Brake disc dynamic balance detection system based on laser doppler vibration measurement

Abstract

The application relates to the technical field of rotating part dynamic balance detection, and discloses a brake disc dynamic balance detection system based on laser Doppler vibration measurement, which comprises the following steps: a system acquires complete running deformation vibration modes of a brake disc to be detected, and carries out orthogonal projection of the complete running deformation vibration modes to prestored theoretical rigid body modal vectors, so that pure rigid body vibration components are separated to carry out unbalance calculation; meanwhile, the system verifies the matching degree of a model relied on by the projection process online, the technical problem that structure modal vibration and unbalance vibration cannot be separated due to the same frequency is solved, and the dynamic balance of a flexible disc part can be accurately tested; and the online diagnosis mechanism can also identify the model mismatching problem caused by structure defects of a workpiece itself or errors of a clamping state synchronously in the testing process.

Description

TECHNICAL FIELD

[0001] The application relates to a brake disc dynamic balance detection system based on laser Doppler vibration measurement. BACKGROUND

[0002] Currently, dynamic balance testing of components is a basic quality control process. Mass imbalance on a rotating body will generate a synchronous vibration response with rotation speed. The size and angular position of the imbalance mass can be calculated reversely through measurement of the vibration amplitude and phase, and correction can be made accordingly. With the requirements of lightweight and high performance of the automobile industry, the structure of disc-shaped rotating parts such as brake discs tends to be thin-walled and complex, which makes it impossible to accurately describe the real physical behavior of the parts under dynamic working conditions by regarding the parts as quasi-rigid bodies. When such a brake disc with flexible characteristics rotates at high speed on a balancing machine, in addition to the centrifugal force excitation caused by mass imbalance, other factors such as aerodynamic load will also excite the structural bending or torsional modes of the part itself, generating a synchronous structural modal vibration with rotation speed. This vibration and the rigid body vibration caused by mass imbalance are superimposed on each other in the signal collected by the sensor, forming a composite vibration mode.

[0003] The conventional technical idea is usually to seek more advanced signal filtering algorithms. However, since the quality imbalance vibration and the structural modal vibration are signals of the same frequency on the rotating frequency, they are not noise and signals in the traditional sense, and therefore it is difficult to separate them based on the frequency domain characteristics. In addition to the above fundamental defects in the dynamic balance calculation model, the detection methods for other physical properties of the brake disc also have limitations in the measurement principle and control logic level, and cannot solve the key dynamic performance evaluation problem. For example, the Chinese utility model patent with the publication number CN210625579U discloses a kind of airplane brake dynamic disc polishing detection device, which detects the disc surface reflection area by means of photosensitive sensor array to evaluate whether the polishing amount reaches 30% to 50% required by manual. However, this scheme is essentially a static geometric quantity evaluation based on surface optical characteristics, which focuses on the completion of grinding operation, rather than the dynamic performance of the brake disc core. This method completely ignores the complex vibration form of the component under rotating working condition, and cannot effectively identify the quality imbalance and structural modal coupled vibration affecting the vehicle NVH (Noise, Vibration and Harshness) performance. The detection logic starting from the apparent characteristics does not touch the fundamental technical problem that the precision is limited due to the inseparability of the same frequency vibration source in the dynamic balance test of the flexible disc. This phenomenon causes a fundamental technical problem, that is, the classical dynamic balance calculation algorithm, whose physical model is based on the rigid body assumption, will include the structural modal vibration component in the response caused by the quality imbalance when receiving a composite signal, and calculate an imbalance vector deviating from the true mass distribution. The accuracy of the correction result based on this vector will be affected.

[0004] Specifically, the existing technology mainly has the following deficiencies: 1. Mismatch of calculation model. The rigid body theory premise relied on by the existing dynamic balance calculation method does not match the actual dynamic characteristics of modern flexible brake discs, causing applicability problems of the calculation model; 2. Difficulty in separating the same frequency vibration source. The rigid body vibration caused by mass imbalance and the structural modal vibration caused by other factors are signals of the same frequency at the measurement end, which are difficult to distinguish using conventional signal processing methods; 3. Accuracy of balance result is limited. The vibration signal input to the calculation model contains vibration components generated by non-mass imbalance factors, resulting in distortion of the calculated imbalance vector and limiting the precision of the balance correction work. Therefore, how to identify and separate the vibration component generated only by mass imbalance from a composite vibration signal containing multiple physical sources to obtain accurate dynamic balance test results has become a technical problem to be solved by the present invention. SUMMARY

[0005] The application provides a brake disc dynamic balance detection system based on laser Doppler vibration measurement, which mainly aims to solve the problem that in the prior art, the quality imbalance vibration and the structural modal vibration of the same frequency cannot be effectively distinguished, the physical model for dynamic balance calculation does not match the actual dynamic characteristics of the measured part, and the test result accuracy is affected.

[0006] To achieve the above-mentioned purpose, the application provides a brake disc dynamic balance detection system based on laser Doppler vibration measurement, which comprises:

[0007] a rotating drive module configured to clamp and drive the measured brake disc to rotate at a preset rotating speed;

[0008] a vibration signal acquisition module configured to: when the measured brake disc generates composite vibration containing rigid body vibration and structural modal vibration due to rotation, acquire vibration signals of multiple preset measurement points on the surface of the measured brake disc by using the laser Doppler vibration measurement method, and construct a measured running deformation mode vector according to the acquired vibration signals and a phase reference.

[0009] a controller internally storing a modal base library matched with the model of the measured brake disc, containing at least a theoretical rigid body modal vector, the controller being configured to: orthogonally project the measured running deformation mode vector onto the theoretical rigid body modal vector stored in the modal base library, so as to obtain a purified rigid body vibration vector; and calculate a mass imbalance vector of the measured brake disc only by taking the purified rigid body vibration vector as input; and calculate a residual vector between the measured running deformation mode vector and the modal base library, and generate and output a diagnostic alarm signal indicating model mismatch when the norm of the residual vector exceeds a preset threshold.

[0010] Preferably, the controller is configured to obtain the purified rigid body vibration vector by performing the following calculation formula: wherein, the purified rigid body vibration vector is, the measured running deformation mode vector is, the theoretical rigid body modal vector is, and the vector dot product operation is, the norm of the theoretical rigid body modal vector is.

[0011] Preferably, the specific operation of the vibration signal acquisition module configured to construct the measured running deformation mode vector comprises: acquiring vibration velocity time domain signals of the multiple preset measurement points respectively; the controller performs Fourier transform on each vibration velocity time domain signal according to the phase reference, so as to extract complex amplitudes at the rotating frequency synchronized with the preset rotating speed; and the controller combines the complex amplitudes of all measurement points at the rotating frequency according to a preset spatial order, so as to construct the measured running deformation mode vector.

[0012] Preferably, the modal basis library further comprises a plurality of theoretical flexible modal vectors corresponding to the structural characteristics of the brake disc to be measured; the theoretical rigid modal vectors and the theoretical flexible modal vectors are both obtained and stored in advance by performing finite element analysis on a digital model of the brake disc to be measured or by performing experimental modal testing on a physical brake disc.

[0013] Preferably, the diagnostic alarm signal indicates that the physical cause leading to the model mismatch is one of the following three: an error exists in the clamping state of the brake disc to be measured, the model of the brake disc to be measured does not match the modal basis library currently loaded by the controller, or the brake disc to be measured itself has a structural defect.

[0014] Preferably, the controller is configured to calculate the mass imbalance vector based on the influence coefficient method and based on the purified rigid vibration vector, the mass imbalance vector comprising information about the magnitude and angular position of the imbalance.

[0015] Preferably, the specific operation of the controller to calculate the residual vector comprises: projecting the measured operating deformation mode shape vector onto the theoretical rigid modal vectors and all the theoretical flexible modal vectors stored in the modal basis library respectively to obtain a set of modal participation coefficients; using the set of modal participation coefficients and all the theoretical modal vectors in the modal basis library to linearly superimpose and reconstruct a theoretical operating deformation mode shape vector; and calculating the vector difference between the measured operating deformation mode shape vector and the theoretical operating deformation mode shape vector to obtain the residual vector.

[0016] Preferably, the controller is further configured to: receive an input brake disc model information before the dynamic balancing detection starts; and select and load the modal basis library that is unique to the model of the brake disc to be measured from the stored modal basis libraries of a plurality of brake discs of different models according to the model information.

[0017] Preferably, the controller is further configured to: when the diagnostic alarm signal is not generated, output the final mass imbalance vector; and when the diagnostic alarm signal is generated, prohibit the output of the mass imbalance vector and output the diagnostic alarm signal to an external system.

[0018] Preferably, the diagnostic alarm signal output by the controller further comprises the norm value of the residual vector and the value of the preset threshold, which are used to quantitatively evaluate the degree of model mismatch.

[0019] Compared with the prior art, the present application has the following beneficial effects:

[0020] 1. A new information processing method for dynamic balance testing is established by obtaining the complete running deformation mode of the brake disc to be tested online and mathematically projecting it to the pre-stored rigid modal vector corresponding to the physical properties of the measured part. This method makes the calculation basis of the unbalance directly locked in the component in the vibration information that is unique to the rigid motion form. This process separates the calculation of the balance value from the deformation vibration component that is not directly related to the mass distribution caused by the structural flexibility or aerodynamic load of the brake disc, and further makes the calculation distortion problem caused by model mismatch in the traditional method no longer an interference factor that needs to be pre-processed in the logical framework of the method.

[0021] 2. While using the projection method to separate the rigid vibration component, the pre-stored complete modal base library including rigid and flexible is also used to reconstruct the measured original vibration deformation mode, and the residual error between the measured mode and the reconstructed mode is calculated. The introduction of this residual value establishes a real-time model matching degree self-checking mechanism for the current single measurement for the system, so that the output of each unbalance is accompanied by a simultaneous confirmation of the validity of its calculation premise. This operation mode that integrates purification processing and premise checking into one measurement process ensures the credibility of the test results when facing the inevitable clamping variation or product individual tolerance in the production line. Through the calculation and monitoring of the residual error, the dynamic balance testing process, which is a single target measurement process, is expanded to a working process with online monitoring capability of the structure state of the measured part. When the residual error exceeds the preset threshold, the system indicates not only the deviation of the balance testing condition, but also the abnormal vibration component that cannot be explained by the normal physical model appears in the measured vibration. This abnormality directly points to the possible crack damage or material unevenness structural defects of the brake disc itself, so that the vibration measurement data originally only serving the quality balance degree judgment is upgraded to the function of preliminary screening of structural integrity. BRIEF DESCRIPTION OF DRAWINGS

[0022] Fig. 1 The flow chart of the unbalance calculation and model diagnosis dual-path processing of the present application;

[0023] Fig. 2 The residual vector norm change curve diagram in the model mismatch diagnosis of the present application;

[0024] Fig. 3 The hardware composition and core signal interaction block diagram of the system of the present application. DETAILED DESCRIPTION

[0025] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments; based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0026] The disclosed brake disc dynamic balance detection system based on laser Doppler vibration measurement comprises a rotating drive module, a vibration signal acquisition module and a controller. When performing dynamic balance detection, the rotating drive module is used to clamp the brake disc to be measured and provide a stable and accurate rotating speed for the brake disc to be measured, the vibration signal acquisition module is used to non-contactly acquire complete vibration shape information of the brake disc in a rotating state, and the vibration data is transmitted to the controller. The controller analyzes and separates the composite signal containing multiple vibration sources according to the data processing method in the controller, finally outputs a physical quantity related only to mass imbalance, and simultaneously checks the matching degree of the model relied on by the analysis process. In the quality detection station of the automobile brake disc production line, due to the thin-walled structure of the brake disc, when rotating at high speed, not only the centrifugal force caused by the mass imbalance of the brake disc will cause the rigid body vibration of the disc body, but also the disc body will be excited to vibrate in the structural bending or torsional mode due to the factors such as main shaft runout or pneumatic load. The structural modal vibration and the rigid body vibration caused by the imbalance are the same frequency signals in the rotating frequency, and the conventional filtering method is difficult to separate the two, which leads to the error of the structural modal vibration to the imbalance response in the subsequent balance calculation algorithm, thereby causing the deviation of the calculation result. In order to solve this problem, the controller of the present application is configured to execute a data processing procedure based on modal projection, and the procedure processes the acquired vibration signal as a spatial mode corresponding to the physical shape of the brake disc.

[0027] The specific implementation mode of the data processing procedure is as follows. Before detection, the controller loads a modal base library matched with the brake disc to be measured according to the external input brake disc model information from the internal memory. The modal base library is a digital model of the brake disc of the type, which is pre-established through finite element analysis or experimental modal test, and the data structure of the modal base library is a matrix, which at least contains a theoretical rigid body modal vector , which describes the theoretical vibration shape of the brake disc when the brake disc is purely rigidly tilted due to mass imbalance, in addition, the modal base library can also contain one or more theoretical flexible modal vectors , to describe the main structure of the brake disc vibration mode; then, the rotating drive module drives the brake disc to be tested to a preset rotating speed, for example, 900 RPM; after the rotating speed is stable, the vibration signal acquisition module starts, and the scanning laser Doppler vibration meter in the module scans the surface of the brake disc at M preset measuring points and collects the vibration speed signal of each measuring point in the time domain. At the same time, a phase reference sensor synchronized with the main shaft of the rotating drive module provides a real-time rotating angle reference signal; the controller receives the M vibration speed time domain signals and the phase reference signal, performs Fourier transform on the time domain signal of each measuring point, and extracts the complex amplitude at the rotating frequency synchronized with the preset rotating speed. The complex amplitude contains the vibration amplitude and phase of the point; finally, the controller combines the complex amplitudes of all M measuring points at the rotating frequency according to the preset spatial order to construct an M-dimensional complex vector, which is the measured running deformation mode vector . It is necessary to point out that before the controller constructs the measured running deformation mode vector according to the collected M vibration speed time domain signals, a signal quality pre-checking procedure is also performed. The procedure first calculates the signal-to-noise ratio of each time domain signal at the rotating frequency, that is, the ratio of the narrow-band energy centered on the rotating frequency to the average noise energy in the adjacent frequency band, and confirms that the signal-to-noise ratio is not less than a preset signal-to-noise ratio threshold, for example, 20 dB. At the same time, the procedure also performs full-spectrum analysis on the signal to extract the amplitude at the rotating frequency and the amplitude at twice the frequency of the rotating frequency , and confirms that the amplitude of the rotating frequency is at least one order of magnitude higher than the amplitude of the rotating frequency and all other harmonic or noise peaks. Only when all M signals pass the above signal-to-noise ratio and main frequency purity check, the controller calls these signals for subsequent mode vector construction and modal projection calculation. If any signal fails the check, the system will output a prompt indicating that there is excessive synchronous interference or signal acquisition link abnormality at the current rotating speed, and pause the subsequent calculation.

[0028] In addition, in order to ensure the long-term stability and data effectiveness of the entire measurement chain, the system is configured to automatically perform a measurement system state self-calibration procedure once before starting each batch of production tasks or after accumulating running for more than a preset time, for example, 4 hours. In the procedure, the rotating drive module does not install any brake disc to be tested, and the vibration signal acquisition module performs a vibration measurement on the surface of the clamped tooling in a stationary state. The controller collects the background vibration signal at this moment and calculates the vibration speed root mean square value in the entire effective frequency band ; then, the controller compares the real-time measured a value of the device background noise baseline representing the electronic noise inherent to the measuring chain and the environmental background vibration, calibrated and stored when the system is first installed and commissioned comparing the value of the measured vibration level with the value of the device background noise baseline if the value of the measured vibration level does not exceed 1.2 times the value of the device background noise baseline, it is determined that the current state of the measuring system is normal and the subsequent dynamic balancing detection is permitted to start, otherwise, if the value of the measured vibration level exceeds the limit value the system prohibits the execution of the dynamic balancing detection task and outputs to an external system a maintenance instruction indicating that the measuring device needs to be checked or the environmental vibration is excessive; immediately after, the controller performs the core separation operation, i.e. orthogonally projects the measured operating deformation mode vector onto the theoretical rigid body modal vectors stored in the modal base library previously loaded, thereby obtaining a purified rigid body vibration vector ; the projection operation is completed by performing the following calculation: wherein is the purified rigid body vibration vector, is the measured operating deformation mode vector, is the theoretical rigid body modal vector, and · is the vector dot product operation, is the norm of the theoretical rigid body modal vector; the projection operation separates from the measured operating deformation mode vector the component that is in spatial form consistent with the pure rigid body unbalance vibration form, all the structural flexible deformation vibration components that are in spatial form orthogonal to the rigid body modal are filtered out; the controller calls the dynamic balancing solution algorithm, such as the influence coefficient method, taking as input only the purified rigid body vibration vector and calculates the mass unbalance vector of the brake disc to be tested, which contains the size of the correction mass and its angular position information.

[0029] It should be noted that the effectiveness of the above processing flow is based on the premise that the pre-stored theoretical modal base library can accurately describe the actual physical characteristics of the brake disc to be tested; in the production process, the manufacturing tolerance of the brake disc individual, the variation of the clamping state, or the structural defects such as cracks in the disc body, may cause a mismatch between the theoretical model and the actual physical behavior; to cope with this situation, the controller is also configured to perform an online verification mechanism of the model matching degree while completing the projection separation; the specific operation of the mechanism is that the controller calculates the residual vector between the measured operating deformation mode vector and the modal base library; in a specific embodiment, the calculation includes projecting the measured operating deformation mode vector , respectively, to the theoretical rigid modal vector and all the theoretical flexible modal vectors stored in the modal base library, a set of modal participation coefficients is obtained, and then the set of coefficients and all the theoretical modal vectors in the modal base library are used to linearly superimpose and reconstruct a theoretical running deformation mode vector, and the vector difference between the measured running deformation mode vector and the theoretical running deformation mode vector is calculated to obtain a residual vector; then, the controller calculates the norm of the residual vector and compares it with a preset threshold; if the norm does not exceed the threshold, it indicates that the current model matching degree is within the allowable range, and the current detection is valid, and the controller outputs the finally calculated mass imbalance vector; otherwise, if the norm exceeds the threshold, the controller determines that model mismatch occurs, and at this time, the possible inaccurate mass imbalance vector is not output, and a diagnostic alarm signal indicating model mismatch is generated, which is used to indicate process abnormalities such as clamping error, inconsistent workpiece model or workpiece structural defects.

[0030] Embodiment 1: In a brake disc automatic production line for high-performance electric vehicles, the dynamic balance detection system of the application is deployed in the quality control link; a large-diameter thin-walled brake disc produced by the production line has a large structural flexibility, and the balance result repeatability is poor in traditional dynamic balance testing. Some of the balanced workpieces still have shaking problems in the subsequent vehicle noise, vibration and sound roughness, i.e. NVH testing, resulting in production bottlenecks; when one of the brake discs to be tested is sent into the detection system of the application by the automatic tooling and is clamped, the system controller loads the corresponding modal base library according to the received workpiece model, and the rotating drive module drives the brake disc to a preset speed; after the speed stabilizes, the system constructs a measured running deformation mode vector containing rigid vibration and structural modal vibration through the vibration signal acquisition module ; the controller performs two core data processing operations in parallel, one of which is to orthogonally project to the theoretical rigid modal vector , to obtain a purified rigid vibration vector , and based on this, an initial mass imbalance vector is calculated; the other is to calculate the residual vector between and the entire modal base library, and to calculate its norm; in this specific measurement, the residual vector norm value calculated by the controller exceeds the preset threshold, and the system determines that the theoretical model on which the current measurement is based and the actual vibration behavior of the workpiece are mismatched, therefore, the system executes the preset response procedure, i.e. prohibits output of the initial mass imbalance vector, and sends a diagnostic alarm signal indicating model mismatch to the central control system of the production line.

[0031] After receiving the alarm signal, the production line maintenance personnel inspected the automatic clamping mechanism at the workstation and found that a locating pin was worn, causing a slight tilt in the clamping posture of the brake disc under test. This tilt altered the dynamic boundary conditions during disc rotation, causing its actual vibration mode to deviate from the theoretical model. This deviation was captured by the norm change of the residual vector. After replacing the locating pin and re-clamping and testing the brake disc, the system executed the complete measurement and data processing flow again. This time, the calculated residual vector norm was within the preset threshold, and the system determined that the model match was valid. Therefore, the system used the purified rigid body vibration vector obtained from the second measurement... The calculated mass imbalance vector is output as the final valid result. After the brake disc is deweighted and corrected based on this result, it successfully passes the subsequent NVH test. This process shows that combining the functions of mass imbalance calculation and online model matching verification makes dynamic balancing detection no longer an isolated numerical measurement process, but a process diagnostic link that can simultaneously monitor the validity of the measurement preconditions. The system provides confirmation of the physical meaning of each imbalance correction value it outputs by self-verifying each measurement behavior.

[0032] Example 2: To quantitatively verify the effectiveness of the present invention's technical solution in identifying the mass imbalance vector under the presence of co-frequency structural modal vibration interference, this example constructs a comparative verification experiment. The experiment is conducted on a standard dynamic balancing test platform, which consists of a servo motor drive system with a rotational speed control accuracy of ±1 RPM, a set of highly repeatable clamping fixtures, and a vibration signal acquisition system composed of a scanning laser Doppler vibrometer and a synchronous acquisition controller. The vibration velocity measurement resolution of the vibrometer is 0.1 µm / s. The test object is a known thin-walled brake disc that easily excites a first-order umbrella-shaped structural mode at a specific rotational speed. Before the experiment, the brake disc is precisely dynamically balanced to ensure that its initial imbalance is less than 1 g·cm. Then, at a designated position on the disc (radius 120 mm, angle...),... A 5.0g test weight is installed on the brake disc to apply a mass imbalance vector to be measured. To simulate structural modal vibration interference, a miniature piezoelectric ceramic sheet is attached to the non-measurement area of ​​the brake disc and driven by a function generator that is phase-locked with the signal of the spindle rotary encoder. This causes the sheet to generate a continuous excitation force at a frequency (1x) that is exactly the same as the spindle rotation frequency. The amplitude and phase of this excitation force are adjusted to be sufficient to excite a first-order umbrella-shaped structural modal vibration, and the amplitude of this vibration at the measurement point is on the same order of magnitude as the unbalanced vibration amplitude generated by the aforementioned 5.0g test weight.

[0033] Two treatment groups are set in the test, namely a control group and a test group; the control group adopts a conventional dynamic balance test method, which only collects the vibration signal at the bearing seat supported by the tool shaft, and performs Fourier transform on the signal to extract the complex amplitude at the rotation frequency for imbalance calculation; the test group adopts the complete technical solution of the application, that is, the vibration signals of multiple measuring points on the disc surface are collected to construct the measured running deformation mode vector , and subsequent modal orthogonal projection separation and model matching degree online verification are performed; the brake disc in the above state with test weight and under structural excitation is tested at the same speed of 900 RPM, and the mass imbalance vector results calculated by the two methods are recorded in Table 1; referring to Table 1, the table presents the measurement results of the known imbalance amount by the two methods.

[0034] Table 1: Comparison table of measurement results of known imbalance amount by different test methods.

[0035]

[0036] The data in Table 1 shows that in the presence of the same frequency structural modal vibration interference, the measurement results of the control group have large deviations in the amplitude and phase of the imbalance amount, and the amplitude measurement error reaches 58.3%, which shows that the conventional single-point measurement method cannot distinguish the two sources of the same frequency vibration, and incorrectly counts the structural modal vibration into the imbalance response; in contrast, the measurement results of the test group are highly consistent with the known true imbalance amount, and the amplitude measurement error is only 2.8%, which benefits from the modal orthogonal projection processing method used, which separates the structural modal vibration component that does not conform to the rigid body motion mode in space by analyzing the spatial form characteristics of the vibration, thereby obtaining the vibration information caused only by the mass imbalance; the test result confirms that the technical solution of the application can identify and separate the vibration component caused only by the mass imbalance in a composite vibration signal containing multiple physical sources, thereby obtaining accurate dynamic balance test results; in order to further verify the necessity and superiority of the modal orthogonal projection processing method used in the application compared with other conventional signal processing methods in the art which are also based on multi-point measurement but have different technical paths, a following comparative example is added.

[0037] Comparative Example 1: To verify the superiority of the technical solution of the present application compared with a conventional method which is technically more complex, the comparative example constructs an experimental environment which is completely identical with Example 2 in terms of hardware configuration, test object, known unbalance amount application mode and structural modal excitation mode. The only difference between the comparative example and the test group of Example 2 is that the data processing algorithm executed by the controller is different: the comparative example uses a multi-point vibration signal space vector average method to replace the modal orthogonal projection separation step in the technical solution of the present application; the specific processing process of the multi-point vibration signal space vector average method is as follows: first, the vibration signals of M preset measurement points on the brake disc surface are collected, and the Fourier transform is performed on each signal to extract the complex amplitude at the rotation frequency; then, the method performs vector summation on the complex amplitudes obtained by all M measurement points in the complex plane, and calculates the arithmetic mean value to obtain a single space average vibration vector; finally, the method takes the space average vibration vector as input, calls the influence coefficient method which is completely identical with Example 2, to calculate the mass unbalance vector. Under the same 900 RPM speed, the same brake disc with a 5.0 g test weight and subjected to the same frequency structural excitation is tested, and the mass unbalance vector calculated by the method is recorded in Table 2.

[0038] Table 2: Comparison table of measurement results of known unbalance amount by the comparative example and Example 2 of the present application.

[0039]

[0040] The test results of Comparative Example 1 show that under the condition of the presence of the same frequency structural modal vibration interference, the measurement results obtained by the multi-point vibration signal space vector average method have deviations in amplitude and phase, and the amplitude measurement error reaches 34.2%. The results confirm that although the conventional method also uses the vibration information of multiple points on the disc surface, the space vector average processing method cannot effectively distinguish the rigid body vibration with different spatial forms from the structural modal vibration in terms of physical principle, and the calculated space average vibration vector is still a composite quantity seriously contaminated by the structural modal vibration, so it cannot obtain accurate dynamic balancing test results. This result indirectly proves that the technical path of the present application based on the physical model (theoretical rigid body modal vector) for orthogonal projection to separate the vibration components has technical advantages and necessity for solving the same frequency vibration interference problem of flexible disc parts.

[0041] Example 3: This example combines Figs. 1 to 3 the description of the brake disc dynamic balancing detection system based on laser Doppler vibration measurement, as follows: Fig. 1As shown, it shows the complete information path from obtaining the original vibration to outputting the final result, a brake disc to be tested produces physical vibration when rotating, the vibration velocity time domain signal is obtained through the vibration signal acquisition link, and the mode shape vector is constructed accordingly, the mode shape vector is sent to two parallel processing paths, one path retrieves the theoretical modal vector from A1: modal base library, performs rigid body vibration separation operation on the input mode shape vector to obtain the purified rigid body vibration, and calculates the unbalance based on the purified rigid body vibration, and finally outputs the mass unbalance for the line control system / user to use, the other parallel path also retrieves the theoretical modal vector from A1: modal base library, performs model matching degree operation on the input mode shape vector to obtain the residual vector norm, and generates a diagnosis result based on the norm, and outputs a diagnosis alarm signal to the line control system / user when the model is mismatched.

[0042] As shown in Fig. 2 The horizontal coordinate of the graph is the test sequence number, and the vertical coordinate is the residual vector norm. A dashed line in the graph represents a preset threshold value as a judgment basis, which is constant at 0.20. Another point-splined line with data points represents the actual value of the residual vector norm calculated in the previous tests. The value is lower than the preset threshold value in most test sequence numbers, but a peak value with a maximum value exceeding 0.30 appears near the test sequence number 24. It clearly shows that when the system detects abnormalities such as workpiece clamping errors, the residual vector norm will suddenly change and exceed the preset threshold value, thereby triggering the working state of the diagnosis alarm signal.

[0043] As shown in Fig. 3 The core is a rotating drive module that drives the brake disc to be tested. The module includes a servo motor and a precision spindle, and fixes the brake disc through a clamping tool. A laser Doppler vibrometer projects a laser vibration beam to a preset measurement point on the surface of the brake disc and performs scanning measurement. The collected vibration signal is sent to the controller through data transmission. A phase reference sensor provides a rotation reference for the system through phase detection, and sends this information to the controller through a phase synchronization link. The controller integrates modal base library storage, data processing unit and unbalance calculation function, and sends control signals to the rotating drive module according to the processing result.

[0044] Example 4: Before the dynamic balance detection system of the application is put into a batch detection task of a new type brake disc, a standardized offline calibration procedure needs to be performed to establish a modal base library matched with the workpiece of this type and set the residual norm threshold value for model mismatch diagnosis. This procedure aims to ensure that the accuracy of subsequent online detection and the reliability of diagnosis function are based on an experimentally verified and reproducible engineering reference. The initial step of the procedure is to construct a preliminary modal base library. For the new type brake disc, first, the theoretical rigid body modal vector is calculated through finite element analysis method and a plurality of low-order theoretical flexible modal vectors ; subsequently, to verify and correct the performance of the theoretical model under the actual clamping condition, a batch (for example, 20 pieces) of the brake disc of the model confirmed to have no structural defects and the size within the tolerance center range is selected as the reference sample, each reference sample is installed one by one on the actual tooling of the dynamic balance detection system, and the experimental modal test is carried out through a small force value hammering or acoustic excitation and the like in combination with the vibration signal acquisition module to obtain the actual modal parameters under the real constraint boundary condition; the controller performs correlation analysis on the measured modal shape and the theoretical modal shape obtained by the finite element analysis, if the correlation degree is higher than the preset standard, the effectiveness of the theoretical model is confirmed, if there is a slight deviation caused by clamping stress and the like, the theoretical modal vector is appropriately corrected according to the measured data, so as to form a final modal base library verified and matched with the actual working condition and stored in the controller.

[0045] After the modal base library is established, the next step of the procedure is to set a quantitative criterion for model mismatch diagnosis, that is, the norm threshold of the residual vector; the operator repeatedly carries out normal clamping and unloading on the test tooling for the aforementioned 20 reference samples, and performs a standard dynamic balance test process for each normal clamping state, for example, repeatedly tests each sample 5 times, and a total of 100 groups of measured running deformation modal vectors under normal working conditions are obtained ; for each of the 100 groups of data, the controller calculates the norm of the residual vector according to the final modal base library stored; after all the tests are completed, the controller statistically analyzes the 100 residual norm values, and calculates the mean value and the standard deviation ; in order to ensure the detection sensitivity of the abnormal state while avoiding false alarms for normal production fluctuations, the preset threshold of the residual norm is set according to the statistical process control theory, which is the sum of the mean value and several times the standard deviation, for example, the threshold is determined as ; at this time, the modal base library establishment and the diagnostic threshold setting for the new model brake disc are completed, and the system can switch to the online automatic detection mode; this calibration procedure changes the determination of the diagnostic threshold from a fuzzy setting depending on experience to a quantitative parameter with clear physical meaning and statistical basis obtained after statistical learning of the batch of workpieces and process state.

[0046] Embodiment 5: The influence coefficient used for balance solving in the dynamic balance detection system controller of the application is determined by the following calibration procedure: a reference brake disc that has been precisely balanced is selected, installed on the test station and rotated at a preset speed, the system collects its vibration signals and calculates to obtain the initial purified rigid body vibration vector ; subsequently, the influence coefficient of the reference brake disc In the angular position, the test mass of known mass is added, and the system is rotated again at the same speed. The new purified rigid body vibration vector is calculated The controller calculates the vector difference between the two measurements of the purified rigid body vibration vector The vibration response change caused by the test mass is obtained, and the ratio of the vibration response change to the applied test mass vector is calculated as the influence coefficient of the test system at the speed, which is stored in the controller as the basis for subsequent imbalance calculation.

[0047] To cope with the modal drift of the brake disc structure caused by environmental temperature changes, the dynamic balance detection system of the application also configures a modal base library adaptive updating procedure; the system workstation integrates a temperature sensor to monitor the environmental temperature. When the monitored temperature deviates from the reference temperature at the initial calibration by more than a preset range, such as ±10 When the monitored temperature deviates from the reference temperature at the initial calibration by more than a preset range, such as ±10

[0048] Before applying the dynamic balance detection system of the application to batch detection of a new type of brake disc, a set of pre-test parameter optimization procedures need to be performed to determine the specific configuration of the preset speed and multiple preset measurement points; the purpose of this procedure is to determine a set of engineering parameters that can obtain effective vibration pattern information with the highest signal-to-noise ratio in the shortest test time for subsequent online detection; the procedure first determines the test speed, the steps are as follows: select a reference sample, after clamping on the test station, the rotating drive module drives it to slowly increase from a lower speed (such as 100 RPM) to a higher speed (such as 2000 RPM), during this speed-up process, the vibration signal acquisition module continuously monitors the vibration response of specific points on the brake disc surface, and the controller synchronously draws the vibration frequency vs. speed relationship map; by analyzing the map, the structural modal resonance regions of the brake disc can be identified, i.e. the speed range where the vibration amplitude peaks with speed; to avoid exciting excessive structural modal vibrations during conventional testing and thereby interfering with the measurement of unbalanced vibrations, the preset speed is selected at a speed between two adjacent main resonance regions, which can produce sufficient amplitude of unbalanced response signal and is far away from each order resonance peak.

[0049] After the test speed is determined, the procedure continues to optimize the number and spatial distribution of the measurement points for the laser Doppler vibrometry: still using the aforementioned reference specimen, at the determined test speed, it is driven in rotation and an artificial excitation of a known modal is applied, so that it simultaneously generates a rigid body modal vibration and a low order flexible modal vibration; the system first performs a scan with an ultra-high density (e.g. 500) of measurement points, far exceeding the conventional number, to construct a high-fidelity reference operating deformation mode shape vector; subsequently, the controller performs an iterative down-sampling analysis, i.e. from the high-density data, data subsets are extracted with different numbers (e.g. from 100 to 20) and different spatial distribution patterns (e.g. along 2, 4 or 8 diameters), and a low-density measured operating deformation mode shape vector is constructed for each data subset; by calculating the modal confidence criterion correlation between each low-density mode shape vector and the aforementioned high-fidelity reference mode shape vector, the ability of the measurement point configuration to reproduce the key modes (rigid body and low order flexible) is evaluated; finally, the number and spatial distribution of measurement points that are able to reproduce all key modes with a correlation higher than a pre-set threshold (e.g. 0.98) are determined, i.e. the multiple pre-set measurement points for the batch detection of the model brake disc are determined.

[0050] It is apparent for a person skilled in the art that the present application is not limited to the details of the above exemplary embodiments, but can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not to limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A brake disc dynamic balance detection system based on laser Doppler vibrometry, characterized in that, The system comprises: a rotating drive module configured to clamp and drive a brake disc to be tested to rotate at a preset rotating speed; a vibration signal acquisition module configured to, when the brake disc to be tested generates a composite vibration containing rigid body vibration and structural modal vibration due to rotation, acquire vibration signals of a plurality of preset measurement points on the surface of the brake disc to be tested by using a laser Doppler vibration measurement method, and construct an actually measured operating deformation mode shape vector based on the acquired vibration signals and a phase reference; a controller, which internally stores a modal basis library containing at least one theoretical rigid body modal vector matched with the model of the brake disc to be tested, and is configured to: orthogonally project the actually measured operating deformation mode shape vector onto the theoretical rigid body modal vector stored in the modal basis library, thereby obtaining a purified rigid body vibration vector; calculate a mass imbalance vector of the brake disc to be tested only by taking the purified rigid body vibration vector as input; and calculate a residual vector between the actually measured operating deformation mode shape vector and the modal basis library, and generate and output a diagnostic alarm signal indicating model mismatch when the norm of the residual vector exceeds a preset threshold; and the controller is configured to obtain the purified rigid body vibration vector by performing the following calculation formula: where, is the purified rigid body vibration vector, is the measured operating deformation mode vector, is the theoretical rigid body modal vector, is the vector dot product operation, is the norm of the theoretical rigid body modal vector; The modal basis library further contains a plurality of theoretical flexible modal vectors corresponding to the structural characteristics of the brake disc to be tested; the theoretical rigid body modal vector and the theoretical flexible modal vector are both obtained and stored in advance by performing finite element analysis on a digital model of the brake disc to be tested or experimental modal testing on a physical brake disc; The specific operation of the controller to calculate the residual vector includes: projecting the actually measured operating deformation mode shape vector onto the theoretical rigid body modal vector and all theoretical flexible modal vectors stored in the modal basis library respectively to obtain a set of modal participation coefficients; reconstructing a theoretical operating deformation mode shape vector by linear superposition using the set of modal participation coefficients and all theoretical modal vectors in the modal basis library; and calculating the vector difference between the actually measured operating deformation mode shape vector and the theoretical operating deformation mode shape vector to obtain the residual vector.

2. The brake disc dynamic balance detection system based on laser Doppler vibration measurement according to claim 1, characterized in that, The specific operation of the vibration signal acquisition module to construct the actually measured operating deformation mode shape vector includes: acquiring vibration velocity time domain signals of each of the plurality of preset measurement points; the controller performs Fourier transform on each vibration velocity time domain signal according to the phase reference; and the controller combines the complex amplitudes of all measurement points at the rotating frequency in accordance with a preset spatial order.

3. The brake disc dynamic balance detection system based on laser Doppler vibration measurement according to claim 1, characterized in that, The diagnostic alarm signal indicates that the physical cause leading to model mismatch is one of the following three: the clamping state of the brake disc to be tested is incorrect, the model of the brake disc to be tested does not match the modal basis library currently loaded by the controller, or the brake disc to be tested itself has structural defects.

4. The brake disc dynamic balance detection system based on laser Doppler vibration measurement according to claim 1, characterized in that, The controller is configured to calculate the mass imbalance vector based on the influence coefficient method and based on the purified rigid body vibration vector, and the mass imbalance vector contains the size and angular position information of the unbalance amount.

5. The brake disc dynamic balance detection system based on laser Doppler vibration measurement according to claim 1, characterized in that, The controller is further configured to: receive an inputted brake disc model information before the dynamic balance detection starts; and select and load a modal base library that is uniquely matched with the current brake disc model from the stored modal base libraries of different brake disc models according to the model information.

6. The brake disc dynamic balance detection system based on laser Doppler vibration measurement according to claim 1, characterized in that, The controller is further configured to: output a final mass imbalance vector when no diagnostic alarm signal is generated; and output no mass imbalance vector and output the diagnostic alarm signal to an external system when the diagnostic alarm signal is generated.

7. The brake disc dynamic balance detection system based on laser Doppler vibration measurement according to claim 1, characterized in that, The diagnostic alarm signal outputted by the controller further comprises a norm value of the residual vector and a preset threshold value.

Images