measurement technology currently uses speed measuring gears and electromagnetic / optical displacement sensors. Due to the limitations of the number of teeth, manufacturing and assembly accuracy, and the frequency response of the sensor, the resolution of transient pulse events at low and medium speeds is insufficient, making it difficult to accurately identify abnormal conditions. The heterodyne laser Doppler vibrometer (LDV) has the advantages of high resolution and wide frequency response, and has been applied to scenarios such as structural health monitoring and modal analysis. However, LDV is still relatively rarely used in the measurement of rotating shaft torsional vibration. The main problem is that the signal-to-noise ratio is relatively low when measuring the surface of the rotating shaft. The reason is that the single-channel LDV measures the vibration velocity that is colinear with the laser. Therefore, in the torsional vibration measurement, the laser needs to deviate from the center of rotation and be incident on the curved shaft surface to obtain the linear velocity component of the shaft surface. The magnitude of this component is proportional to the distance that the incident light deviates from the axis. Theoretically, it reaches the maximum value when the laser is incident along the tangent of the shaft surface. At this time, the curved shaft surface seriously affects the focusing state of the light spot and the collection effect of scattered light. In addition, the translation, pitch, deflection and rotation of the shaft cause the laser speckle pattern to change all the time. Under the combined effect of the above factors, the sudden change of the light signal intensity will cause pulse noise in the demodulated signal, which is called speckle noise. Some researchers have designed a rotation speed measurement system based on laser Doppler velocimetry, but have not involved the evaluation and control of speckle noise. Some researchers have proposed a method for suppressing colored noise and random noise based on post-processing algorithms, but the spectral method is difficult to effectively reduce pulse noise. The current LDV generally only uses the intensity of the light signal as the basis for adjustment, which is also difficult to meet the actual monitoring needs.

[0004] Therefore, in order to control the intensity of speckle noise, it is urgent and necessary to seek a signal optimization method for laser Doppler vibrometer used for transmission shaft torsional vibration measurement to achieve real-time and rapid measurement of shaft torsional vibration. Summary of the invention

[0005] In view of the defects in the above-mentioned prior art, the present invention proposes a method for optimizing the signal of a laser Doppler vibrometer for measuring the torsional vibration of a transmission shaft system. The method includes estimating the limit speed of the measured shaft and setting the range of the laser Doppler vibrometer based on the measured shaft in the transmission shaft system; adjusting the laser Doppler vibrometer and determining the offset point to be traversed based on the measured shaft; collecting the vibration signal of the measured shaft based on the offset point to be traversed, and calculating the improved signal-to-noise ratio index; determining the optimal incident offset based on the improved signal-to-noise ratio index by means of interpolation; measuring the vibration signal based on the optimal incident offset to obtain a smooth torsional vibration signal for measuring the torsional vibration of the transmission shaft system. The present invention is based on the Vold-Kalman filtering algorithm, the EEMD algorithm and the median filtering algorithm, and has a fast computing speed and high accuracy. It can assist in the real-time and rapid measurement of the torsional vibration of the shaft system, reduce the dependence on the operator's experience, and has reliable operation and strong practicality.

[0006] The present invention provides a laser Doppler vibrometer signal optimization method for measuring torsional vibration of a transmission shaft system, which comprises the following steps:

[0007] S1. Based on the transmission shaft system, determine the limit speed of the shaft system to be measured and set the range of the laser Doppler vibrometer;

[0008] S2. Based on the measured axis system, adjust the laser Doppler vibrometer and determine the offset point to be traversed:

[0009] S3, based on the offset point to be traversed, collecting the vibration signal of the measured shaft system, and calculating the improved signal-to-noise ratio index;

[0010] S31. Based on the offset point to be traversed, collect the vibration signal u of the measured shaft system t ,

[0011] S32. Calculate the instantaneous speed ω of the nth offset point of the measured shaft system according to the relationship between the linear velocity and the angular velocity. t :

[0012] ω t =-u t / (nΔy) (1)

[0013] Wherein, Δy represents the feed step size of the offset adjustment;

[0014] S33, instantaneous phase function c of the vibration harmonics of the measured shaft system t for:

[0015] c t=2cos(ω t f r Δt) (2)

[0016] Wherein, Δt represents the vibration signal u t The sampling rate F s The reciprocal of r Indicates the first-order torsional vibration frequency of the measured shaft system;

[0017] S34, based on the instantaneous phase function c t , solve the vibration signal u t The harmonic components x of each order to be monitored t ;

[0018] S35, subtracting each order harmonic component x from the vibration signal t The sum of the residual signal res t , and for each order harmonic component x t sum and the residual signal res t Calculate the corresponding L2 / L1 norms respectively:

[0019]

[0020] Where k represents the order;

[0021] S36, normalizing the light intensity signal output by the laser Doppler vibrometer to obtain a basic signal-to-noise ratio SNR, and based on each order harmonic component x t sum The L2 / L1 norm and residual signal res t The L2 / L1 norm of , and the improved signal-to-noise ratio indicator ISNR is obtained:

[0022]

[0023] S4. Based on the improved signal-to-noise ratio index, determine the optimal incident bias by interpolation: establish a rectangular coordinate system with the cross-sectional view of the rotating shaft, take the height y of the bracket as the independent variable, take the improved signal-to-noise ratio index ISNR under each bias as the dependent variable, perform interpolation densification, and take the Mth position y corresponding to the maximum value of the improved signal-to-noise ratio index ISNR m is the optimal incident bias;

[0024] S5. Measure the vibration signal based on the optimal incident bias to obtain a smooth torsional vibration signal for measuring the torsional vibration of the transmission shaft system: measure the vibration signal u to be analyzed under the optimal incident bias condition m (t), design an adaptive window width median filter to suppress speckle noise and obtain a smooth torsional vibration signal The smoothed torsional vibration signal Used for torsional vibration measurement of transmission shafting.

[0025] Preferably, the step S5 specifically includes the following steps:

[0026] S51. Measure the vibration signal u to be analyzed under the optimal incident bias condition m (t), and based on step S34, obtain the torsional vibration signal u to be analyzed m The residual component res of (t) m (t);

[0027] S52, decomposing the residual component res based on an integrated empirical mode decomposition algorithm m (t), obtain the IMF of each order of intrinsic mode components i (t):

[0028]

[0029] Among them, AM i (t) and FM i (t) represent the amplitude modulation component and frequency modulation component of the eigenmode component respectively;

[0030] S53, determine the time when the impulse noise event occurs: for the first-order intrinsic mode component IMF1, use Hilbert transform to calculate the envelope signal AM1(t) of the first-order intrinsic mode component IMF1, and refer to the preset pulse amplitude detection threshold A min , determine the time when the impulse noise event occurs;

[0031] S54, estimating the noise impact interval: for the retrieved T impulse noise events, the actual width W of each impulse noise event is used t a times of the corresponding noise impact interval length, where a represents the impact interval length coefficient;

[0032] S55. For each noise impact interval, the length of The median filter is used to filter the residual component to obtain a smooth residual component, which is added to the harmonic component to obtain a smooth torsional vibration signal. The smoothed torsional vibration signal Used for torsional vibration measurement of transmission shaft system; where b represents the filter length coefficient.

[0033] Preferably, the step S2 specifically includes the following steps:

[0034] S21, adjusting the indicator light spot of the laser Doppler vibrometer: placing the indicator light spot of the laser Doppler vibrometer at a point on the surface of the measured shaft system, so that the measurement plane where the light beam of the laser Doppler vibrometer is located is coplanar with the cross section of the measured shaft system and avoids the low signal quality area of ​​the laser Doppler vibrometer;

[0035] S22, determine the upper limit of the offset adjustment: adjust the bracket height of the laser Doppler vibrometer through the servo control system, the indicator light spot of the laser Doppler vibrometer is close to the upper edge of the measured shaft system and focused, if the return light intensity RSSI read by the laser Doppler vibrometer is greater than or equal to 80%, then record the first offset distance Y u Used as a bias to adjust the upper boundary;

[0036] S23, determining the lower limit of the offset adjustment: starting the continuous measurement mode of the laser Doppler vibrometer, collecting the initial vibration signal x of the measured shaft system t , input the vibration signal into the control system of the bracket, take the mean value of the vibration signal within a small interval [-Δε, Δε] close to 0 as the target, adjust the height of the bracket through feedback control, and record the stable second offset distance Y d Used as a bias to adjust the lower boundary;

[0037] S24, determine the offset points to be traversed: set the feed step length Δy of the offset adjustment on the host computer, so that the set of offset points to be traversed is {Y u ,Y u -Δy,…,Y u -NΔy}, and satisfy Y u -NΔy≈Y d , where N represents the number of offset points to be traversed.

[0038] Preferably, the vibration signal u in step S34 t The harmonic components x of each order to be monitored t The solution equation is:

[0039]

[0040] Where m represents the sequence length; ε t Indicates the error in the filtering process; η t Indicates measurement error.

[0041] Preferably, the step S1 specifically includes the following steps:

[0042] S11. Calculate the limit speed V of the shaft system under test: Based on the transmission shaft system, determine the limit value of the surface linear speed of the shaft system under test, that is, the limit speed V;

[0043] S12. Based on the limiting rotational speed V of the measured shaft system, a measuring range of the laser Doppler vibrometer is set, wherein the measuring range of the laser Doppler vibrometer is greater than or equal to the limiting rotational speed V of the measured shaft system.

[0044] Preferably, the filtering bandwidth of the filter set in step S34 is the instantaneous speed ω of the measured shaft system.t The filter order is set to 2nd order, and the number of harmonic orders extracted is 10.

[0045] Preferably, the sampling rate F of the initial vibration signal in step S23 is s0 is greater than or equal to 25.6kHz; the sampling rate F of the vibration signal in step S31 s Greater than or equal to the first-order torsional vibration frequency f of the measured shaft system r The acquisition length of the vibration signal covers 15 to 30 cycles of the first-order torsional vibration harmonic of the measured shaft system.

[0046] Preferably, the interpolation function used for the interpolation in step S4 is a spline function and the interpolation step length is 0.2Δy.

[0047] Preferably, the number of decomposition modes of the integrated empirical mode decomposition algorithm in step S52 is 10, the number of integrations is 20, and the added noise amplitude index is 0.4σ, where σ represents the residual component res m (t) standard deviation; the pulse amplitude detection threshold A in step S53 min Take the 85% to 95% quantile of the envelope signal AM1(t); the influence interval length coefficient a in step S54 is 3 to 5; the filter length coefficient b in step S55 is an even number between 6 and 12

[0048] Compared with the prior art, the technical effects of the present invention are:

[0049] 1. The present invention proposes a laser Doppler vibrometer signal optimization method for measuring torsional vibration of a transmission shaft system. The proposed method is a non-contact vibration measurement method designed based on the LDV vibration measurement system. It is convenient for inspecting or monitoring rotating shaft equipment at the engineering site, and has simple operation and strong practicality.

[0050] 2. The present invention proposes a signal optimization method for a laser Doppler vibrometer for measuring torsional vibration of a transmission shaft system. The proposed method is based on the Vold-Kalman filtering algorithm, the EEMD algorithm and the median filtering algorithm. The method has a fast operation rate and is simple and efficient. It can assist in the real-time and rapid measurement of torsional vibration of the shaft system. The improved signal-to-noise ratio of the laser vibrometer is innovatively constructed. The proposed method avoids the limitations of manually setting the laser incident bias parameters and the smoothing filter parameters, and can improve the vibration measurement signal-to-noise ratio in different engineering application scenarios, reduce the dependence on the operator's experience, provide strong support for monitoring and diagnosis, and realize the automated measurement of torsional vibration of rotating shaft systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Other features, objects and advantages of the present application will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings.

[0052] Figure 1 It is a flow chart of a laser Doppler vibrometer signal optimization method for measuring torsional vibration of a transmission shaft system according to the present invention;

[0053] Figure 2 is a schematic diagram of a measurement method of a specific embodiment of the present invention;

[0054] Figure 3 is a graph of harmonic components of various orders according to a specific embodiment of the present invention;

[0055] Figure 4 is a residual signal curve diagram of a specific embodiment of the present invention;

[0056] Figure 5 is a bias-signal-to-noise ratio curve diagram obtained by interpolation and densification of each measuring point in a specific embodiment of the present invention;

[0057] Figure 6 is a diagram showing the smoothing and denoising results of a torsional vibration signal according to a specific embodiment of the present invention;

[0058] Figure 7 is a torsional vibration signal curve diagram of a specific embodiment of the present invention;

[0059] Figure 8 It is a curve diagram of the encoder torsional vibration signal processing result of a specific embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It will be appreciated that the specific embodiments described herein are only used to explain the relevant inventions, rather than to limit the inventions. It should also be noted that, for ease of description, only portions related to the relevant inventions are shown in the accompanying drawings. It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0061] Figure 1 The present invention shows a laser Doppler vibrometer signal optimization method for transmission shaft torsional vibration measurement, the method comprising the following steps:

[0062] S1. Based on the transmission shaft system, determine the limit speed of the shaft system to be measured and set the range of the laser Doppler vibrometer.

[0063] S11. Calculate the limit rotational speed V of the measured shaft system: Based on the transmission shaft system, determine the limit value of the surface linear speed of the measured shaft system, and calculate the limit rotational speed V of the measured shaft system.

[0064] S12. Based on the limiting rotational speed V of the measured shaft system, a measuring range of the laser Doppler vibrometer is set. The measuring range of the laser Doppler vibrometer is greater than or equal to the limiting rotational speed V of the measured shaft system.

[0065] S2. Based on the measured axis system, adjust the laser Doppler vibrometer and determine the offset point to be traversed.

[0066] S21. Adjust the indicator light spot of the laser Doppler vibrometer: place the indicator light spot of the laser Doppler vibrometer at a point on the surface of the measured shaft system. The measuring plane where the light beam of the laser Doppler vibrometer is located is coplanar with the cross section of the measured shaft system and avoids the low signal quality area of ​​the laser Doppler vibrometer.

[0067] S22, determine the upper limit of the offset adjustment: adjust the bracket height of the laser Doppler vibrometer through the servo control system, and the indicator light spot of the laser Doppler vibrometer is close to the upper edge of the measured shaft system and focused. If the return light intensity RSSI read by the laser Doppler vibrometer is greater than or equal to 80%, then record the first offset distance Y u Used as a bias to adjust the upper boundary.

[0068] S23, determine the lower limit of the offset adjustment: start the continuous measurement mode of the laser Doppler vibrometer and collect the initial vibration signal x of the measured shaft system t , input the vibration signal into the control system of the bracket, aiming at keeping the mean value of the vibration signal within a small interval [-Δε, Δε] close to 0, adjust the height of the bracket through feedback control, and record the stable second offset distance Y d As a bias to adjust the lower boundary. The sampling rate F of the initial vibration signal s0 Greater than or equal to 25.6kHz.

[0069] S24, determine the offset points to be traversed: set the feed step length Δy of the offset adjustment on the host computer, so that the set of offset points to be traversed is {Y u ,Y u -Δy,…,Y u -NΔy}, and satisfy Y u -NΔy≈Y d , where N represents the number of offset points to be traversed.

[0070] S3. Based on the offset points to be traversed, collect the vibration signal of the measured shaft system and calculate the improved signal-to-noise ratio index: At each offset point, collect the vibration signal u of the measured shaft system t , the sampling rate of the vibration signal F s Greater than or equal to the first-order torsional vibration frequency f of the measured shaft system r10k times, it is empirically set that the k-order frequency multiplication needs to be retained. According to the application practice of the sampling theorem in engineering, Fs ≥ 10 × kf r The acquisition length of the vibration signal should cover at least 15 to 30 cycles of the first-order torsional vibration harmonic of the measured shaft system to ensure the filtering quality. For the vibration signal of each offset point, the improved signal-to-noise ratio index is calculated based on Vold-Kalman filtering and L2 / L1 norm.

[0071] S31. Based on the offset point to be traversed, collect the vibration signal u of the measured shaft t .

[0072] S32. Calculate the instantaneous speed ω of the nth offset point of the measured shaft system according to the relationship between the linear velocity and the angular velocity. t :

[0073] ω t =-u t / (nΔy) (1)

[0074] S33, instantaneous phase function c of the vibration harmonic of the measured shaft t for:

[0075] c t =2cos(ω t f r Δt) (2)

[0076] Where Δt represents the vibration signal u t The sampling rate F s The reciprocal of r Indicates the first-order torsional vibration frequency of the measured shaft system.

[0077] S34, based on the instantaneous phase function c t According to the Vold-Kalman filtering principle, the vibration signal u is solved t The harmonic components x of each order to be monitored t :

[0078]

[0079] Where m represents the sequence length; ε t Indicates the error in the filtering process; η t Indicates measurement error.

[0080] In a specific embodiment, the filter bandwidth of the Vold-Kalman filter is set to the instantaneous speed ω of the measured shaft. t The average value of the Vold-Kalman filter is set to 2nd order, and the number of harmonic orders extracted is 10.

[0081] S35. Subtract each order harmonic component x from the vibration signal t The sum of the residual signal res t , and each order harmonic component x t sum and the residual signal res t Calculate the corresponding L2 / L1 norms respectively:

[0082]

[0083] Where k represents the order. The impulsive speckle noise is mainly retained in the residual signal.

[0084] S36, normalize the 0-5V light intensity signal output by the laser Doppler vibrometer to obtain the basic signal-to-noise ratio SNR, and based on the harmonic components x t sum The L2 / L1 norm and residual signal res t The L2 / L1 norm of , and the improved signal-to-noise ratio indicator ISNR is obtained:

[0085]

[0086] S4. Based on the improved signal-to-noise ratio index, the optimal incident bias is determined by interpolation: a rectangular coordinate system is established with the cross-sectional view of the rotating shaft, the height y of the bracket is taken as the independent variable (vertical direction), and the improved signal-to-noise ratio index ISNR under each bias is taken as the dependent variable, and interpolation densification is performed. The interpolation function used is the spline function, and the interpolation step size is preferably 0.2Δy. The Mth position y corresponding to the maximum value of the improved signal-to-noise ratio index ISNR is taken. m is the optimal incident bias.

[0087] S5. Measure the vibration signal based on the optimal incident bias to obtain a smooth torsional vibration signal for measuring the torsional vibration of the transmission shaft system: measure the vibration signal u to be analyzed under the optimal incident bias condition m (t), design an adaptive window width median filter to suppress speckle noise and obtain a smooth torsional vibration signal Smooth torsional vibration signal Used for torsional vibration measurement of transmission shafting.

[0088] S51. Measure the vibration signal u to be analyzed under the optimal incident bias condition m (t), and based on the Vold-Kalman filtering principle in step S34, the torsional vibration signal u to be analyzed is obtained m The residual component res of (t) m (t).

[0089] S52, based on the ensemble empirical mode decomposition (EEMD) algorithm, further decompose the residual component res m (t), obtain the IMF of each order of intrinsic mode components i (t)(Intrinsic modefunction, IMF):

[0090]

[0091] Among them, AM i (t) and FM i (t) represent the amplitude modulation component and frequency modulation component of the eigenmode component respectively.

[0092] The integrated empirical mode decomposition adds a certain amplitude of white noise to the decomposed signal, and takes the average of multiple empirical mode EMD decomposition results to improve the decomposition effect. The number of decomposition modes of the integrated empirical mode decomposition algorithm is 10, the number of integrations is 20, and the added noise amplitude index is 0.4σ, where σ represents the residual component res m The standard deviation of (t).

[0093] S53, determine the time when the impulse noise event occurs: for the first-order intrinsic mode component IMF1, use Hilbert transform to calculate the envelope signal AM1(t) of the first-order intrinsic mode component IMF1, and refer to the preset pulse amplitude detection threshold A min , determine the time when the impulse noise event occurs. Pulse amplitude detection threshold A min Take the 85% to 95% quantile of the envelope signal AM1(t).

[0094] S54, estimate the noise impact interval: for the retrieved T impulse noise events, take the actual width W of each impulse noise event t a times of the corresponding noise impact interval length, where a represents the impact interval length coefficient and is 3 to 5.

[0095] S55. For each noise impact interval, the length of The median filter is used to filter the residual component to obtain a smooth residual component, which is then added to the harmonic component to obtain a smooth torsional vibration signal. Smooth torsional vibration signal Used for torsional vibration measurement of transmission shaft system; where b represents the filter length coefficient and is an even number between 6 and 12.

[0096] In a specific embodiment, taking the test of an industrial planetary gear transmission shaft system test platform as an example, a planetary single gear tooth wear gearbox with a reduction ratio of 5:1 is used in this embodiment, and a high-precision rotary encoder installed on the output shaft is used to obtain a torsional vibration reference signal.

[0097] The whole system includes a laser Doppler vibrometer 1, a bracket 2, a host computer and an acquisition system 3, a drive motor 4, a planetary gearbox to be tested 5, an output shaft section to be tested 6, a simulated load 7, and Figure 1 As shown in the figure, 1 to 3 are the vibration test and processing equipment used, and 4 to 7 are the test benches.

[0098] In a specific embodiment, the radius of the measured output shaft is 25mm, and the maximum average rotational speed of the output shaft is about 2.513Hz, then the average linear velocity of the surface is about 395mm / s, and the range of the heterodyne laser Doppler vibrometer LDV is set to 500mm / s to meet the requirements. The distance from the light output point to the surface is set to ΔX≈450mm. The height of the bracket is adjusted by a servo control system such as an electric drive or hydraulic system. According to engineering experience, traversing 20 measuring points can meet the optimal requirements, that is, Δy is set to 25 / 20=1.25mm. In this example, the torsional vibration of the shaft system is mainly caused by gear meshing. Therefore, the gear meshing frequency is obtained by theoretical calculation as the first-order torsional vibration frequency f of the shaft system. r At a certain measuring point, the sum of the harmonic components of each order is obtained as Figure 3 As shown, the residual signal is Figure 4 A rectangular coordinate system is established with the cross-sectional view of the rotating shaft, the height y of the bracket is taken as the independent variable (vertical direction), and the improved signal-to-noise ratio index ISNR under each offset is taken as the dependent variable. For all offset measurement point data, a smooth y-ISNR curve is obtained using the interpolation function, and the interpolation step is set to 0.25 mm. The curve is plotted as shown in the figure. Figure 5 The offset-SNR curve shown in the figure takes the offset corresponding to the maximum ISNR value as the optimal incident offset. The actual offset is set to 6.8mm as shown in the figure, and the smoothing noise reduction effect of the median filter with adaptive window width is as follows Figure 6 The results of the shaft torsional vibration analysis obtained by the laser Doppler vibrometer are shown in Figure 7 As shown, Figure 8 The impact torsional vibration characteristics measured by the rotary encoders shown are basically consistent.

[0099] The present invention proposes a laser Doppler vibrometer signal optimization method for measuring torsional vibration of a transmission shaft system. The proposed method is a non-contact vibration measurement method designed based on an LDV vibration measurement system, which is convenient for patrol inspection or monitoring of rotating shaft system equipment at a construction site, and has simple operation and strong practicability. The proposed method is based on a Vold-Kalman filtering algorithm, an EEMD algorithm and a median filtering algorithm, and has a fast operation rate, is simple and efficient, and can assist in real-time and rapid measurement of torsional vibration of a shaft system. The improved laser vibrometer signal-to-noise ratio is innovatively constructed, and the proposed method avoids the limitations of manually setting laser incident bias parameters and smoothing filter parameters, and can improve the vibration measurement signal-to-noise ratio in different engineering application scenarios, reduce dependence on operator experience, provide strong support for monitoring and diagnosis, and realize automatic measurement of torsional vibration of a rotating shaft system.

[0100] Finally, it should be noted that the above embodiments are only intended to illustrate rather than limit the technical solutions of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the present invention can still be modified or replaced by equivalents. Any modification or partial replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for optimizing laser Doppler vibrometer signals for measuring torsional vibration of a transmission shaft system, characterized in that: It includes the following steps: S1. Based on the transmission shaft system, determine the limit speed of the shaft system to be measured and set the range of the laser Doppler vibrometer; S2. Based on the measured axis system, adjust the laser Doppler vibrometer and determine the offset point to be traversed: S3, based on the offset point to be traversed, collecting the vibration signal of the measured shaft system, and calculating the improved signal-to-noise ratio index; S31. Based on the offset point to be traversed, collect the vibration signal u of the measured shaft system t , S32. Calculate the instantaneous speed ω of the nth offset point of the measured shaft system according to the relationship between the linear velocity and the angular velocity. t : ωt=-ut / (nΔy) (1) Wherein, Δy represents the feed step size of the offset adjustment; S33, instantaneous phase function c of the vibration harmonics of the measured shaft system t for: c t =2cos(ω t f r Δt) (2) Wherein, Δt represents the vibration signal u t The sampling rate F s The reciprocal of r Indicates the first-order torsional vibration frequency of the measured shaft system; S34, based on the instantaneous phase function c t , solve the vibration signal u t The harmonic components x of each order to be monitored t ; S35, subtracting each order harmonic component x from the vibration signal t The sum of the residual signal res t , and for each order harmonic component x t sum and the residual signal res t Calculate the corresponding L2 / L1 norms respectively: Where k represents the order; S36, normalizing the light intensity signal output by the laser Doppler vibrometer to obtain a basic signal-to-noise ratio SNR, and based on each order harmonic component x t sum The L2 / L1 norm and residual signal res t The L2 / L1 norm of , and the improved signal-to-noise ratio indicator ISNR is obtained: S4. Based on the improved signal-to-noise ratio index, determine the optimal incident bias by interpolation: establish a rectangular coordinate system with the cross-sectional view of the rotating shaft, take the height y of the bracket as the independent variable, take the improved signal-to-noise ratio index ISNR under each bias as the dependent variable, perform interpolation densification, and take the Mth position y corresponding to the maximum value of the improved signal-to-noise ratio index ISNR m is the optimal incident bias; S5. Measure the vibration signal based on the optimal incident bias to obtain a smooth torsional vibration signal for measuring the torsional vibration of the transmission shaft system: measure the vibration signal u to be analyzed under the optimal incident bias condition m (t), design an adaptive window width median filter to suppress speckle noise and obtain a smooth torsional vibration signal The smoothed torsional vibration signal Used for torsional vibration measurement of transmission shafting.

2. The laser Doppler vibrometer signal optimization method for transmission shaft torsional vibration measurement according to claim 1, characterized in that: The step S5 specifically comprises the following steps: S51. Measure the vibration signal u to be analyzed under the optimal incident bias condition m (t), and based on step S34, obtain the torsional vibration signal u to be analyzed m The residual component res of (t) m (t); S52, decomposing the residual component res based on an integrated empirical mode decomposition algorithm m (t), obtain the IMF of each order of intrinsic mode components i (t): Among them, AM i (t) and FM i (t) represent the amplitude modulation component and frequency modulation component of the eigenmode component respectively; S53, determine the time when the impulse noise event occurs: for the first-order intrinsic mode component IMF1, use Hilbert transform to calculate the envelope signal AM1(t) of the first-order intrinsic mode component IMF1, and refer to the preset pulse amplitude detection threshold A min , determine the time when the impulse noise event occurs; S54, estimating the noise impact interval: for the retrieved T impulse noise events, the actual width W of each impulse noise event is used t a times of the corresponding noise impact interval length, where a represents the impact interval length coefficient; S55. For each noise impact interval, the length of The median filter is used to filter the residual component to obtain a smooth residual component, which is added to the harmonic component to obtain a smooth torsional vibration signal. The smoothed torsional vibration signal Used for torsional vibration measurement of transmission shaft system; where b represents the filter length coefficient.

3. The laser Doppler vibrometer signal optimization method for transmission shaft torsional vibration measurement according to claim 1, characterized in that: The step S2 specifically includes the following steps: S21, adjusting the indicator light spot of the laser Doppler vibrometer: placing the indicator light spot of the laser Doppler vibrometer at a point on the surface of the measured shaft system, so that the measurement plane where the light beam of the laser Doppler vibrometer is located is coplanar with the cross section of the measured shaft system and avoids the low signal quality area of ​​the laser Doppler vibrometer; S22, determine the upper limit of the offset adjustment: adjust the bracket height of the laser Doppler vibrometer through the servo control system, the indicator light spot of the laser Doppler vibrometer is close to the upper edge of the measured shaft system and focused, if the return light intensity RSSI read by the laser Doppler vibrometer is greater than or equal to 80%, then record the first offset distance Y u Used as a bias to adjust the upper boundary; S23, determining the lower limit of the offset adjustment: starting the continuous measurement mode of the laser Doppler vibrometer, collecting the initial vibration signal x of the measured shaft system t , input the vibration signal into the control system of the bracket, take the mean value of the vibration signal within a small interval [-Δε, Δε] close to 0 as the target, adjust the height of the bracket through feedback control, and record the stable second offset distance Y d Used as a bias to adjust the lower boundary; S24, determine the offset points to be traversed: set the feed step length Δy of the offset adjustment on the host computer, so that the set of offset points to be traversed is {Y u ,Y u -Δy,…,Y u -NΔy}, and satisfy Y u -NΔy≈Y d , where N represents the number of offset points to be traversed.

4. The laser Doppler vibrometer signal optimization method for transmission shaft torsional vibration measurement according to claim 1, characterized in that: The vibration signal u in step S34 t The harmonic components x of each order to be monitored t The solution equation is: Where m represents the sequence length; ε t Indicates the error in the filtering process; η t Indicates measurement error.

5. The laser Doppler vibrometer signal optimization method for transmission shaft torsional vibration measurement according to claim 1, characterized in that: The step S1 specifically includes the following steps: S11. Calculate the limit speed V of the shaft system under test: Based on the transmission shaft system, determine the limit value of the surface linear speed of the shaft system under test, that is, the limit speed V; S12. Based on the limiting rotational speed V of the measured shaft system, a measuring range of the laser Doppler vibrometer is set, wherein the measuring range of the laser Doppler vibrometer is greater than or equal to the limiting rotational speed V of the measured shaft system.

6. The method for optimizing laser Doppler vibrometer signal for transmission shaft torsional vibration measurement according to claim 1, characterized in that: The filtering bandwidth of the filter in step S34 is set to the instantaneous speed ω of the measured shaft system. t The filter order is set to 2nd order, and the number of harmonic orders extracted is 10.

7. The method for optimizing laser Doppler vibrometer signal for measuring torsional vibration of a transmission shaft system according to claim 3, characterized in that: The sampling rate F of the initial vibration signal in step S23 is s0 is greater than or equal to 25.6kHz; the sampling rate F of the vibration signal in step S31 s Greater than or equal to the first-order torsional vibration frequency f of the measured shaft system r The acquisition length of the vibration signal covers 15 to 30 cycles of the first-order torsional vibration harmonic of the measured shaft system.

8. The method for optimizing laser Doppler vibrometer signal for measuring torsional vibration of a transmission shaft system according to claim 1, characterized in that: The interpolation function used in the interpolation in step S4 is a spline function and the interpolation step length is 0.2Δy.

9. The laser Doppler vibrometer signal optimization method for transmission shaft torsional vibration measurement according to claim 2, characterized in that: The number of decomposition modes of the integrated empirical mode decomposition algorithm in step S52 is 10, the number of integrations is 20, and the added noise amplitude index is 0.4σ, where σ represents the residual component res m (t) standard deviation; the pulse amplitude detection threshold A in step S53 min The 85% to 95% quantiles of the envelope signal AM1(t) are taken; the influence interval length coefficient a in the step S54 is taken as 3 to 5; the filter length coefficient b in the step S55 is taken as an even number between 6 and 12.

Images