Highly integrated detection vehicle combining autonomous walking and active excitation

Abstract

The invention discloses a highly integrated detection vehicle combining autonomous walking and active excitation, which integrates an autonomous walking system, an active excitation system and a detection identification system, and is characterized in that an autonomous walking module realizes automatic navigation, obstacle avoidance and path planning through a sensor and an intelligent control algorithm; the active excitation module controls excitation equipment through a user-defined algorithm, and generates excitation signals with specific frequency according to detection requirements of different to-be-detected structures, so that damage information of the to-be-detected structures is fed back in a high signal-to-noise ratio manner; the detection and identification module collects information through an accelerometer arranged at a vibration excitation position, extracts rigidity sudden change information of the to-be-detected structure through a time-frequency analysis result, determines a damage position and evaluates the damage degree. The overall structural design of the detection vehicle is optimized, compared with a traction type passive excitation detection vehicle, the detection precision, the detection efficiency and the portability are improved, and the detection vehicle is suitable for automatic damage detection of large civil structures such as bridges, wharfs and roads.

Description

Technical Field

[0001] The present invention belongs to the technical field of bridge inspection, and particularly relates to a highly integrated inspection vehicle that combines autonomous walking with active excitation. The vehicle achieves autonomous walking through wheel hub motors and an intelligent control algorithm, uses an excitation device provided on a single excitation wheel for active excitation at a fixed excitation frequency, and collects information through an accelerometer provided at the excitation location. Time-frequency analysis results are used to extract information on sudden changes in bridge stiffness, thereby determining the location of bridge damage and assessing the extent of the damage. Background Art

[0002] Currently, the identification of highway bridge defects still relies primarily on manual methods. However, the sheer number of bridges and the vast mileage of the bridges present significant challenges for bridge maintenance. Failure to promptly detect defects and assess their severity to develop appropriate countermeasures can lead to accidents and loss of life and property. Furthermore, manual identification poses significant safety risks for certain bridges with hazardous conditions. Therefore, utilizing automated inspection technologies to improve efficiency and accuracy has become an essential option for bridge defect detection.

[0003] Patent application number CN201080020956.8, titled "A Structural Damage Detection System, Equipment, and Method," proposes a towed active vibration vehicle. Meanwhile, patent application number CN202210127441.2, titled "A Passive Vibration Bridge Damage Location Method," introduces an active, walking passive vibration vehicle. Both vehicles collect signals from the inspection area through sensors and calculate damage indication values ​​from the spectral envelope. They then identify areas where the indication value suddenly drops as damage locations and assess the severity of the damage based on the degree of sudden change.

[0004] However, these inspection vehicles have some limitations. First, the speed control of the tractor is relatively subjective for towing active vibration vehicles, and usually requires an experienced driver to achieve smooth driving at a specific speed. This limits the driving stability and damage identification accuracy of the inspection vehicle. On the other hand, active walking passive vibration vehicles can only generate an excitation force of a specific frequency when traveling at a constant speed. The magnitude of this excitation force is limited by the axle weight of the excitation wheel and the driving speed, making it difficult to flexibly adjust. In addition, this type of inspection vehicle is also unable to achieve signal recognition with a high signal-to-noise ratio, resulting in limited damage identification accuracy, which to some extent limits the flexibility and portability of the system. Summary of the Invention

[0005] In response to the above problems, the present invention provides a highly integrated inspection vehicle that combines autonomous walking and active vibration. The autonomous walking of the inspection vehicle is achieved through wheel hub motors and intelligent control algorithms. The excitation equipment set on a single excitation wheel is used for active excitation with a fixed excitation frequency. The accelerometer set at the excitation point collects information. The acceleration signal of the excitation vehicle is processed to extract the bridge stiffness mutation information, determine the location of bridge damage, and evaluate the degree of damage.

[0006] A highly integrated inspection vehicle combining autonomous walking and active vibration excitation, comprising: an autonomous walking module, an active vibration excitation module, and a signal acquisition and processing module, wherein: The autonomous driving module uses sensors and intelligent control algorithms to ensure stable autonomous driving in various environments through real-time data processing. It also performs emergency obstacle avoidance based on received feedback information, significantly improving the stability of driving speed and thus the accuracy of damage location. The active vibration module can accurately adjust the frequency and amplitude of the excitation signal based on the characteristics of the different structures to be tested through a user-defined control algorithm, independent of the speed of the inspection vehicle, ensuring high-quality damage information feedback under various working conditions, thereby significantly improving inspection accuracy; The signal acquisition and processing module collects information through an accelerometer set at the excitation point. The accelerometer is used to collect the vibration signal of the detection vehicle. The signal is determined by the maximum point of the equivalent acceleration extracted from the time-frequency analysis results to determine the moment of stiffness mutation. The inflection point in the corresponding power spectrum diagram is used to determine the location of bridge damage and evaluate the degree of damage.

[0007] Furthermore, the autonomous walking module adopts a three-wheel layout, including a pair of front wheels driven by hub motors and a driven rear wheel for excitation; wherein, the front wheels provide forward power for the inspection vehicle and achieve precise steering through differential control. Users can customize the walking mode of the hub motor through the program to adapt to different inspection needs; a gyroscope is provided on the axle of the front wheel, and the stability and smooth driving of the inspection vehicle during driving are ensured through the feedback control system; the driven rear wheel provides planar support for the inspection vehicle, maintaining the structural stability and driving balance of the inspection vehicle.

[0008] Furthermore, the active excitation module includes a top rod type excitation device for actively exciting the bridge deck with a fixed excitation frequency. The excitation device is placed directly above the driven rear wheel through a limiting sleeve. A fixed strap is extended upward from both ends of the central bearing of the driven rear wheel through an end groove to be connected to the limiting sleeve to provide a knocking plane for the excitation device. The excitation device achieves precise knocking of the fixed strap through a mechanical top rod, and transmits the knocking signal to the bridge deck through the fixed strap and the driven rear wheel.

[0009] Furthermore, the signal acquisition and processing module performs data processing on the acquired acceleration signal in the following manner: (1) The acceleration signal is processed using time-frequency methods (such as short-time Fourier transform) to obtain the time domain signal J signals (corresponding to different positions). Calculate the spectrum of each part of the signal and record it in the vector ; and calculate the total power of each signal , mean and standard deviation ,in K is the number of data points in the analysis band; (2) Calculate the mean value in each analysis frequency band , standard deviation , coefficient of variation , mean of the coefficient of variation and standard deviation The coefficient of variation is used to measure the fluctuation of the spectrum at that point. If the fluctuation is large, a sudden change in stiffness may occur. (3) Record the position point vector after removing noise by the following filtering algorithm , the filtering parameters are : ; (4) Calculate the equivalent acceleration of each signal , calculate its mean . Record the peak point position to vector ,in is the sensitive frequency segment vector; (5) The location with obvious stiffness mutation is , whose weighted stiffness ratio is evaluated by: , The dimensionless coefficients G and H are constant coefficients.

[0010] (6) The weighted stiffness ratio can be reversely calculated into the actual stiffness ratio using the following formula: , in( EI ) d is the damage stiffness, EI is the original bending stiffness, w i is the window function, N s is the position vector of the damage in the transmission function.

[0011] Furthermore, the internal equipment has an optimized layout, which includes active excitation equipment, an on-board power supply, a counterweight device and an equipment support frame. The internal equipment achieves structural balance of the entire vehicle through optimized layout, ensuring that the mode generated when the equipment is running is a single mode within a specific detection frequency range, and the front and rear axle weight ratio is approximately 1:1, thereby increasing the driving stability of the inspection vehicle.

[0012] The beneficial technical effects brought about by the present invention are: 1) The present invention has flexible mobility through the autonomous walking module and can adapt to complex terrain and environments. Compared with traditional traction-type active vibration vehicles, the improved speed stability can effectively improve the accuracy of damage location.

[0013] 2) The active excitation module can effectively excite the bridge deck at different speeds, avoiding strict speed restrictions. Compared with the passive excitation vehicle, the frequency and magnitude of the excitation force are no longer determined by the speed of the inspection vehicle, further improving the flexibility and accuracy of the inspection.

[0014] 3) The real-time signal acquisition and analysis module can quickly and accurately locate bridge damage, ensuring the reliability and accuracy of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a front view of the bridge damage inspection vehicle body; Figure 2 It is a side view of the bridge damage inspection vehicle body; Figure 3 This is a schematic diagram of the basic model of the knock scanning method; Figure 4 is a schematic diagram of the damaged beam model; Figure 5 It is a control flow diagram; Description of reference numerals: 1-Vibrator, 2-Fixed sleeve, 3-Vibration wheel, 4-Battery and power amplifier, 5-Hardware box, 6-Drive wheel, 7-Frame, 8-Counterweight connection part. DETAILED DESCRIPTION

[0016] In order to enable those skilled in the art to better understand the technical solution of the present invention, the structure of the inspection vehicle and the principle of the active vibration bridge flaw detection method will be described in detail below.

[0017] The present invention provides a bridge damage detection vehicle body structure, including: a driving part, a counterweight connection part and an excitation part. Figure 1Schematic diagram of the body structure of the bridge damage inspection vehicle. The autonomous driving system consists of two sets of drive wheels 6 fixed to the frame 7, two 10-inch 200W hub motors, and a hardware box 5. The frame 7 can be equipped with hydraulic nitrogen shock absorbers to reduce the noise of the drive wheels, thereby reducing interference with the excitation signal. A dedicated position for installing an attitude sensor can be set on the top of the frame 7. The attitude sensor is used to implement semi-supervised inertial navigation functions. The hardware box 5 can be installed with batteries, power amplifiers and control modules, and provides storage space for the power amplifier of the active excitation system.

[0018] like Figure 2 As shown, the counterweight connection section 8 consists of a square channel member and a strap. The strap is welded to the square channel member, which connects the front body drive section and the rear body excitation section via four slots and bolts. The excitation section includes an electromagnetic exciter 1 positioned directly above the excitation wheel 3. The excitation wheel axle is equipped with a sensor mounting point, which can be an IEPE accelerometer. The excitation section is designed to ensure that low-order modes contain only vertical vibration modes as much as possible.

[0019] After installing acceleration sensors and signal analysis components on the body of the bridge flaw detection vehicle, by analyzing the degree of mutation at the stiffness mutation point on the beam, we can approximate the quadratic curve relationship between the mutation degree and the vehicle acceleration amplitude or the square root of the power spectrum density, thereby finding the corresponding characteristic points for defect location.

[0020] The rapid bridge stiffness abnormal point detection mode of the bridge damage detection vehicle is as follows: The working mode of this method is as follows Figure 5 As shown in the figure, the module is divided into the user end, control mode selection, host computer, slave computer, and underlying hardware according to the command transmission sequence. The user end must select the control mode based on the specific scenario. For example, when moving equipment quickly at the inspection site, short-range remote control can be used to directly control the slave computer, achieving flexible steering or U-turns. During the inspection process, semi-supervised inertial navigation mode is used. In this mode, the inspection vehicle mainly relies on inertial navigation to conduct inspections at a fixed speed. However, if the inertial navigation sensor deviates significantly due to expansion joints, the user end can make timely corrections.

[0021] The host computer's functions include receiving and processing user commands, collecting inertial navigation sensor data and detection data, and performing data analysis and storage. In terms of control flow, after receiving user commands, the host computer sends detection tasks to the slave computer for execution, receives feedback, and ultimately transmits the processing results to the user for review.

[0022] The functions of the lower computer mainly include completing various hardware control tasks, such as driving the motor, vibrator control, speed acquisition, power and temperature monitoring, and obstacle avoidance and emergency stop.

[0023] The structural stiffness identification algorithm principle of the bridge damage detection vehicle is as follows: Figure 3 The basic model of the tap scanning method, in which the inspection vehicle is simplified as a spring-mass system and moves at a constant speed v It moves on a simply supported Euler-Bernoulli beam. The mass of the inspection vehicle is M V , the stiffness of the suspension system is k V , the damping coefficient is μ V The bending stiffness of the beam is EI , the mass per unit length is m , the damping coefficient is μ B , the road roughness is expressed as r ( x Active exciting force F V The passive excitation force can be applied directly to the vehicle body through the exciter. F T Impact from the wheel knock tooth and the beam. The wheel knock tooth is simplified to the contact interface between the vehicle and the beam.

[0024] All displacements and forces are y Direction is positive; subscript V and B Represent the vehicle body and beam respectively. The static equilibrium state before the inspection vehicle enters the beam is regarded as the zero displacement state of the inspection vehicle and beam. x = vt The following equation of motion is established:

[0025]

[0026] Among them: the dot represents the time t Find partial derivatives; represent Material derivative at ; δ is the Dirac delta function. According to the force balance condition in the y direction, the support force of the beam on the wheel at the contact interface can be obtained. F :

[0027] The pulse excitation applied to the inspection vehicle can be approximately expanded into a series form:

[0028] If the preset knock force FV Passive knock force caused by road roughness F T Large enough, that is A i >> ε s , the influence of roughness can be ignored, and the mass of the exciting wheel can also be regarded as part of the mass of the inspection vehicle; therefore, according to ~, the origin displacement impedance of the beam at the exciting wheel can be obtained:

[0029] It can be seen that the acceleration of the car is related to the displacement impedance of the origin, and therefore contains information about local damage:

[0030] Since the bridge is very rigid, it can be assumed that the deformation of the bridge under the impact of the car is very small and the vibration response is linear. Therefore, the bridge displacement can be expressed as:

[0031] in: The bridge j order mode; q Bj ( t ) is the j The modal coordinates.

[0032] Substitute equation (7) into equations (1) to (3), and then multiply both sides of the equation by φ j ( x ) and then integrating along the length of the beam, we get:

[0033] in, ω Bj The bridge j Order natural circular frequency:

[0034] Under the premise that the car moves at a low speed and the mass of the inspection car is relatively small compared to the bridge, it is easy to meet , so we can ignore the , so the decoupled equation can be obtained. Substituting (4) into (8) and applying Duhamel integral, the forced vibration solution can be obtained (Note: Because slight local damage will not cause ω Bj There is a big change, so it is believed that ω Bj is a constant; in addition, the initial displacement and initial velocity of the beam are considered to be zero):

[0035] in:

[0036] Substituting (10) into (1), we obtain:

[0037] in: 2ξ V =μ V / M V ; k It is related to the position on the beam, so the equivalent natural frequency is a function of time:

[0038] It can be seen that Equation (12) is a second-order linear ordinary differential equation with variable coefficients. In order to derive the analytical solution, this mechanical model equates the actual damaged beam + intact trolley to an intact beam + damaged trolley, that is, k Suspension stiffness of the car k V It is related to the damage of the beam at the contact point, so the equivalent natural frequency in (12) is a function of time. t = t 0 appears, and assumes ω V In the time interval Δ t The inside is a constant, and the detection vehicle is t = t +Δ t The displacement at the moment can be calculated by Duhamel integration. Considering that under the impact force amplification, the response of the detection vehicle is mainly the impact force frequency ω 0 component, so the displacement response can be simplified as:

[0039] in , , the coefficient is related to the specific structure to be measured.

[0040] Assume that the stiffness distribution of the damaged beam is as follows Figure 4 As shown, according to the stiffness equivalent model:

[0041]

[0042]

[0043] in: θ= ( EI )d / ( EI ), D s is the damage coefficient, which is related to the cross-sectional structure of the beam and can be obtained from Figure 4 The simply supported beam of is analytically obtained under unit moving load:

[0044] in: α = l 1 / L , β = ( l 2 -l 1) / L , γ = x / L=vt / L .

[0045] Substituting equations (15) to (18) into equation (14) and performing the second-order time differentiation of equation (14), we can obtain the relationship between the car acceleration and the beam stiffness variation coefficient:

[0046] The coefficient A , B and C Depends on the specific structure to be tested.

[0047] In actual detection, low-frequency noise has a greater impact on time domain signals. Therefore, the STFT processing method is usually used to analyze signals in higher sensitive frequency bands. The relationship between the effective value of acceleration and the coefficient of variation of beam stiffness can be obtained:

[0048] The coefficient It is the coefficient to be calibrated for the calibration system.

[0049] Based on the above principles, it can be seen that the degree of sudden change in stiffness at a beam point is approximately quadratically related to the square root of the vehicle acceleration amplitude or power spectrum density. The key lies in finding the characteristic point corresponding to the derivative term. For actual bridges, acceleration is easily affected by low-frequency noise, so power spectrum density can be used for analysis. To distinguish different degrees of sudden change in stiffness, the following algorithm is proposed for classification: (1) The acceleration signal is processed using time-frequency methods (such as short-time Fourier transform) to obtain the time domain signal J signals (corresponding to different positions). Calculate the spectrum of each part of the signal and record it in the vector ; and calculate the total power of each signal , mean and standard deviation ,inK is the number of data points in the analysis band; (2) Calculate the mean value in each analysis frequency band , standard deviation , coefficient of variation , mean of the coefficient of variation and standard deviation The coefficient of variation is used to measure the fluctuation of the spectrum at that point. If the fluctuation is large, a sudden change in stiffness may occur. (3) Record the position point vector after removing noise by the following filtering algorithm , the filtering parameters are : (twenty one); (4) Calculate the equivalent acceleration of each signal , calculate its mean . Record the peak point position to vector ,in is the sensitive frequency segment vector; (5) The location with obvious stiffness mutation is , whose weighted stiffness ratio is evaluated by: (twenty two) The dimensionless coefficients G and H are constant coefficients.

[0050] (6) The actual stiffness ratio can be calculated by reverse calculation using the following formula: (twenty three) in( EI ) d is the damage stiffness, EI is the original bending stiffness, w i is the window function, N s is the position vector of the damage in the transmission function.

[0051] The examples of the present invention are described in detail above in conjunction with the embodiments, but the present invention is not limited to the above examples. Various changes can be made within the knowledge of ordinary technicians in this field without departing from the purpose of the present invention, and should also be regarded as the scope of protection of the present invention.

Claims

1. A highly integrated inspection vehicle combining autonomous walking and active vibration, comprising: The autonomous walking module, active vibration module and signal acquisition and processing module are characterized by: The autonomous driving module uses sensors and intelligent control algorithms to ensure stable autonomous driving in various environments through real-time data processing. It also performs emergency obstacle avoidance based on received feedback information, significantly improving the stability of driving speed and thus improving the accuracy of damage location. The active excitation module precisely adjusts the frequency and amplitude of the excitation signal based on the characteristics of the structure under test through a user-defined control algorithm, independent of the speed of the inspection vehicle. This ensures that damage information of the structure under test is fed back with a high signal-to-noise ratio under various working conditions, thereby significantly improving inspection accuracy. The signal acquisition and processing module collects the vibration signal of the detection vehicle through an accelerometer set at the excitation point, determines the moment of stiffness mutation detected by the maximum point of the equivalent acceleration extracted from the time-frequency analysis results, and determines the bridge damage location and assesses the damage extent at the corresponding inflection point in the power spectrum.

2. The highly integrated inspection vehicle combining autonomous walking and active excitation according to claim 1 is characterized in that: The autonomous walking module adopts a three-wheel layout, including a pair of front wheels driven by hub motors and a driven rear wheel for excitation; the front wheels provide forward power for the inspection vehicle and achieve precise steering through differential control. Users can customize the walking mode of the hub motors through programs to adapt to different inspection needs; a gyroscope is set on the axle of the front wheel, and the stability and smooth driving of the inspection vehicle during driving are ensured through a feedback control system; the driven rear wheel provides planar support for the inspection vehicle, maintaining the structural stability and driving balance of the inspection vehicle.

3. The highly integrated inspection vehicle combining autonomous walking and active excitation according to claim 1 is characterized in that: The active excitation module includes a jack-type excitation device for actively exciting the bridge deck at a fixed excitation frequency. The excitation device is arranged directly above the driven rear wheel through a limiting sleeve. A fixed plate extends upward from both ends of the central bearing of the driven rear wheel through an end groove to be connected to the limiting sleeve, providing a knocking plane for the excitation device. The excitation device achieves precise knocking of the fixed plate through a mechanical jack, and transmits the knocking signal to the bridge deck through the fixed plate and the driven rear wheel.

4. The highly integrated inspection vehicle combining autonomous walking and active excitation according to claim 1 is characterized in that: The signal acquisition and processing module performs data processing on the collected acceleration signal as follows: (1) The acceleration signal is processed using the time-frequency method to obtain the time domain signal corresponding to different positions. J signals, calculate the spectrogram of each part of the signal, and record it in the vector ; And calculate the total power of each signal , mean and standard deviation ,in K is the number of data points in the analysis band; (2) Calculate the mean value in each analysis frequency band , standard deviation , coefficient of variation , mean of the coefficient of variation and standard deviation ,The coefficient of variation is used to measure the fluctuation of the spectrum at that point. Larger fluctuations correspond to sudden changes in stiffness; (3) Record the position point vector after removing noise by the following filtering algorithm , the filtering parameters are : ； (4) Calculate the equivalent acceleration of each signal , calculate its mean , record the position of the peak point to the vector ,in is the sensitive frequency segment vector; (5) The location where the stiffness mutation occurs is , and its weighted stiffness ratio is expressed by the following formula: ， The dimensionless coefficients G and H are constant coefficients; (6) The weighted stiffness ratio is reversely calculated to obtain the actual stiffness ratio using the following formula: ， in( EI ) d is the damage stiffness, EI is the original bending stiffness, w i is the window function, N s is the position vector of the damage in the transmission function.

5. The highly integrated inspection vehicle combining autonomous walking and active excitation according to claim 1 is characterized in that: The internal equipment achieves vehicle structural balance through optimized layout, ensuring that the mode generated when the equipment is running is a single mode within a specific detection frequency range, and the front and rear axle weight ratio is approximately 1:1.

Images