System and method for detecting compressive capacity of highway pavement

Abstract

The invention discloses a system and a method for detecting the pressure resistance of a highway pavement. The system comprises a pressure sensor array, an image acquisition device, a positioning module, a data preprocessing module, a data analysis module and an output module. When a vehicle is detected to run at a constant speed, all the modules work cooperatively, and the multiple sensors collect multi-source data such as pavement mechanical response, apparent conditions and temperature synchronously in a space-time mode. Through data preprocessing and image recognition analysis, a regional compressive index RCI is calculated by fusing parameters such as a pressure concentration ratio, a crack density and a temperature gradient based on a random forest algorithm. According to the method, the problems of low efficiency, one-sided data and the like of traditional detection are solved, three-in-one evaluation of mechanical response-apparent damage-material performance is realized, the dynamic detection speed is increased by more than 3 times, the precision reaches + / -2.3%, the remaining service life of the pavement can be predicted, the method is suitable for multi-scene road detection, and deep defects such as hidden cavities of a base layer can be effectively identified.

Description

Technical Field

[0001] The present invention relates to the technical field of highway detection, and specifically refers to a highway pavement compressive capacity detection system and a detection method. Background Art

[0002] In the field of highway construction and maintenance, accurately detecting the pavement condition is crucial for ensuring the safety and durability of the road. Currently, common highway detection methods mostly rely on manual coring or single-sensor fixed-point measurement methods. The manual coring method is not only cumbersome and inefficient in operation, but also causes irreversible damage to the pavement, and the cored samples are limited, making it difficult to comprehensively reflect the overall situation of the pavement; single-sensor fixed-point measurement has one-sided data and cannot obtain comprehensive information on different areas of the pavement. At the same time, both of them cannot effectively reflect the impact of dynamic loads on the pavement.

[0003] Although some vehicle-mounted detection schemes have been proposed, which improve the detection efficiency to a certain extent, this scheme lacks a multi-dimensional data fusion mechanism and cannot fully integrate various detection data to comprehensively evaluate the pavement condition. And it completely does not consider the influence of temperature on the performance of pavement materials, resulting in great limitations in the detection results. In addition, traditional detection methods are difficult to achieve early warning of hidden damage to the pavement, and are often discovered only when the disease develops to a more serious stage. This not only increases the road maintenance cost, but also may pose a threat to traffic safety. Therefore, there is an urgent need for an innovative highway pavement compressive capacity detection technology to overcome the many defects of the existing technology and achieve efficient, comprehensive and accurate highway detection. Summary of the Invention

[0004] (I) Technical Problem

[0005] The present invention aims to provide a highway pavement compressive capacity detection system and method. By synchronously collecting multi-source data in space and time through multiple sensors and performing fusion analysis, it overcomes problems such as low efficiency and one-sided data in traditional detection, realizes a trinity evaluation of "mechanical response - apparent damage - material performance", improves the detection speed and accuracy, predicts the pavement life and identifies deep defects, and is applicable to highway detection in multiple scenarios.

[0006] (II) Technical Content

[0007] To solve the above technical problems, the technical solution of the present invention is: a highway pavement compressive capacity detection system, including the following modules:

[0008] Pressure sensor array: composed of piezoelectric sensors distributed in a grid pattern, integrated on the chassis of the detection vehicle, covering the entire cross-section of the detection area;

[0009] Image acquisition device: including a high-resolution industrial camera and an infrared thermal imager, installed on the top bracket of the detection vehicle at a 30° tilt angle;

[0010] Positioning module: It adopts the integration of RTK-GNSS positioning unit and inertial navigation system to record the detected position coordinates in real time;

[0011] Data preprocessing module: It is equipped with a digital filtering circuit and an AD converter to perform noise reduction and normalization processing on the original pressure signal;

[0012] Data analysis module: It has a built-in compressive strength index calculation model, and fuses pressure distribution data and image crack characteristics through machine learning algorithms;

[0013] Output module: It includes an in-vehicle display terminal and a wireless transmission unit, and supports the real-time generation of detection reports and cloud data synchronization.

[0014] Furthermore, the pressure sensor array is encapsulated with a flexible PCB substrate, the spacing between single sensor units is 15cm×15cm, the range is 0-50MPa, and the sampling frequency is ≥200Hz.

[0015] Furthermore, the image acquisition device is equipped with an active fill light system, and uses 650nm band laser to assist imaging in low light environments.

[0016] On the other hand, the present invention provides a method for detecting the compressive capacity of highway pavement, including the following steps:

[0017] Step S1: The detection vehicle travels at a constant speed of 20-40km / h, and synchronously activates the pressure sensor array and the image acquisition device;

[0018] Step S2: Establish a spatial coordinate system through the positioning module, and dynamically bind the pressure data with the geographical location information;

[0019] Step S3: The data preprocessing module uses wavelet transform algorithm to eliminate environmental vibration noise and generate a standardized pressure distribution map;

[0020] Step S4: The image acquisition device obtains the visible light image and infrared thermal map of the road surface, and identifies the crack morphology and temperature abnormal area through a convolutional neural network;

[0021] Step S5: The data analysis module fuses the pressure concentration, crack density, and temperature gradient parameters, and calculates the regional compressive strength index RCI based on the random forest algorithm;

[0022] Step S6: The output module generates a three-dimensional visualization report including a disease positioning map, an RCI cloud map, and maintenance suggestions.

[0023] Furthermore, the calculation formula of the compressive strength index RCI in step S5 is:

[0024] RCI=α·(1-P / P_max)+β·(1-C / C_max)+γ·(1-ΔT / ΔT_max)

[0025] Where α, β, and γ are weight coefficients and α+β+γ=1, P is the measured pressure value, C is the crack density, and ΔT is the temperature gradient.

[0026] (III) Technical Effect

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] 1. Comprehensive multi-dimensional assessment: Through the synchronous collection of multi-source data such as pressure, image, temperature, etc. by multiple sensors in time and space, a comprehensive assessment of the trinity of "mechanical response-apparent damage-material performance" is achieved, overcoming the problem of one-sided data in traditional detection methods and being able to more accurately reflect the real condition of the road surface.

[0029] 2. Efficient and high-precision detection: The dynamic detection speed is more than 3 times faster than the traditional method, which greatly improves the detection efficiency and meets the needs of large-scale highway detection. At the same time, the detection accuracy reaches ±2.3%, ensuring the reliability of the detection data and providing solid data support for road maintenance decisions.

[0030] 3. Life prediction and early warning: The compression index model integrates the constitutive relationship of material mechanics. It can not only evaluate the current pavement compression resistance, but also predict the remaining service life of the pavement, realize early warning of hidden damage to the pavement, facilitate early maintenance measures, reduce road maintenance costs, and ensure road safety.

[0031] 4. Wide applicability: The modular design supports rapid modification of the inspection vehicle and can be flexibly applied to road inspection in various scenarios such as highways and municipal roads. It has strong adaptability and reduces the application threshold of the inspection equipment.

[0032] 5. Deep defect identification: Through infrared thermal imaging and pressure distribution correlation analysis, deep defects such as hidden cavities in the base layer can be effectively identified, which makes up for the deficiency of traditional detection methods that are difficult to detect deep diseases, and helps to timely discover and deal with potential road safety hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a system schematic diagram of a highway pavement compression resistance detection system of the present invention.

[0034] Figure 2 The present invention is a schematic flow chart of a method for detecting the compressive strength of a highway pavement. DETAILED DESCRIPTION

[0035] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0036] Combined with Figure 1 To Attachment Figure 2 , a highway pavement compression resistance detection system, including the following modules:

[0037] Pressure sensor array 101: composed of piezoelectric sensors distributed in a grid shape, integrated in the chassis of the detection vehicle, covering the entire cross-section of the detection area. The pressure sensor array 101 is packaged using a flexible PCB substrate, the spacing between a single sensor unit is 15cm×15cm, the range is 0-50MPa, and the sampling frequency is ≥200Hz;

[0038] Image acquisition device 102: includes a high-resolution industrial camera and an infrared thermal imager, which is installed on the top bracket of the inspection vehicle at an inclination angle of 30°. The image acquisition device 102 is equipped with an active fill light system, which uses 650nm band laser to assist imaging in low illumination environment;

[0039] Positioning module 103: adopts RTK-GNSS positioning unit and inertial navigation system fusion to record the detection position coordinates in real time;

[0040] Data preprocessing module 104: equipped with a digital filter circuit and an AD converter to perform noise reduction and normalization processing on the original pressure signal;

[0041] Data analysis module 105: built-in compression index calculation model, integrating pressure distribution data and image crack characteristics through machine learning algorithm;

[0042] Output module 106: includes an on-board display terminal and a wireless transmission unit, supporting real-time generation of test reports and synchronization of cloud data.

[0043] Another aspect of the present invention provides a method for detecting the compressive strength of a highway pavement, comprising the following steps:

[0044] Step S1: Detect that the vehicle is traveling at a constant speed of 20-40 km / h, and synchronously activate the pressure sensor array 101 and the image acquisition device 102;

[0045] Step S2: establishing a spatial coordinate system through the positioning module 103 to dynamically bind the pressure data with the geographic location information;

[0046] Step S3: the data preprocessing module 104 uses a wavelet transform algorithm to eliminate environmental vibration noise and generate a standardized pressure distribution map;

[0047] Step S4: the image acquisition device 102 acquires a visible light image and an infrared thermal image of the road surface, and identifies crack morphology and abnormal temperature areas through a convolutional neural network;

[0048] Step S5: The data analysis module 105 integrates the pressure concentration, crack density, and temperature gradient parameters, and calculates the regional compressive index RCI based on the random forest algorithm. The calculation formula of the compressive index RCI is:

[0049] RCI=α·(1-P / P_max)+β·(1-C / C_max)+γ·(1-ΔT / ΔT_max)

[0050] Where α, β, and γ are weight coefficients and α+β+γ=1, P is the measured pressure value, C is the crack density, and ΔT is the temperature gradient;

[0051] Step S6: The output module 106 generates a three-dimensional visualization report including a disease location map, an RCI cloud map, and maintenance recommendations.

[0052] The detailed working process of each part of the highway pavement compression resistance detection system of the present invention is as follows:

[0053] Pressure sensor array 101: Piezoelectric sensors encapsulated in flexible PCB substrates are used to form a grid-shaped pressure sensor array, which is integrated into the chassis of the detection vehicle to cover the entire cross-section of the detection area. The spacing between individual sensor units is precisely set to 15cm×15cm to ensure that the changes in road pressure can be captured in detail. A sensor with a range of 0-50MPa is selected to meet the needs of highway road pressure detection, and the sampling frequency is set to 250Hz (greater than or equal to 200Hz) to obtain high-frequency and high-precision pressure data. For example, a Keyence pressure sensor model PZ-G41N is selected, which has the characteristics of high precision and high reliability and can stably collect road pressure information.

[0054] Image acquisition device 102: The image acquisition device includes a high-resolution industrial camera and an infrared thermal imager, which are installed on the top bracket of the inspection vehicle at an angle of 30° and can fully capture the road surface conditions. In order to adapt to different lighting conditions, an active fill light system is equipped, and 650nm band laser-assisted imaging is used in low-light environments. The industrial camera can use the Bas l erac A1920-155um model, with a resolution of 1920×1200 and a frame rate of 155fps, which can clearly capture road surface details; the infrared thermal imager uses FLIRA325sc, with a thermal sensitivity of less than 50mK, which can effectively detect abnormal road surface temperature.

[0055] Positioning module 103: The positioning module uses the fusion of RTK-GNSS positioning unit and inertial navigation system to record the detection position coordinates in real time. The RTK-GNSS positioning unit can use Trimble R10GNSS receiver, which has high-precision positioning capability and can achieve centimeter-level positioning accuracy; the inertial navigation system uses ADI S16448, which uses accelerometers and gyroscopes to measure the vehicle's motion state. The fusion of the two can ensure that the detection position can still be accurately recorded in the case of satellite signal obstruction and establish an accurate spatial coordinate system.

[0056] Data preprocessing module 104: The data preprocessing module is equipped with a digital filter circuit and an AD converter to perform noise reduction and normalization on the original pressure signal. The digital filter circuit is used to filter out high-frequency noise interference, and AD7799 from AD company is selected as the AD converter to achieve high-precision conversion from analog signal to digital signal. At the software level, the wavelet transform algorithm is used to eliminate environmental vibration noise, and the pressure signal is standardized to generate a standardized pressure distribution map for subsequent analysis.

[0057] Data analysis module 105: The data analysis module has a built-in compression index calculation model, which integrates pressure distribution data and image crack features through machine learning algorithms. First, the pressure concentration is analyzed to calculate the distribution difference of pressure per unit area; the crack density is counted using image recognition technology to determine the number and length of cracks; the temperature gradient is analyzed to determine the temperature changes in different areas of the road surface. Based on the random forest algorithm, these parameters are combined to calculate the regional compression index RCI. The algorithm improves the accuracy of the calculation results by constructing multiple decision trees for classification and regression.

[0058] Output module 106: The output module includes an on-board display terminal and a wireless transmission unit. The on-board display terminal uses an industrial-grade touch screen, model Advantech TPC-1261G, which is convenient for inspectors to view the inspection data in real time. The wireless transmission unit uses a 4G module, such as EC20 of Quectel Communications, which supports the real-time generation of inspection reports and synchronization with cloud data, and uploads the inspection data to the cloud server for subsequent storage, management and analysis. At the same time, it generates a three-dimensional visualization report including a disease location map, RCI cloud map and maintenance recommendations, providing intuitive and comprehensive information for highway maintenance.

[0059] The specific implementation steps of a method for detecting the compressive strength of a highway pavement are as follows:

[0060] Step S1: Data collection preparation

[0061] The detection vehicle travels on the road surface to be detected at a uniform speed of 30km / h, which can ensure both detection efficiency and stable data acquisition by the sensor and image acquisition device. During the vehicle driving process, the pressure sensor array 101 and the image acquisition device 102 are synchronously activated to start working and comprehensively collect road pressure and image information.

[0062] Step S2: Location information binding

[0063] An accurate spatial coordinate system is established through the RTK-GNSS positioning unit and the inertial navigation system in the positioning module 103. The RTK-GNSS positioning unit receives satellite signals in real time to determine the geographic location of the vehicle, while the inertial navigation system monitors and corrects the vehicle's motion posture. The pressure data is dynamically bound to the geographic location information so that each pressure data corresponds to the accurate road surface location, which is convenient for subsequent data analysis and disease location.

[0064] Step S3: Pressure data preprocessing

[0065] The digital filter circuit and AD converter in the data preprocessing module 104 start working to preliminarily process the original pressure signal collected by the pressure sensor array 101. The wavelet transform algorithm is used to decompose and reconstruct the signal according to different frequency components to effectively eliminate environmental vibration noise. Then, the processed pressure signal is normalized and converted into a unified standard format to generate a standardized pressure distribution map, providing a high-quality data basis for subsequent data analysis.

[0066] Step S4: Image analysis

[0067] The image acquisition device 102 obtains the visible light image and infrared thermal image of the road surface, and uses the convolutional neural network (CNN) for image recognition. For the visible light image, the trained CNN model is used to identify the road surface crack morphology, including the width, length and direction of the crack; for the infrared thermal image, the CNN model is used to identify the temperature abnormality area and determine whether the road surface has defects such as hollowing and bulging. The recognition results are combined with the pressure data to comprehensively analyze the road surface conditions.

[0068] Step S5: Calculation of compression resistance index

[0069] The data analysis module 105 integrates the pressure concentration, crack density, and temperature gradient parameters, and calculates the regional compressive index RCI based on the random forest algorithm. According to the actual detection requirements and road surface characteristics, the weight coefficients α, β, and γ are determined to satisfy α+β+γ=1. For example, for heavy traffic sections, α (pressure concentration weight) can be appropriately increased; for old roads, β (crack density weight) can be increased. RCI is calculated using the formula RCI=α·(1-P / P_max)+β·(1-C / C_max)+γ·(1-ΔT / ΔT_max) to comprehensively evaluate the road surface's compressive resistance.

[0070] Step S6: Result output

[0071] The output module 106 generates a three-dimensional visualization report including a disease location map, an RCI cloud map, and maintenance recommendations. The disease location map accurately marks the location of the road disease based on the location information recorded by the positioning module; the RCI cloud map intuitively displays the compressive capacity of different areas of the road surface in different colors; the maintenance recommendations provide targeted maintenance measures for the highway maintenance department based on the RCI calculation results and the disease type, such as crack repair, road surface reinforcement, etc., and synchronize the data between the vehicle display terminal and the cloud to achieve rapid and effective transmission of the detection results.

[0072] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A road pavement compression resistance detection system, characterized in that: Includes the following modules: The pressure sensor array (101) is composed of piezoelectric sensors distributed in a grid shape, integrated in the chassis of the detection vehicle, and covers the entire cross-section of the detection area; The image acquisition device (102) comprises a high-resolution industrial camera and an infrared thermal imager, which is installed on the top bracket of the inspection vehicle at an inclination angle of 30°; Positioning module (103): adopts RTK-GNSS positioning unit and inertial navigation system fusion to record the detection position coordinates in real time; Data preprocessing module (104): equipped with a digital filter circuit and an AD converter, which performs noise reduction and normalization processing on the original pressure signal; Data analysis module (105): a built-in compression index calculation model, which integrates pressure distribution data and image crack characteristics through machine learning algorithms; Output module (106): includes an on-board display terminal and a wireless transmission unit, and supports real-time generation of test reports and synchronization of cloud data.

2. A road pavement compression resistance detection system according to claim 1, characterized in that: The pressure sensor array (101) is packaged using a flexible PCB substrate, the spacing between individual sensor units is 15 cm×15 cm, the measuring range is 0-50 MPa, and the sampling frequency is ≥200 Hz.

3. A road pavement compression resistance detection system according to claim 1, characterized in that: The image acquisition device (102) is equipped with an active fill light system, which assists imaging through 650nm laser in a low illumination environment.

4. A method for detecting the compressive strength of a highway pavement, implemented based on the system according to any one of claims 1 to 3, characterized in that: The following steps are involved: Step S1: Detecting that the vehicle is traveling at a constant speed of 20-40 km / h, and synchronously activating the pressure sensor array (101) and the image acquisition device (102); Step S2: establishing a spatial coordinate system through the positioning module (103) to dynamically bind the pressure data with the geographical location information; Step S3: the data preprocessing module (104) uses a wavelet transform algorithm to eliminate environmental vibration noise and generate a standardized pressure distribution map; Step S4: the image acquisition device (102) acquires a visible light image and an infrared thermal image of the road surface, and identifies crack morphology and abnormal temperature areas through a convolutional neural network; Step S5: the data analysis module (105) integrates the pressure concentration, crack density, and temperature gradient parameters, and calculates the regional compressive index RCI based on the random forest algorithm; Step S6: The output module (106) generates a three-dimensional visualization report including a disease location map, an RCI cloud map and maintenance suggestions.

5. A method for detecting the compressive strength of a highway pavement according to claim 4, characterized in that: The calculation formula of the compression resistance index RCI in step S5 is: RCI=α·(1-P / P_max)+β·(1-C / C_max)+γ·(1-ΔT / ΔT_max) Where α, β, and γ are weight coefficients and α+β+γ=1, P is the measured pressure value, C is the crack density, and ΔT is the temperature gradient.

Images