A laser detection system and method for expansion flange

Abstract

The present invention discloses an expansion flange laser detection system and a method thereof. An expansion flange laser detection system comprises: a power supply module; a microcontroller module, a laser measurement module, and a servo drive module, which drives multiple motors to accurately adjust the spatial position and incident angle of the laser. A data processing method for an expansion flange laser detection system comprises the following steps: (1) surface modeling and deformation analysis; (2) laser posture control; (3) scanning detection; (4) signal processing and feature extraction; (5) real-time control. The present invention has the following beneficial effects: the expansion flange laser detection system can automatically adjust the laser incident angle, and realize high-precision detection of inner hole deformation through mathematical models and control algorithms. The system establishes an inner hole surface model in a three-dimensional cylindrical coordinate system, and adopts a real-time adaptive control strategy to ensure that the laser beam always maintains an optimal incident angle with the detection surface.

Description

Technical Field

[0001] The invention relates to an expansion flange detection system, and in particular to an expansion flange laser detection system and a method thereof. Background Art

[0002] The expansion flange is an important mechanical connection part, and its quality directly affects the sealing and safety of the pipeline system. With the continuous improvement of industrial production requirements, the detection of the internal quality of the expansion flange is becoming more and more important. At present, the expansion flange detection on the market mainly adopts laser scanning technology to detect the surface and internal defects of the flange through optical principles. This detection method has been applied to a certain extent in industrial production.

[0003] The existing expansion flange laser detection process usually includes the following steps: first, fix the flange on the detection platform; then use the laser emitting device to emit a laser beam to scan the flange surface; finally, use the photoelectric receiving device to capture the reflected light signal, and use the computer system to perform data analysis and defect determination. This detection method can achieve rapid detection of flange surface defects.

[0004] Existing expansion flange laser detection systems have significant technical limitations. Due to the special structure of the expansion flange, its inner hole will expand or shrink during the processing process. This geometric feature poses a huge challenge to the detection of internal defects. Traditional laser detection methods are often difficult to accurately locate and identify internal defects when faced with such a deformed structure. In the prior art, laser detection equipment usually uses a fixed angle to emit lasers, resulting in the laser incident direction not always being perpendicular to the inner hole surface, causing problems such as abnormal signal reflection and detection blind spots, which seriously affects the detection accuracy and reliability.

[0005] Therefore, it is of great practical significance to develop a laser detection system that can adapt to the special structural characteristics of expansion flanges and realize high-precision internal defect detection. This system should not only solve the detection problem caused by the deformed inner hole, but also have intelligent signal processing capabilities to meet the strict requirements of modern industrial production for quality inspection. Summary of the invention

[0006] The purpose of the present invention is to provide an expansion flange laser detection system and method thereof. The expansion flange laser detection system and method thereof can automatically adjust the laser incident angle, and realize high-precision detection of inner hole deformation through mathematical models and control algorithms. The system establishes an inner hole surface model in a three-dimensional cylindrical coordinate system, and adopts a real-time adaptive control strategy to ensure that the laser beam always maintains the best incident angle with the detection surface.

[0007] The above technical objectives of the present invention are achieved through the following technical solutions:

[0008] A laser detection system for expansion flanges comprises: a power module, which is a 3.3V power module of a linear voltage regulator and is used to provide a stable working power supply for the system; a microcontroller module, which is electrically connected to the power module and is used to control the operation of the entire system and process detection data; a laser measurement module, which is electrically connected to the microcontroller module and is used to emit a laser beam and receive a reflection signal, so as to detect and measure the internal hole of the expansion flange; and a servo drive module, which is electrically connected to the microcontroller module and is a servo motor drive module based on a DRV8813PWPR chip and is used to accurately adjust the spatial position and incident angle of a laser by driving multiple motors.

[0009] The present invention is further configured as follows: the power module includes: two filter capacitors of 10 μF and 1 μF at the input end, used to filter out the ripple of the input power supply; a 1 μF voltage stabilizing capacitor at the output end, used to stabilize the output voltage; the chip enable terminal CE is connected to a high level to start the chip, VSS is grounded to form a complete loop, and the NC pin is left floating to prevent interference.

[0010] The present invention is further configured as follows: the microcontroller module includes: a U1 chip, which is used to process data information and control algorithms; multiple VDD pins and an analog power supply pin VDDA, which provide a stable working power supply for the chip through a decoupling capacitor network; the corresponding VSS pin and the analog ground VSSA are connected to the system ground to form a complete power supply loop; the OSC_IN and OSC_OUT pins are connected to an external crystal oscillator to generate a system clock signal; the NRST pin realizes the system power-on reset function through an RC reset circuit; the BOOT0 pin configures the chip startup mode through a pull-up or pull-down resistor.

[0011] The present invention is further configured as follows: the servo drive module includes: a first motor interface, driving a stepper motor to achieve precise adjustment of the laser pitch angle; a second motor interface, driving a DC motor to control the position movement of the laser on the longitudinal axis; and a third motor interface, driving the stepper motor to achieve rotational scanning movement of the laser around the axis.

[0012] The present invention is further configured as follows: the servo drive module also includes: an LC filter circuit composed of a 10μH inductor and a 1N4007 diode, which is used to suppress electromagnetic interference generated during the motor drive process; two 100nF filter capacitors, which are used to filter out high-frequency noise in the drive signal; a 74HC14D Schmitt trigger and a 10kΩ pull-up resistor, which are used to shape the drive signal and improve the anti-interference ability.

[0013] The present invention is further configured as follows: the laser measurement module includes: a 100nF decoupling capacitor connected to the power supply pin to filter out power supply noise during operation; I2C communication with the main controller is realized through two signal lines SCL and SDA to transmit measurement data; the GPIO1 pin outputs an interrupt signal or a status indication for synchronous control; the XSHUT pin controls the enable state of the module to realize switch control of the laser.

[0014] A data processing method for an expansion flange laser detection system comprises the following steps:

[0015] (1) Surface modeling and deformation analysis: The laser performs a rapid pre-scan under the control of the detection system to obtain the actual contour information of the inner hole of the expansion flange. Based on the obtained contour data, the incident angle required for each detection position is calculated;

[0016] (2) Laser attitude control: adjusting the angle state of the laser according to the calculated incident angle of each detection point;

[0017] (3) Scanning detection: After the angle adjustment is completed, the motor drives the laser to rotate 360 ​​degrees, and its rotation speed is adaptively adjusted according to the actual defect situation;

[0018] (4) Signal processing and feature extraction: Filter the laser receiving signal to remove high-frequency noise and interference during the measurement process, analyze the signal through wavelet transform, obtain the time domain and frequency domain information of the signal, and extract the details of the defect characteristics;

[0019] (5) Real-time control: Through the improved PID algorithm, the precise positioning and attitude adjustment of the laser are controlled, and the rotation speed, tilt angle and axial position of the laser are controlled to adapt to the deformation characteristics of the inner hole surface of the expansion flange.

[0020] The present invention is further configured as follows: surface modeling and deformation analysis include: establishing a mathematical model S(r,θ,z) of the inner hole surface in a three-dimensional cylindrical coordinate system; calculating a deviation function δ(θ,z)=r(θ,z)-R of the measured radius and the nominal radius; 0 ; Decompose the deformation characteristics using the N-order Fourier series: r(θ,z)=R 0 +∑[A n (z)cos(nθ)+B n (z)sin(nθ)], n=1,2,...,N, use the least squares method to solve the Fourier coefficients: A n (z)=(2 / π)∫[r(θ,z)-R 0 ]cos(nθ)dθ; B n (z)=(2 / π)∫[r(θ,z)-R 0]sin(nθ)dθ; used to quantitatively describe deformation.

[0021] The present invention is further configured as follows: the laser attitude control process includes: real-time calculation of the detection point (r i ,θ i ,z i ) at the local surface normal vector; ; Construct the angle transformation matrix R=R x (α)×R y (β), realize the spatial attitude adjustment of the laser; calculate the required pitch angle α=arctan(N z / √(N x 2 +N y 2 )) and yaw angle β = arctan(N y / N x ).

[0022] The present invention is further configured as follows: the scanning detection process includes: dynamically adjusting the laser rotation speed according to the deformation amount and the change rate: v(t)=v 0 ×exp(-k 1 |δ(θ,z)|)×exp(-k 2 |dδ / dθ|); acceleration limit control: |dv / dt|≤a max and |d 2 θ / d t 2 |≤ε max , to improve the stability of detection.

[0023] The present invention is further configured as follows: the signal processing and feature extraction process includes: the received signal is processed by the filter equation: H(s)=1 / [1+(s / ω c )^(2n)]; wavelet transform is used for feature extraction: W(a,b)=(1 / √a)∫s(t)ψ((tb) / a)dt; the feature vector is defined as: F=[E 1 ,E 2 ,...,E L ,V 1 ,V 2 ,...,VL]; the defect identification adopts the Mahalanobis distance criterion: D²=(F-μ)^T×Σ^(-1)×(F-μ);

[0024] The real-time control process includes: Improved PID algorithm: u(k)=K p [e(k)+(T / T i )∑e(j)+(T d / T)(e(k)-e(k-1))]; PID parameters are defined by the Ziegler-Nichols method: K p =0.6Ku; T i =0.5T u ; T d =0.125T u .

[0025] In summary, the present invention has the following beneficial effects:

[0026] The modular design realizes high-precision automatic detection of the inside of the expansion flange. The system uses the ME6211C33M5G-N linear regulator to provide a stable 3.3V operating voltage, combined with the GD32F303CBT6 microcontroller for data processing and control, and realizes real-time data interaction with the laser measurement module through the I2C interface. The core innovation lies in its adaptive laser angle adjustment mechanism. By establishing a mathematical model of the inner hole surface in a three-dimensional cylindrical coordinate system and using Fourier series to describe the deformation characteristics, the deformation of the inner hole can be accurately characterized. The servo drive module based on the DRV8813PWPR chip can adjust the attitude of the laser in real time according to the calculation results to ensure that the laser beam is always perpendicular to the inner hole surface. The system realizes precise positioning control of the laser by calculating the local surface normal vector and establishing an angle transformation matrix. During the scanning process, the system also has a speed adaptive adjustment function, which can automatically adjust the scanning speed according to the deformation degree of the detection surface, which not only ensures the detection accuracy but also improves the detection efficiency. This intelligent detection solution effectively solves the problems of positioning difficulties and detection blind spots in traditional laser detection methods when facing deformation of the inner hole of the expansion flange, and significantly improves the accuracy and reliability of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of the expansion flange laser detection system;

[0028] Figure 2 This is the circuit diagram of the power module;

[0029] Figure 3 is the circuit diagram of the microcontroller module;

[0030] Figure 4 This is the circuit diagram of the laser measurement module;

[0031] Figure 5 This is the circuit diagram of the servo drive module;

[0032] Figure 6 It is a structural schematic diagram of the expansion flange. DETAILED DESCRIPTION

[0033] like Figure 1As shown, a laser detection system for expansion flanges includes a power module. A microcontroller module connected to the power module. The microcontroller module and the power module are electrically connected to a laser measurement module for detecting and measuring the internal hole of the expansion flange. The microcontroller module receives data measured by the laser measurement module and performs algorithm processing on the data. The microcontroller module is connected to a servo drive module. The servo drive module is used to transmit the data processed by the microcontroller to the servo drive module, and the position and angle of the laser are automatically adjusted by the servo drive module.

[0034] 1. Power module

[0035] like Figure 2 As shown, the power module is a 3.3V power module of the ME6211C33M5G-N linear regulator. LDO1 converts the input voltage VOUT_OUT (5V power supply) into a 3.3V output (VCC_3V3).

[0036] The input end uses two capacitors (C1, C2) of 10μF and 1μF for filtering, and the output end uses a 1μF capacitor (C3) to stabilize the output voltage. The current from VOUT_OUT enters the VIN pin of LDO1 after being filtered by the input capacitor, and is output from the VOUT pin after voltage stabilization inside the chip, and then filtered by the output capacitor to form a stable 3.3V power supply. The CE pin of the LDO1 chip is connected to a high level to enable operation, VSS is grounded, and the NC pin is left floating.

[0037] The above circuit structure can provide a stable 3.3V output and has protection functions such as over-current, short-circuit and over-temperature.

[0038] 2. Microcontroller module

[0039] like Figure 3 As shown, the microcontroller module is a control module based on the GD32F303CBT6 microcontroller. Among them, the U1 chip (GD32F303CBT6) is powered by a single 3.3V power supply. VCC_3V3 provides the working power supply of the chip through multiple VDD pins (VDD_1, VDD_2, etc.) and the analog power supply pin VDDA, and the corresponding VSS pins (VSS_1, VSS_2, etc.) and the analog ground VSSA are connected to the system ground (GND) to form a complete power supply circuit.

[0040] The OSC_IN and OSC_OUT pins can be used to connect an external crystal oscillator to provide a basic clock source for the U1 chip. The NRST pin is used for system reset control and is used to connect an external RC reset circuit. The BOOT0 pin is used to configure the boot mode. This is an existing technology and will not be described here.

[0041] The U1 chip introduces multiple groups of general IO ports such as PA0-PA15 and PB0-PB15, which are used to connect the laser measurement module and the servo drive module. The measurement data of the laser measurement module is input into the U1 chip, and after the U1 chip data processing, the information is passed to the servo drive module, and then the angle and position of the laser are adjusted accordingly.

[0042] The above-mentioned microcontroller module is capable of executing programs, performing data processing, controlling peripherals and communicating with external devices.

[0043] 3. Laser measurement module

[0044] like Figure 4 As shown, the laser measurement module is powered by a 3.3V power supply and is equipped with a 100nF decoupling capacitor C5 for power supply filtering. The circuit uses an I2C communication interface to achieve data transmission with the main controller through two signal lines SCL (PB6) and SDA (PB7). GPIO1 (PB8) can be used as an interrupt signal or status indication, and XSHUT (PB9) is used to control the enable status of the module.

[0045] All signals are led out through the six-pin interface of the laser connector P1, including power, ground and four signal lines. The current mainly powers the module from VCC_3V3 through the decoupling capacitor C5, and the communication signal is transmitted bidirectionally between the main controller and the module. Through the above settings, the external laser is connected to the connector P1 through the connection form of the socket plug, so as to realize the information interaction between the laser measurement module and the microcontroller module.

[0046] 4. Servo drive module

[0047] like Figure 5 As shown, the servo drive module is a servo motor drive module circuit based on the DRV8813PWPR chip U3. The circuit uses a 3.3V power input and is connected to a 10μH inductor L1 and a 1N4007 diode D 1 The LC filter circuit formed by the two 100nF capacitors (C1 and C2) forms a complete power filter network, which provides a stable operating voltage for driving the DRV8813PWPR chip and achieves the technical purpose of effectively suppressing power ripple and switching noise.

[0048] A 74HC14D Schmitt trigger (U4) is used to process and condition the control signals (PA9, PA8, PB15, PB14, PB13) from the microcontroller, and a 10kΩ pull-up resistor R1 is used to ensure the stability of the signal level.

[0049] The control signal is transmitted to the control end of the DRV8813 driver chip through the interface. The output end of the driver chip leads to multiple groups of motor control signals, including M1_A / M1_B, M2_AZ / M2_BZ, and M3_A / M3_B for controlling three different groups of stepper motors. Among them, M1_A / M1_B is used to control the angle adjustment of the laser. It is a rotary motor that transmits information according to the U1 chip in the microcontroller module and drives the laser to change its emission angle. The M2_AZ / M2_BZ interface is used to control the longitudinal position adjustment of the laser. When entering the expansion flange (such as Figure 6 When the laser is in the inner hole of the expansion flange, the laser scanning measurement is performed in a stepping manner around the inner hole of the expansion flange. M3_A / M3_B is used to control the rotation of the laser. After reaching the fixed position and adjusting the angle of the laser, the motor connected to M3 drives the laser to rotate one circle, and then the inner hole at that position is detected. The above three groups of motors can realize the corresponding defect detection inside the expansion flange at different positions.

[0050] In addition, the corresponding output signal of U3 can be reconnected and returned to the chip U1 to determine the position information and angle information of the laser, which will not be repeated here.

[0051] The above expansion flange laser detection system realizes complete automatic detection function through modular design. The power module with ME6211C33M5G-N linear regulator as the core of the system provides a stable 3.3V working voltage, and the multi-stage filter capacitor design ensures the power supply quality and anti-interference ability of the system. The GD32F303CBT6 microcontroller realizes data acquisition, algorithm processing and drive control, and interacts with the laser measurement module in real time through the I2C interface. The laser measurement part adopts a standardized six-pin interface design, combined with the signal processing circuit and GPIO interrupt mechanism to ensure the accuracy and real-time performance of the measurement data. In addition, the servo drive module based on the DRV8813PWPR chip realizes precise control of the laser angle and position through LC filtering and Schmitt trigger signal conditioning, so that it can be adaptively adjusted according to the changing trajectory of the inner hole of the expansion flange, so that the emission direction of the laser is kept perpendicular to the inner hole of the expansion flange, thereby performing a full-scale scanning and detection of the inner wall of the expansion flange.

[0052] Based on the above-mentioned microcontroller module, the present application provides a data processing method for an expansion flange laser detection system, and further performs algorithmic processing on the data processing of chip U1, so that the laser can automatically adjust its laser emission angle at different positions of the expansion flange and scan and detect the inner hole of the expansion flange.

[0053] This application takes an industrial laser with a wavelength of 1064nm and a power of 100mW as an example, and its working environment temperature is 20±2°C and the humidity is ≤65%RH.

[0054] A data processing method for an expansion flange laser detection system comprises the following steps:

[0055] (1) Surface modeling and deformation analysis

[0056] First, the laser performs a quick pre-scan under the control of the detection system to obtain the actual contour information of the inner hole of the expansion flange. Based on the obtained contour data, the chip U1 calculates the optimal incident angle required for each detection position. In the subsequent fine detection process, the laser adjusts the laser angle in real time according to the calculation results to ensure that the laser beam emission angle is always perpendicular to the inner hole surface.

[0057] The specific contour modeling is as follows: In the three-dimensional cylindrical coordinate system (r, θ, z), the inner hole surface is represented as a continuous function S(r, θ, z). Considering the deviation between the inner hole surface and the ideal cylindrical surface, the deviation function is defined as: δ(θ, z) = r(θ, z) - R 0 , where R 0 is the nominal radius.

[0058] The N-order Fourier series is used to describe the deformation characteristics:

[0059] r(θ,z)=R 0 +∑[A n (z)cos(nθ)+B n (z)sin(nθ)],n=1,2,...,N

[0060] The Fourier coefficients are solved by the least squares method:

[0061] A n (z)=(2 / π)∫[r(θ,z)-R 0 ]cos(nθ)dθ

[0062] B n (z)=(2 / π)∫[r(θ,z)-R 0 ]sin(nθ)dθ.

[0063] In the above-mentioned expansion flange laser detection system, r represents the actual distance from the flange axis to the inner hole surface (mm), θ represents the rotation angle of the laser in the cross section (rad), and z represents the distance coordinate along the flange axis (mm). 0 is the nominal inner diameter size of the flange design, the S(r,θ,z) function describes the overall shape of the inner hole surface, and δ(θ,z) reflects the deviation between the actual radius and the nominal radius. In the Fourier series expansion, An (z) and B n (z) are the cosine term and sine term coefficients, respectively, which characterize the deformation characteristics of different orders. n determines the complexity of the deformation that the system can describe. In the actual detection process, r(θ,z) corresponds to the distance data obtained by the laser ranging sensor, θ and z correspond to the rotation angle and axial position of the laser, respectively. The system needs to calculate the A n (z) and B n (z) Adjust the laser incident angle in real time to ensure detection accuracy.

[0064] (2) Laser attitude control

[0065] At the detection point (r i ,θ i ,z i ), calculate the local surface normal vector:

[0066] The laser attitude is adjusted by the servo drive module, and the angle transformation matrix is: R=R x (α)×R y (β). α and β are obtained by inverse solution of the normal vector:

[0067] α=arctan(N z / √(N x ²+N y ²))

[0068] β = arctan(N y / N x )

[0069] In laser attitude control, (r i ,θ i ,z i ) represents the current detection point coordinates, where r i is the radial distance, θ i is the circumferential angle, z i is the axial position. N(θ,z) is the normal vector of the point, which consists of three components: It represents the partial derivative of radius with respect to angle, reflecting the circumferential rate of change; It represents the partial derivative of the radius with respect to the axial position, reflecting the axial rate of change; r represents the current radius value.

[0070] Denominator Used for vector normalization.

[0071] R represents the attitude transformation matrix, which is the rotation matrix R around the X axis. x (α) and the rotation matrix R around the Y axis y (β). α is the pitch angle, which is composed of the normal vector Z component N zThe arc tangent of the projection length on the XY plane is obtained; β is the yaw angle, which is obtained by the Y component N of the projection of the normal vector on the XY plane. y With X component N x These angle values ​​are directly used to control the servo drive module to achieve precise posture adjustment of the laser.

[0072] The above process describes how the laser adjusts its incident angle in real time according to the surface features. i ,θ i ,z i ), first calculate the normal vector N(θ,z) of this point, which is perpendicular to the inner hole surface. The normal vector consists of three components: radial derivative , the current radius r, and the axial derivative , and normalized by the denominator.

[0073] In order to keep the laser beam perpendicular to the detection surface, the system needs to adjust the laser to a direction consistent with the normal vector, controlling the pitch angle α around the X axis and the yaw angle β around the Y axis respectively. The angle transformation matrix R represents the overall angle adjustment process.

[0074] The α angle is calculated by the angle between the projection of the normal vector on the XY plane and the Z axis, and the β angle is determined by the angle between the projection of the normal vector on the XY plane and the X axis. The above adjustment ensures that the laser beam always hits the detection surface at the best angle, thus obtaining the most accurate measurement results.

[0075] (3) Scanning detection

[0076] After the angle adjustment is completed, the motor drives the laser to rotate 360 ​​degrees, and its rotation speed can be adaptively adjusted according to the actual defect situation. The adjustment equation is:

[0077] v(t)=v 0 ·exp(-k 1 |δ(θ,z)|)·exp(-k 2 |dδ / dθ|)

[0078] Acceleration constraints:

[0079] |dv / dt|≤a max

[0080] |d²θ / dt²|≤ε max

[0081] In the adjustment equation, v(t) represents the real-time rotation speed, v 0 is the initial set speed. δ(θ,z) is the deformation amount defined previously, and dδ / dθ represents the derivative of the deformation amount with respect to the angle, reflecting the severity of the deformation. k 1 and k2 is the attenuation coefficient, which is used to adjust the sensitivity of velocity to deformation and deformation gradient. In the acceleration constraint, dv / dt represents the velocity change rate, and its absolute value must be less than the maximum allowable acceleration a. max ; d²θ / dt² represents the angular acceleration, and its absolute value must be less than the maximum allowable angular acceleration ε max ,These constraints ensure that the system moves smoothly and controllably.,Through the above speed control, the laser automatically decelerates when detecting a large deformation or a region with drastic deformation changes,,improving the measurement accuracy.

[0082] (4) Signal processing and feature extraction

[0083] The received signal is processed by the filter: H(s)=1 / [1+(s / ω c )^(2n)]; wavelet transform is used for feature extraction: W(a,b)=(1 / √a)∫s(t)ψ((tb) / a)dt; its eigenvector: F=[E 1 ,E 2 ,...,E L ,V 1 ,V 2 ,...,V L ]; The Mahalanobis distance criterion is used for defect identification: D²=(F-μ)^T·Σ^(-1)·(F-μ).

[0084] In the above equation, H(s) is the filter transfer function, where ω c is the cutoff frequency, n is the filter order, which is used to suppress high-frequency noise. W(a,b) is the continuous wavelet transform, s(t) is the original signal, ψ is the wavelet mother function, a is the scale parameter controlling the frequency resolution, and b is the translation parameter determining the time position. The eigenvector F is composed of the energy feature E i and variance feature V i In defect identification, D² is the Mahalanobis distance, F is the feature vector of the sample to be identified, μ is the mean vector of the normal sample, and Σ is the covariance matrix.

[0085] The filter first filters the laser receiving signal to remove high-frequency noise and interference during the measurement process. Then, the signal is analyzed at multiple scales through wavelet transform, which can simultaneously obtain the signal's time domain and frequency domain information and extract the local details of the defect characteristics. i and variance feature V iCombined into a feature vector F, comprehensively characterize the defect characteristics. Finally, the Mahalanobis distance criterion is used for defect discrimination. The above process takes into account the correlation between each feature component and is insensitive to the scale of the feature. It can accurately identify abnormal deformation and defects on the inner hole surface, reduce the misjudgment rate, and effectively improve the detection reliability and accuracy of the system.

[0086] (5) Real-time control

[0087] Using improved PID algorithm:

[0088] u(k)=K p [e(k)+(T / T i )∑e(j)+(T d / T)(e(k)-e(k-1))]

[0089] The above PID parameters are integrated by the Ziegler-Nichols method and defined as:

[0090] K p =0.6Ku; T i =0.5T u ; T d =0.125T u .

[0091] The above is an improved PID control algorithm, where u(k) is the control output at the kth moment, e(k) is the error signal, and T is the sampling period. p is the proportionality coefficient, T i is the integration time constant, T d is the differential time constant. The parameters are adjusted by the Ziegler-Nichols method. Where Ku is the critical proportional gain of the system, T u is the critical oscillation period. The proportional term directly reflects the current error, the integral term eliminates the static error, and the differential term improves the dynamic response performance of the system. This control algorithm can achieve precise control of the laser position and posture, ensuring the stability and response speed of the detection system.

[0092] The improved PID algorithm mentioned above is used to control the precise positioning and attitude adjustment of the laser. In the actual detection process, it is necessary to control the rotation speed, tilt angle and axial position of the laser to adapt to the deformation characteristics of the inner hole surface. The algorithm adjusts the speed of the motor and the angle of the servo system by calculating the control output u(k) in real time, where the proportional term ensures the response speed, the integral term eliminates the steady-state error, and the differential term suppresses overshoot and oscillation. The Ziegler-Nichols tuning method ensures the rationality of the control parameters, so that the system can respond quickly and remain stable, thereby achieving smooth movement and precise tracking of the laser.

[0093] Overall working process:

[0094] The expansion flange laser detection system described in the present invention first obtains the inner hole contour information through rapid pre-scanning, and uses Fourier series to establish a mathematical model to describe the surface deformation characteristics. During the detection process, the system calculates the normal vector of the detection point in real time, controls the attitude of the laser through the angle transformation matrix, and ensures that the laser beam is always perpendicular to the detection surface. The laser adopts an adaptive speed control strategy when scanning 360 degrees, and automatically slows down in the deformation-significant area to improve accuracy. The received signal is filtered and wavelet transformed for feature extraction, and the Mahalanobis distance criterion is used for defect discrimination. The entire process is controlled in real time by an improved PID algorithm to ensure accurate adjustment of the laser position and attitude. The system realizes high-precision and high-reliability automated detection by integrating multiple algorithms, and can effectively identify abnormal deformation and defects of the surface of different inner hole shapes caused by different positions of the expansion flange, and has the characteristics of high detection accuracy, strong adaptability and fast response.

[0095] The following is an example calculation of the above algorithm processing, as follows:

[0096] The following example illustrates the specific calculations during the detection process:

[0097] Assume that a nominal inner diameter R is detected 0 =100mm expansion flange is tested at a certain axial position z=50mm.

[0098] 1. Surface modeling calculation example:

[0099] Assuming that the actual radius r=100.5mm is measured at the θ=45° position, the deviation value can be calculated:

[0100] δ(45°,50mm)=100.5mm-100mm=0.5mm;

[0101] Taking N=3rd order Fourier series, the expansion of a certain measurement cross section is:

[0102] r(θ,50)=100+0.3cos(θ)+0.2sin(θ)+0.15cos(2θ)+0.1sin(2θ)+0.05cos(3θ)+0.03sin(3θ).

[0103] 2. Attitude control calculation:

[0104] At the point (100.5mm, 45°, 50mm), the calculated normal vector may be:

[0105] N=[-0.2,100.5,-0.1] / √(0.04+10100.25+0.01)=[-0.00199,0.99999,-0.00099];

[0106] The angle that needs to be adjusted is calculated from this:

[0107] α=arctan(-0.1 / √(0.04+10100.25))≈-0.057°;

[0108] β=arctan(-0.2 / 100.5)≈-0.114°.

[0109] 3. Speed ​​control example:

[0110] Assume that the initial speed v 0 =10rad / s,k 1 = 0.5, k 2 =0.3,

[0111] When the deformation |δ|=0.5mm and the deformation gradient |dδ / dθ|=0.2mm / rad are detected:

[0112] v(t)=10×exp(-0.5×0.5)×exp(-0.3×0.2)≈7.79rad / s.

[0113] Feature extraction:

[0114] Assume that the measured signal at a certain point is transformed by wavelet to obtain the energy characteristics of three layers of decomposition:

[0115] E 1 =0.8,E 2 =0.5,E 3 =0.2;

[0116] Variance characteristics: V 1 =0.05,V 2 =0.03,V 3 =0.01;

[0117] Feature vector F = [0.8, 0.5, 0.2, 0.05, 0.03, 0.01];

[0118] If the normal sample mean vector μ=[0.7, 0.4, 0.1, 0.04, 0.02, 0.01],

[0119] Calculating the Mahalanobis distance can determine whether there is a defect.

[0120] PID control:

[0121] Assume that the angle error e(k)=0.2° at a certain moment and the sampling period T=0.01s;

[0122] If Ku=1.5, T u =0.4s, then:

[0123] K p =0.9, T i =0.2s, T d =0.05s;

[0124] u(k)=0.9[0.2+(0.01 / 0.2)∑e(j)+(0.05 / 0.01)(0.2-0.18)] is used to adjust the motor output torque.

[0125] This specific embodiment is merely an explanation of the present invention and is not a limitation of the present invention. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed. However, as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A data processing method for an expansion flange laser detection system, characterized in that: include: A power supply module, which is a 3.3V power supply module of a linear regulator, and is used to provide a stable working power supply for the system; A microcontroller module, the microcontroller module is electrically connected to the power module, and the microcontroller module is used to control the operation of the entire system and process detection data; A laser measurement module, which is electrically connected to the microcontroller module and is used to emit a laser beam and receive a reflected signal to detect and measure the internal hole of the expansion flange; A servo drive module, the servo drive module is electrically connected to the microcontroller module, the servo drive module is a servo motor drive module based on the DRV8813PWPR chip, and the spatial position and incident angle of the laser are accurately adjusted by driving multiple motors; The servo drive module comprises: The first motor interface drives the stepper motor to achieve precise adjustment of the laser pitch angle; The second motor interface drives the DC motor to control the position movement of the laser on the longitudinal axis; The third motor interface drives the stepper motor to realize the rotation scanning motion of the laser around the axis; The following steps are also included: (1) Surface modeling and deformation analysis: The laser performs a rapid pre-scan under the control of the detection system to obtain the actual contour information of the inner hole of the expansion flange. Based on the obtained contour data, the incident angle required for each detection position is calculated; (2) Laser attitude control: adjusting the angle state of the laser according to the calculated incident angle of each detection point; (3) Scanning detection: After the angle adjustment is completed, the motor drives the laser to rotate 360 ​​degrees, and its rotation speed is adaptively adjusted according to the actual defect situation; (4) Signal processing and feature extraction: Filter the laser receiving signal to remove high-frequency noise and interference during the measurement process, analyze the signal through wavelet transform, obtain the time domain and frequency domain information of the signal, and extract the details of the defect characteristics; (5) Real-time control: Through the improved PID algorithm, the precise positioning and attitude adjustment of the laser are controlled, and the rotation speed, tilt angle and axial position of the laser are controlled to adapt to the deformation characteristics of the inner hole surface of the expansion flange.

2. The data processing method according to claim 1, characterized in that: The surface modeling and deformation analysis include: Establish the inner hole surface mathematical model S(r,θ,z) in the three-dimensional cylindrical coordinate system; Calculate the deviation function of the measured radius and the nominal radius δ(θ,z)=r(θ,z)- R0; Decomposition of deformation features using N-order Fourier series: r(θ,z)=R0+∑[A n (z)cos(nθ)+B n (z)sin(nθ)],n=1.2,…,N, Solve for the Fourier coefficients using the least squares method: A n (z)=(2 / π)∫[r(θ,z)-R0]cos(nθ)dθ; B n (z)=(2 / π)∫[r(θ,z)-R0]sin(nθ)dθ; Used to quantitatively describe deformation.

3. The data processing method according to claim 1, characterized in that ,The laser attitude control process includes: Real-time calculation of detection points (r i ,θ i ,z i ) at the local surface normal vector; construct the angle transformation matrix R = R x (α)×R y (β), realize the spatial attitude adjustment of the laser; calculate the required pitch angle α=arctan(N z / √(N x 2 +N y 2 )) and yaw angle β = arctan(N y / N x ); The scanning detection process includes: Dynamically adjust the laser rotation speed according to the deformation amount and change rate: v(t)=v0×exp(-k1|δ(θ,z)|)×exp(-k2|dδ / dθ|); acceleration limit control: |dv / dt|≤a max and |d2θ / d t 2 |≤ε max , to improve the stability of detection.

4. The data processing method according to claim 1, characterized in that: The signal processing and feature extraction process includes: The received signal is processed by the filter: H(s)=1 / [1+(s / ω c )^(2n)]; wavelet transform is used for feature extraction: W(a,b)=(1 / √a)∫s(t)ψ((tb) / a)dt; the feature vector is defined as: F=[E1,E2,...,E L ,V1,V2,...,VL]; the defect identification adopts the Mahalanobis distance criterion: D²=(F-μ)^T×Σ^(-1)×(F-μ); u(k)=K p [e(k)+(T / T i )∑e(j)+(T d / T)(e(k)-e(k-1))]; PID parameters are defined by the Ziegler-Nichols method: K p =0.6Ku; T i =0.5T u ; T d =0.125T u .

Images