Rapid identification method and monitoring system for fatigue damage stage of asphalt mixture

Abstract

The invention discloses a rapid identification method and a monitoring system for a fatigue damage stage of an asphalt mixture. The identification method comprises the following steps: acquiring an acoustic emission signal generated by a test piece in a fatigue loading process and DIC image data of surface deformation of the asphalt mixture test piece; drawing a curve chart of acoustic emission accumulated impact number and energy changing along with loading times; performing graying processing on the DIC image data, and preliminarily judging spatio-temporal information of the crack; fitting an acoustic emission parameter curve; based on the impact number in the window and the change of acoustic emission energy, the asphalt mixture fatigue damage stage is divided; and aligning the time of the acoustic emission signal with the time of the crack space-time information, and determining the fatigue instability precursor information of the asphalt mixture and the instability failure initial position of the corresponding test piece according to the acoustic emission parameter characteristics before the instability failure of the test piece. According to the method, the acoustic emission signal impact number, energy and DIC image data are combined, and the instability damage of the asphalt mixture is predicted according to the acoustic emission parameter characteristics before fatigue instability of the asphalt mixture.

Description

Technical Field

[0001] The invention belongs to the technical field of road engineering and relates to a rapid identification method and monitoring system for the fatigue damage stage of an asphalt mixture. Background Art

[0002] Among the many defects of asphalt pavements, cracks are the most common, and fatigue cracking is one of the main causes of cracks in asphalt pavements. Fatigue cracks in asphalt pavements are often subtle, subtle, and difficult to detect. Currently, the main testing methods for identifying fatigue cracks in asphalt pavements include ground-penetrating radar (GPR), ultrasonic testing, and industrial CT. GPR uses an antenna to transmit and receive high-frequency electromagnetic waves to detect the internal material properties and distribution patterns of the asphalt pavement. This method is susceptible to environmental factors and interference signals, requires specialized technicians to operate and identify, and can be challenging to obtain crack information. Ultrasonic testing utilizes the time difference between different acoustic signals as ultrasonic waves penetrate and reflect within an object to infer internal material defects. Due to the heterogeneous composition and viscoelastic nature of asphalt mixtures, ultrasonic testing data exhibits anisotropy and hysteresis. Industrial CT methods primarily reconstruct the spatial structure of asphalt mixtures based on the distribution of radiation absorption coefficients of different materials. However, existing testing equipment is expensive and inconvenient to carry, and the testing accuracy is insufficient to meet the requirements for microcrack detection. In summary, existing methods are mostly active detection methods, resulting in results that are susceptible to interference from material and human factors, and their accuracy and efficiency are insufficient to meet the requirements for real-time monitoring of asphalt pavements. To gain a deeper understanding of the entire process from the initial formation of asphalt mixture cracks to their expansion and evolution into macrocracks, it is necessary to develop real-time, dynamic monitoring equipment that can capture subtle fatigue damage initiation information in asphalt pavements, and to establish diagnostic evaluation methods that can rapidly identify the evolutionary state of cracks in asphalt pavements.

[0003] Acoustic emission (AE) refers to transient elastic waves generated by the rapid release of internal energy during material damage. By extracting and analyzing the parameter and waveform characteristics of AE signals, changes in the material's mechanical properties and the evolution of its damage state can be determined. Among AE parameters, the impact count is particularly sensitive to information about pore closure, aggregate migration, and crack propagation within asphalt mixtures. AE energy, on the other hand, reflects the energy accumulation and release patterns during fatigue loading in asphalt mixtures, accurately reflecting the evolution of internal damage.

[0004] As a non-contact measurement technology, DIC can obtain local real-time displacement and strain of materials with high precision, timely perceive the changes in surface displacement and strain of materials during cyclic loading, capture fatigue crack initiation information, and characterize the displacement and strain characteristics of the entire process of fatigue crack initiation and expansion.

[0005] Existing technology 1 (patent application number CN201810224227.2) uses a drop weight deflectometer to measure the deflection changes at different measuring points on both sides of the crack to determine the development of cracks inside the asphalt pavement. It can only detect the development of existing pavement cracks and cannot conduct long-term monitoring of the formation and expansion of cracks. Moreover, the evaluation basis of this method is closely related to the selection of measuring points. The results obtained are easily affected, highly subjective, and not intuitive and objective enough.

[0006] Prior art 2 (patent application number CN202221240468.4) uses high-definition, high-speed cameras to capture asphalt pavement surface image data and visually assess the development of cracks on the pavement surface. However, this method only captures the development of cracks on the pavement surface, but cannot assess the internal cracks in the pavement, nor can it provide a qualitative assessment of the extent of crack development.

[0007] Existing technology 3 (patent application number CN201910394016.8) emits electromagnetic waves or microwaves through electromagnetic wave or microwave devices, causing the temperature field near the road area to gradually increase, and uses an infrared camera to perform temperature imaging, and uses the road area temperature infrared imaging data to clarify the development of cracks inside the asphalt pavement; however, by emitting electromagnetic waves or microwaves to heat the road to form a temperature field with a certain temperature, this process undoubtedly exacerbates the aging process of the road surface, and this method is only for post-detection, and cannot achieve real-time monitoring of cracks in the asphalt pavement. Summary of the Invention

[0008] In order to solve the above problems, the present invention provides a method for quickly identifying the fatigue damage stage of asphalt mixture. The method combines the impact number and energy of the acoustic emission signal generated during the fatigue loading process of the asphalt mixture with the DIC image data of the surface deformation to jointly judge the fatigue damage state of the asphalt mixture. The instability and failure of the asphalt mixture are predicted based on the acoustic emission parameter characteristics before the asphalt mixture fatigue instability, thereby solving the problems existing in the prior art.

[0009] Another object of the present invention is to provide a monitoring system for the fatigue damage stage of asphalt mixture.

[0010] The technical solution adopted by the present invention is a method for quickly identifying the fatigue damage stage of asphalt mixture, comprising the following steps:

[0011] S1, fatigue loading is performed on the asphalt mixture specimen, and the acoustic emission signals generated during the fatigue loading process are collected by multiple acoustic emission sensors on the specimen surface. At the same time, DIC image data of the deformation of the asphalt mixture specimen surface during the fatigue loading process is collected by a high-speed industrial CCD camera, and the image data is stored in a DIC acquisition computer;

[0012] S2, based on the real-time acoustic emission data, draw a curve of the cumulative number of acoustic emission impacts and energy versus the number of loading times;

[0013] Select multiple DIC image data during the test, perform grayscale processing, and filter out areas with larger gradients by setting a threshold to preliminarily determine the spatiotemporal information of the cracks.

[0014] S3, based on the curve of the cumulative number of acoustic emission impacts and energy changing with the number of loading times, fit the acoustic emission parameter curve; based on the changes in the number of impacts and acoustic emission energy within the window, divide the asphalt mixture fatigue damage stage;

[0015] S4, aligning the time of the acoustic emission signal obtained in S3 and the crack spatiotemporal information obtained in S2 to obtain combined characteristic information, determining the precursor information of fatigue instability of the asphalt mixture based on the acoustic emission parameter characteristics before the specimen becomes unstable and fails, and determining the starting position of the instability and failure of the corresponding specimen.

[0016] Furthermore, a plurality of the acoustic emission sensors are arranged on the surface of the asphalt mixture specimen according to the arrangement principle of alternating on the same plane, staggered relative to the plane, and globally covering, for receiving the acoustic emission signals on the surface of the asphalt mixture specimen.

[0017] Furthermore, an acoustic emission main box is installed on one side of the asphalt mixture specimen, which is used to send acoustic emission signals to the asphalt mixture specimen; and collect and convert the acoustic emission signals received by the acoustic emission sensor, and finally store them in the acoustic emission acquisition computer.

[0018] Furthermore, in S1, the image data of the calibration plate is collected by a high-speed industrial CCD camera, and after being processed and analyzed by digital image processing software, the calibration data is confirmed, a data processing coordinate system is established, and the position of the test piece is calibrated.

[0019] Furthermore, in S2, I(x, y) in the grayscale image of the DIC data represents the grayscale value of the image, and the magnitude G and direction θ of the gradient are calculated by the following formula:

[0020]

[0021]

[0022]

[0023]

[0024] Among them, G x , G y Represents the components of the gradient amplitude G in the x and y directions, I x and I yis the gradient component of the grayscale value I(x, y) of the image in the x and y directions, and * represents the convolution operation.

[0025] Further, in S4, the detection method for the mutation points of the number of impacts and acoustic emission energy within the window:

[0026] Select the number of impacts per second and the acoustic emission energy as the indicators of the change in the damage state. Use a sliding window in Python to calculate the number of impacts and the acoustic emission energy per second. If the slope change within a certain window exceeds the threshold compared to the previous window, the point corresponding to this window is a mutation point.

[0027] Further, the cumulative number of impacts and acoustic emission energy are cumulative values that change with the number of loadings / time. Let Δt represent the time interval between two loadings, N represent the number of loadings, I N represents the number of impacts at the Nth loading, I N-1 represents the number of impacts at the (N - 1)th loading; E N represents the acoustic emission energy at the Nth loading, E N-1 represents the acoustic emission energy at the (N - 1)th loading; the number of impacts I within the time interval Δt s and the acoustic emission energy E s are expressed as:

[0028]

[0029]

[0030] For the mutation points of the cumulative curve, they are determined by the change rates of the number of impacts and acoustic emission energy within the time interval Δt; the number of loadings is N, and the mutation point is M, 1 < M < N. Then, the change rates of the number of impacts and acoustic emission energy within the time interval Δt before and after the mutation point are expressed as:

[0031]

[0032]

[0033] In the formula: Δt is the time interval between two loadings, H is the change rate of the number of impacts within the time interval Δt before the mutation point, H M represents the number of impacts at the Mth loading, H M-1 represents the number of impacts at the (M - 1)th loading; H' is the change rate of the number of impacts within the time interval Δt after the mutation point, H N represents the number of impacts at the Nth loading; E is the change rate of the acoustic emission energy within the time interval Δt before the mutation point, E M represents the acoustic emission energy at the Mth loading, E M-1 represents the acoustic emission energy at the (M - 1)th loading; E' is the change rate of the acoustic emission energy within the time interval Δt after the mutation point, E NRepresents the acoustic emission energy at the Nth loading.

[0034] Furthermore, the high-speed industrial CCD camera is fixed at a certain distance from the test piece at an angle of 15 to 60 degrees, and the high-speed industrial CCD camera is fixed by a tripod.

[0035] Furthermore, a supplementary light source is installed between the high-speed industrial CCD camera and the asphalt mixture specimen; and the acoustic emission main box and the acoustic emission sensor are connected via an amplifier.

[0036] A monitoring system for the fatigue damage stage of an asphalt mixture adopts the above-mentioned rapid identification method for the fatigue damage stage of an asphalt mixture, comprising:

[0037] A loading system, used to fix the asphalt mixture specimen and apply fatigue loading to the specimen;

[0038] The DIC image acquisition module is used to grayscale multiple DIC image data during the test, screen areas with larger gradients by setting thresholds, and preliminarily determine the spatiotemporal information of the cracks;

[0039] The acoustic emission acquisition module is used to input the collected acoustic emission data into the software analysis terminal. Based on the real-time acoustic emission data, a graph of the cumulative number of acoustic emission impacts and energy changes with the number of loading times is plotted. Based on the graph of the cumulative number of acoustic emission impacts and energy changes with the number of loading times, an acoustic emission parameter curve is fitted. Based on the changes in the number of impacts and acoustic emission energy within the window, the fatigue damage stage of the asphalt mixture is divided.

[0040] The data fusion module is used to align the time of the obtained acoustic emission signal and the crack spatiotemporal information, determine the precursor information of fatigue instability of the asphalt mixture based on the acoustic emission parameter characteristics before the specimen becomes unstable and fail, and determine the starting position of the instability and failure of the corresponding specimen.

[0041] The beneficial effects of the present invention are:

[0042] 1. The present invention grayscales the DIC image to obtain gradient information. Cracks are usually located at positions with larger gradients, which can be used to preliminarily judge the damage state of the asphalt mixture.

[0043] 2. The parameter changes in the acoustic emission signals monitored by this invention are directly related to the degree of material damage. By continuously monitoring the acoustic emission signals, when the acoustic emission parameters (such as the number of impacts and energy) suddenly change or change in trend, the damage state of the asphalt mixture can be determined, including the stages of pore closure, microcrack initiation, stable macrocrack expansion, and instability failure.

[0044] 3. This invention combines acoustic emission signals and DIC image data for a more accurate assessment of fatigue damage states. The integration of these two distinct data sources overcomes the limitations of single-source analysis and improves the reliability of the assessment. Acoustic emission signals and DIC image data provide a comprehensive understanding of the specimen's damage behavior, enabling identification of stages such as pore closure, microcrack initiation, and crack propagation, and prediction of instability and failure.

[0045] 3. The fatigue state monitoring method of the present invention can predict signs of instability, allowing road maintenance personnel to respond quickly and take preventive maintenance measures before damage occurs, thereby preventing sudden damage from affecting the safe driving of vehicles. This helps reduce maintenance costs, improve road maintenance efficiency, and enhance road quality. The early warning function helps identify potential problems in advance, reduces the risk of traffic accidents, and improves traffic safety. This preventive maintenance can avoid large-scale repair work, thereby reducing traffic disruptions during maintenance. At the same time, by monitoring the fatigue state of the asphalt mixture in real time, road maintenance personnel can carry out targeted maintenance and repairs based on the damage situation, thereby extending the service life of the road.

[0046] 4. The rapid identification method proposed in this invention enables continuous, real-time, and dynamic monitoring, avoiding the drawback of traditional nondestructive testing techniques, which often lack the ability to provide real-time assessment of asphalt mixture damage after the fact. This method is of great significance for early warning of asphalt pavement instability and preventive maintenance. Furthermore, the data generated by this invention provides a new avenue for the research and improvement of road materials, and can provide road managers with detailed information to guide maintenance decisions and planning. This contributes to intelligent pavement management, improving efficiency and quality, enabling more rational maintenance plans, and optimizing resource allocation, thus achieving intelligent pavement management. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 Schematic diagram of the asphalt mixture fatigue damage and cracking monitoring platform according to an embodiment of the present invention.

[0049] Figure 2This is a schematic diagram of the combined characteristics of acoustic emission and DIC of AC-20 asphalt mixture under four-point bending fatigue loading obtained on the monitoring platform under the conditions of temperature 20°C, fatigue loading stress ratio 0.4, and fatigue loading frequency 5 Hz in Test Example 1 of the present invention.

[0050] Figure 3 This is a schematic diagram of the combined characteristics of acoustic emission and DIC of AC-20 asphalt mixture under four-point bending fatigue loading obtained on the monitoring platform under the conditions of temperature 20°C, fatigue loading stress ratio 0.4, and fatigue loading frequency 10 Hz in Test Example 2 of the present invention.

[0051] Figure 4 This is a schematic diagram of the combined characteristics of acoustic emission and DIC of AC-20 asphalt mixture under four-point bending fatigue loading obtained on the monitoring platform under the conditions of temperature 20°C, fatigue loading stress ratio 0.4, and fatigue loading frequency 15 Hz in Test Example 3 of the present invention.

[0052] Figure 5 This is the image data after the DIC image is grayscaled in Example 1 of the present invention.

[0053] Figure 6 Flowchart of an embodiment of the present invention.

[0054] Among them, 1. DIC acquisition computer, 2. High-speed industrial CCD camera, 3. Supplementary light source, 4. Loading system, 5. Acoustic emission sensor, 6. Acoustic emission main box, 7. Acoustic emission acquisition computer. DETAILED DESCRIPTION

[0055] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0056] The basic concept of the embodiment of the present invention is as follows:

[0057] Given the sensitivity and intuitiveness of acoustic emission and DIC technologies in capturing material damage states, their application to monitoring fatigue damage and cracking in asphalt mixtures can help understand the damage state of asphalt mixtures, clarify asphalt mixture damage information, and provide guidance for asphalt pavement maintenance. Based on acoustic emission and DIC technologies, a monitoring system and rapid identification method for the fatigue damage stage of asphalt mixtures have been proposed, enabling early warning of fatigue instability failure in asphalt mixtures.

[0058] Example 1,

[0059] Asphalt mixture fatigue damage cracking monitoring platform, such as Figure 1As shown, it includes a loading system 4 for fixing the asphalt mixture specimen and applying fatigue loading to the specimen;

[0060] An acoustic emission main box 6 is installed on one side of the asphalt mixture specimen, which is used to send acoustic emission signals to the asphalt mixture specimen;

[0061] Multiple acoustic emission sensors 5 are arranged on the surface of the asphalt mixture specimen according to the layout principles of alternating on the same plane, staggered relative planes, and global coverage to receive acoustic emission signals. This ensures that the acoustic emission information is fully collected, saves the number of sensors arranged, and reduces cost investment. The number of acoustic emission sensors 5 is determined according to the size of the specimen, and the acoustic emission sensors 5 are VS45-H type piezoelectric sensors.

[0062] The acoustic emission main box 6 is connected to the acoustic emission collection computer 7, and the acoustic emission collection computer 7 is used to input the collected acoustic emission data into the software analysis terminal.

[0063] A high-speed industrial CCD camera 2 is installed on the other side of the asphalt mixture specimen. The high-speed industrial CCD camera 2 is fixed at a certain distance in front of the specimen at an angle of 15 to 60 degrees. It is used to collect image data during the fatigue loading process of the asphalt mixture and ensure that the specimen occupies about 3 / 4 of the entire display screen. The high-speed industrial CCD camera 2 is fixed by a tripod; the supplementary light source 3 is placed between the high-speed industrial CCD camera 2 and the asphalt mixture specimen, and is located on both sides of the lens.

[0064] The high-speed industrial CCD camera 2 is connected to the DIC acquisition computer 1, and the DIC acquisition computer 1 is used for the operation of the image acquisition system and the image calculation and processing software.

[0065] The acoustic emission main box 6, the acoustic emission sensor 5, the acoustic emission acquisition computer 7, the amplifier, the acoustic emission acquisition system (with functions such as positioning and event screening) and the wires constitute the acoustic emission equipment; the collected acoustic emission raw data files can be read through the Python Vallenae package.

[0066] The DIC system consists of a high-speed industrial CCD camera 2, a supplementary light source 3, a DIC acquisition computer 1, a tripod and a pan / tilt head, a speckle tool, a calibration kit, a data collector, an image acquisition system, and image calculation software.

[0067] After setting the corresponding acquisition parameters on DIC acquisition computer 1 and acoustic emission acquisition computer 7, a fatigue loading test of asphalt mixture under certain conditions is carried out through loading system 4. Both DIC acquisition computer 1 and acoustic emission acquisition computer 7 are simultaneously turned on to collect the corresponding data. The acoustic emission signals generated during the fatigue loading process are collected by acoustic emission sensor 5, collected and converted by acoustic emission mainframe 6, and finally stored in acoustic emission acquisition computer 7.

[0068] The acoustic emission mainframe 6 and the acoustic emission sensor 5 are connected via an amplifier, and the two ends of the amplifier are connected to the acoustic emission mainframe 6 and the acoustic emission sensor 5 respectively via wires. The functions of the amplifier are: (1) providing impedance matching between the high-impedance sensor and the low-impedance transmission cable to prevent signal attenuation; (2) improving the signal-to-noise ratio related to cable noise by amplifying weak input signals; (3) reducing the common-mode electrical noise introduced by the sensor and its cable through differential amplification; and (4) providing a frequency filter.

[0069] A rapid identification method for fatigue damage stage of asphalt mixture, such as Figure 6 As shown, the following steps are included:

[0070] S1, using a high-speed industrial CCD camera 2 to collect image data of the calibration plate, and after processing and analysis by digital image processing software, confirming the calibration data, establishing a data processing coordinate system, and calibrating the specimen position; fatigue loading the treated asphalt mixture specimen according to certain fatigue loading conditions, and collecting acoustic emission signals generated during the fatigue loading process through multiple acoustic emission sensors 5 on the specimen surface. At the same time, DIC image data of the deformation of the asphalt mixture specimen surface during the fatigue loading process is collected by the high-speed industrial CCD camera 2, and the collected image is ensured to be clear and visible by using a supplementary light source 3, and finally the image data is stored in the DIC acquisition computer 1;

[0071] S2, the acoustic emission collection computer 7 draws a graph of the cumulative number of acoustic emission impacts and energy changes with the number of loading times based on the acoustic emission data collected in real time;

[0072] Select 8 DIC image data in the experimental process and use Python to grayscale the image, as shown in the following example: Figure 5 As shown, grayscale value = 0.299 × red channel + 0.587 × green channel + 0.114 × blue channel. In grayscale images, cracks typically appear as high-contrast linear or nonlinear structures. Using the image's gradient information, cracks are typically located at locations with large gradients. Assuming I(x, y) represents the image's grayscale value, the gradient magnitude G and direction θ can be calculated using the following formula:

[0073]

[0074]

[0075]

[0076]

[0077] Among them, G x , G y Represents the components of the gradient amplitude G in the x and y directions, Ix and I y It is the gradient component of the grayscale value I(x,y) of the image in the x and y directions, and * represents the convolution operation.

[0078] By setting a threshold, we can filter out areas with larger gradients and preliminarily judge the spatiotemporal information of the cracks. In practical applications, the threshold is adjusted based on large amounts of data analysis and actual conditions.

[0079] In step S3, the acoustic emission acquisition computer 7 uses a curve fitting method to fit the acoustic emission parameter curve based on the accumulated and summed acoustic emission parameter characteristic curve. The number of impacts per second and the acoustic emission energy are selected as indicators of damage state changes. A sliding window is used in Python to calculate the number of impacts and the acoustic emission energy per second. The change in slope within the window is then used to detect the mutation point.

[0080] For the data (x i ,y i ), where x i Indicates the number of loads, y i Indicates the corresponding acoustic emission parameters (impact number I s and energy E s )The linear regression model is:

[0081] y i =kx i +b

[0082] Where k is the slope and b is the intercept. The slope k and intercept b of the best fitting line are solved using the least squares method, and the minimum residual sum of squares is calculated to obtain the slope k of each window.

[0083]

[0084] Then compare the changes in the slopes of adjacent windows. Assume that the current window slope is k i , the slope of the previous window is k i-1 , then the slope changes to:

[0085] Δk i =|k i -k i-1 |

[0086] When Δk i When the threshold is exceeded, a mutation point is detected. By setting a threshold to detect the mutation point, the cumulative number of impacts and the cumulative acoustic emission energy curve are calculated separately. The Δk of the two in the same window i When both are greater than the threshold, the point corresponding to the window is a mutation point.

[0087] The cumulative impact number and acoustic emission energy are cumulative values that change with the number of loadings (time). By judging the position of the mutation point window through the change of the slope, and then calculating the impact number H within the Δt time inside the window s and the acoustic emission energy E s , to determine the time point information of the specific number of loadings. Let Δt represent the time interval between two loadings, N represent the number of loadings, H N represent the impact number at the Nth loading, H N-1 represent the impact number at the (N - 1)th loading; E N represent the acoustic emission energy at the Nth loading, E N-1 represent the acoustic emission energy at the (N - 1)th loading; the impact number H within the Δt time s and the acoustic emission energy E s can be expressed as:

[0088]

[0089]

[0090] In the formula: Δt is the time interval between two loadings, H s is the change rate of the impact number within the Δt time, H N represents the impact number at the Nth loading, H N-1 represents the impact number at the (N - 1)th loading. E s is the change rate of the acoustic emission energy within the Δt time, E N represents the acoustic emission energy at the Nth loading, E N-1 represents the acoustic emission energy at the (N - 1)th loading.

[0091] For the mutation point of the cumulative curve, it is determined by analyzing the change rates of the impact number and acoustic emission energy within the Δt time. The number of loadings is N, and the mutation point is M (1 < M < N). Then the change rates of the impact number and acoustic emission energy within the Δt time before and after the mutation point can be expressed as:

[0092]

[0093]

[0094] In the formula: Δt is the time interval between two loadings, H is the change rate of the impact number within the Δt time before the mutation point, H M represents the impact number at the Mth loading, H M-1 represents the impact number at the (M - 1)th loading; H' is the change rate of the impact number within the Δt time after the mutation point, H N represents the impact number at the Nth loading. E is the change rate of the acoustic emission energy within the Δt time before the mutation point, E Mrepresents the acoustic emission energy at the Mth loading, E M-1 represents the acoustic emission energy during the M-1th loading; E' is the rate of change of acoustic emission energy within Δt after the mutation point, E N Represents the acoustic emission energy at the Nth loading.

[0095] Based on the changes in the number of impacts and acoustic emission energy within the window, the fatigue damage stages of asphalt mixture are divided.

[0096] S4, the acoustic emission acquisition computer 7 obtains the time information of the acoustic emission signal and the DIC signal through Python data processing, aligns the initial signal time, and ensures the consistency of the acoustic emission signal and the DIC image data on the time scale, so as to determine the surface deformation and crack image data of the asphalt mixture at the corresponding fatigue damage stage (time point) and clarify the development status of the asphalt mixture damage. The precursor information of fatigue instability of the asphalt mixture is determined based on the grayscale gradient of the DIC image and the acoustic emission parameter characteristics before the specimen is unstable and damaged, and the starting position of the specimen instability and damage is determined. The sliding window is used to detect the slope change of the acoustic emission parameter per second, which can accurately map the change of the acoustic emission parameter to different fatigue damage stages.

[0097] From the beginning of fatigue loading, the number of acoustic emission signal impacts and the cumulative energy value generated during the entire fatigue loading stage are counted, and the curve of their change with the number of loading times is drawn (i.e. the curve of the cumulative number of acoustic emission impacts and energy change with the number of loading times, such as Figure 2-4 The transition points of the fatigue damage stages of the asphalt mixture are determined based on the slope mutation points of the cumulative acoustic emission parameter curve. Due to the varying degrees of material damage within the specimen, this is reflected in the changes in the acoustic emission parameters. When the acoustic emission parameters change slowly and remain at a relatively low level, the asphalt mixture is in the elastic deformation stage with pore closure and microcrack initiation. When the acoustic emission parameters accelerate and develop steadily at a relatively high level, the asphalt mixture is in the plastic deformation stage with rapid crack expansion. When the acoustic emission parameters undergo large-scale mutations, it indicates that drastic changes have occurred within the asphalt mixture specimen, and the specimen is heading towards an unstable failure process. Based on the cumulative number of acoustic emission impacts and the energy slope mutation, the fatigue loading failure process of the asphalt mixture specimen can be divided into the pore closure stage (stage I), the microcrack initiation stage (stage II), the stable macrocrack expansion stage (stage III), and the unstable failure stage (stage IV).

[0098] Based on extensive acoustic emission test data analysis, both the cumulative acoustic emission impact count and cumulative acoustic emission energy curves are closely correlated with the development of asphalt mixture cracks, exhibiting a "synchronous" phased pattern. In the crack initiation precursor stage, the acoustic emission signal is relatively weak, with a slight increase in acoustic emission energy and impact count, but the change is not significant. The macrocrack propagation precursor stage is accompanied by a greater increase in impact count and energy, with abrupt changes occurring at shorter intervals. The instability precursor stage manifests as a sharp increase in energy and impact count, with the impact points exhibiting a sharp change in slope. Depending on the specific situation, extensive experimentation and data analysis are required to identify the characteristics of the instability precursors and establish reliable feature extraction algorithms. Extracting these features can help predict and identify instability precursors, providing road maintenance personnel with timely warnings and preventive maintenance measures. This process requires a deep understanding of acoustic emission data and damage behavior to obtain accurate results.

[0099] Example 2,

[0100] A monitoring system for the fatigue damage stage of an asphalt mixture adopts the above-mentioned rapid identification method for the fatigue damage stage of an asphalt mixture, comprising:

[0101] The loading system 4 is used to fix the asphalt mixture specimen and apply fatigue loading to the specimen;

[0102] The DIC image acquisition module is used to grayscale multiple DIC image data during the test, screen areas with larger gradients by setting thresholds, and preliminarily determine the spatiotemporal information of the cracks;

[0103] The acoustic emission acquisition module is used to input the collected acoustic emission data into the software analysis terminal. Based on the real-time acoustic emission data, a graph of the cumulative number of acoustic emission impacts and energy changes with the number of loading times is plotted. Based on the graph of the cumulative number of acoustic emission impacts and energy changes with the number of loading times, an acoustic emission parameter curve is fitted. Based on the changes in the number of impacts and acoustic emission energy within the window, the fatigue damage stage of the asphalt mixture is divided.

[0104] The data fusion module is used to align the time of the obtained acoustic emission signal and the crack spatiotemporal information, determine the precursor information of fatigue instability of the asphalt mixture based on the acoustic emission parameter characteristics before the specimen becomes unstable and fail, and determine the starting position of the instability and failure of the corresponding specimen.

[0105] The advantages of the embodiments of the present invention are:

[0106] DIC image data can provide information on crack propagation on the specimen surface. By combining it with acoustic emission parameter data, DIC image data can be used to verify the fatigue damage stage indicated by changes in acoustic emission parameters. For example, when acoustic emission parameters indicate crack initiation, DIC image data can show the propagation of microcracks. This comprehensive analysis can improve the reliability of the determination of fatigue damage states. If we rely solely on acoustic emission data, there may be uncertainty in identifying different fatigue damage stages. This is because acoustic emission parameters are more sensitive to the acoustic wave signals during crack propagation, but they do not always provide enough information to clearly distinguish different damage stages. DIC image data provides visual information that can verify the correctness of the acoustic emission parameter characteristics, thereby improving the accuracy of fatigue damage state identification.

[0107] DIC is used to obtain surface image data of the specimen during loading, determine the surface deformation and crack distribution development of the asphalt mixture in the corresponding fatigue damage stage, and clarify the damage development status of the asphalt mixture: under continuous cyclic loading, the weak points of the specimen are in a process of continuous stress concentration and dispersion, gradually forming crack initiation locations. Ultimately, cracks nucleate, initiate, and expand here, which can be clearly reflected in the image data obtained by DIC.

[0108] By combining the above steps, the present invention overcomes the difficulties of multi-source data fusion, parameter threshold determination, and instability precursor feature extraction. Through data synchronization, mathematical analysis, and signal processing, it effectively integrates acoustic emission data and DIC image data, accurately determines the fatigue damage stage, and extracts instability precursor information, providing real-time and accurate support for road maintenance decision-making.

[0109] The embodiment of the present invention can realize the precursor warning of fatigue instability damage of asphalt mixture. The fatigue damage state of the asphalt mixture is judged by combining the number and energy of acoustic emission signal impacts generated during the fatigue loading process of the asphalt mixture and the asphalt mixture surface image data obtained based on DIC technology, and the instability damage of the asphalt mixture is predicted by the acoustic emission parameter characteristics before the fatigue instability of the asphalt mixture. The trend of the change of the internal damage signal of the asphalt mixture received and sensed by the acoustic emission sensor can quickly and effectively identify the damage stage of the current specimen. At the same time, the specimen surface image data obtained by DIC can further support and judge the damage state of the asphalt mixture, and the obtained image data can clearly show the crack distribution and expansion trend on the surface of the asphalt mixture specimen, and the through cracks that seriously damage the pavement structure can also be perceived in time.

[0110] Test Example 1,

[0111] The test object is AC-20 asphalt mixture, and the fatigue loading method is four-point bending fatigue loading. The fatigue loading conditions are temperature 20°C, stress ratio 0.4, and loading frequency 5Hz. The schematic diagram of the combined characteristics of acoustic emission and DIC of AC-20 asphalt mixture under four-point bending fatigue loading obtained by the monitoring system of Example 1 is as follows: Figure 2 shown.

[0112] Figure 2 In the figure, the x-axis represents the number of loadings, ranging from 0 to 1800, and the y-axis represents the cumulative number of acoustic emission impacts and the cumulative acoustic emission energy, ranging from 0 to 600 and 0 to 200,000 eu, respectively. The acoustic emission parameter levels also experienced a small increase in the early stages, remaining at a relatively low level. The values ​​remained stable at a high level in the middle stages, and then underwent a dramatic sudden change in the late stages, with the values ​​rising vertically. The specimen surface image data collected by DIC acquisition computer 1 can intuitively reflect the initiation and development of cracks in the specimens at different stages of fatigue damage cracking.

[0113] Figure 2 In the figure, I represents the pore closure stage, II the microcrack initiation stage, III the stable macrocrack expansion stage, and IV the unstable failure stage. The image data a, b, c, d, e, f, g, and h represent the image information of the asphalt mixture crack growth process collected by the DIC system, and feature images of each stage are selected. The cumulative acoustic emission impact count curve a, b, c, d, e, f, g, and h points represent the acoustic emission status information at the time the DIC system captured the image.

[0114] Test Example 2,

[0115] The test object is AC-20 asphalt mixture, and the fatigue loading method is four-point bending fatigue loading. The fatigue loading conditions are temperature 20°C, stress ratio 0.4, and loading frequency 10Hz. The schematic diagram of the combined characteristics of acoustic emission and DIC of AC-20 asphalt mixture under four-point bending fatigue loading obtained by the monitoring system of Example 1 is as follows: Figure 3 shown.

[0116] Figure 3 In the figure, the x-axis represents the number of loadings, ranging from 0 to 2300, and the y-axis represents the cumulative number of acoustic emission impacts and the cumulative acoustic emission energy, ranging from 0 to 1000 and 0 to 320,000 eu, respectively. The acoustic emission parameter levels also experienced a small increase in the early stages, remaining at a relatively low level. The values ​​remained stable at a high level in the middle stages, and then underwent a dramatic sudden change in the late stages, with the values ​​rising vertically. The specimen surface image data collected by DIC acquisition computer 1 can intuitively reflect the initiation and development of cracks in the specimen at different stages of fatigue damage cracking.

[0117] Test Example 3,

[0118] The test object is AC-20 asphalt mixture, and the fatigue loading method is four-point bending fatigue loading. The fatigue loading conditions are temperature 20°C, stress ratio 0.4, and loading frequency 15Hz. The schematic diagram of the combined characteristics of acoustic emission and DIC of AC-20 asphalt mixture under four-point bending fatigue loading obtained by the monitoring system of Example 1 is as follows: Figure 4 shown.

[0119] Figure 4 In the figure, the x-axis represents the number of loadings, ranging from 0 to 14,550, and the y-axis represents the cumulative number of acoustic emission impacts and the cumulative acoustic emission energy, ranging from 0 to 700 and 0 to 140,000 eu, respectively. The acoustic emission parameter levels also experienced a small increase in the early stages, remaining at a relatively low level. The values ​​remained stable at a high level in the middle stages, and then underwent a dramatic sudden change in the late stages, with the values ​​rising vertically. The specimen surface images collected by DIC can intuitively reflect the initiation and development of cracks in the specimens at different stages of fatigue damage cracking.

[0120] from Figures 2 to 4 As can be seen, the cumulative number of acoustic emission impacts and energy show a small initial surge, reflecting the rapid closure of pores under repeated loading. Subsequently, the growth rate of the cumulative number of acoustic emission impacts and energy gradually slows, reflecting the gradual initiation of microcracks within the asphalt mixture specimen. During this period, the specimen surface images collected by DIC reveal the gradual initiation of microcracks at the stress concentration point. Furthermore, the slopes of the cumulative number of acoustic emission impacts and energy curves undergo a small abrupt change, and both values ​​remain at a high dimensionality. At this point, the DIC surface images clearly show macrocrack propagation, indicating that the specimen is in a stage of stable macrocrack propagation. During this process, the acoustic emission parameter curves exhibit multiple "step-like" growth patterns, reflecting the development of multiple cracks within the specimen. In the later stages of loading, the cumulative number of acoustic emission impacts and energy show a large abrupt change, exhibiting a vertical upward trend. Simultaneously, through-cracks appear within the specimen, indicating that the asphalt mixture specimen is in a stage of instability and failure. The characteristic points of the transition from fatigue crack propagation to fatigue crack propagation in asphalt mixtures can be determined by the mutation points of the slope changes in the accumulated acoustic emission parameters. The propagation process and direction of fatigue cracks in asphalt mixtures can be further clarified through the specimen surface image data obtained by DIC. In addition, the vertical explosive growth of acoustic emission parameters can be used as a precursor to fatigue instability cracking and failure in asphalt mixtures.

[0121] The embodiment of the present invention can monitor the development of cracks inside and on the surface of the entire asphalt mixture by arranging acoustic emission sensors (according to the arrangement principles of alternating on the same plane, staggered relative planes, and global coverage), and clarify the expansion stage of the cracks; and further intuitively display the incubation and expansion of the cracks from an optical perspective, thereby realizing full-process monitoring of the incubation, interactive fusion, and expansion of cracks, with little influence from human factors, and accurately grasping the evolution stage of asphalt mixture cracks, and intuitively displaying the development direction and strain distribution of cracks on the surface of the asphalt mixture; and can quickly judge the development status of cracks in the asphalt pavement without any negative impact on the service performance of the pavement.

[0122] Fatigue damage development is characterized by a long cycle, difficulty in effectively capturing crack initiation information, and difficulty in classifying damage states. The embodiment of the present invention uses the acoustic emission parameter slope mutation point as the asphalt mixture damage state change point, and utilizes digital image correlation technology to collect image data to monitor the development of cracks and deformation in the asphalt mixture in real time. By fully leveraging the advantages of acoustic emission technology's sensitivity to material damage development and digital image correlation technology's intuitive presentation of material damage states, and combining the two, the optical and acoustic perspectives can corroborate each other and jointly determine the development of fatigue cracks in the asphalt mixture, greatly improving the accuracy of the results.

[0123] In summary, the embodiments of the present invention offer advantages in the field of nondestructive testing and monitoring of asphalt, including strong sustainability, high sensitivity, and minimal human influence in passive testing. They can fully monitor the initiation, expansion, and instability of latent fatigue cracks in asphalt pavements, assess their stage by stage, and provide early warning of potential instability and damage. Furthermore, collected image data can visually display the development of latent cracks in asphalt pavements. The combined acoustic and optical monitoring of the development of cracks in asphalt pavements can provide constructive guidance for asphalt pavement maintenance and repair.

[0124] The embodiment of the present invention refers to the "Test Procedure for Asphalt and Asphalt Mixtures in Highway Engineering" (JTG E20-2011) and uses an MTS universal loading tester to perform four-point bending fatigue loading, simulating the fatigue crack propagation of roads under repeated vehicle loads for effective characterization, which is closer to actual conditions.

[0125] The monitoring capabilities of acoustic emission and DIC technologies are limited by their respective operating principles and application scopes. While acoustic emission primarily focuses on acoustic signals from internal activity, it can detect minute internal damage but cannot provide information on surface deformation or crack distribution. DIC, while using image processing techniques, can provide high-resolution surface deformation information but cannot capture the internal state of the specimen. Therefore, using either acoustic emission or DIC alone in monitoring fatigue damage and cracking in asphalt mixtures presents challenges such as insufficient information and limited measurement accuracy.

[0126] In practical applications, it is very difficult to accurately match the acoustic emission signal and DIC image data to the same position on the specimen, especially when the specimen is unstable or deformed. Secondly, the analysis of the two types of data poses a great challenge in terms of data volume.

[0127] This embodiment of the present invention ensures temporal and spatial consistency between acoustic emission signals and DIC image data through precise data synchronization and specimen position calibration. These data are then compared and integrated along the same timeline to provide a more accurate analysis of fatigue damage states. S3 identifies asphalt mixture fatigue damage stages based on the changes in the number of impacts and acoustic emission energy within the Δt time window. In S4, the analysis results of the acoustic emission signals and DIC image data are combined according to the corresponding fatigue damage stages, enabling a more accurate assessment of the fatigue damage state and cracking of the asphalt mixture.

[0128] In summary, through rational data processing, the present invention overcomes the technical difficulties encountered when combining acoustic emission and DIC technologies for monitoring fatigue damage and cracking in asphalt mixtures. This helps improve the accuracy and practicality of monitoring, contributing to road maintenance and traffic safety.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A method for quickly identifying the fatigue damage stage of asphalt mixture, characterized in that: Including the following steps: S1. Subject the asphalt mixture specimen to fatigue loading. Collect the acoustic emission signals generated during the fatigue loading through multiple acoustic emission sensors (5) on the surface of the specimen. At the same time, collect the DIC image data of the surface deformation of the asphalt mixture specimen during the fatigue loading through a high-speed industrial CCD camera (2), and store the image data in the DIC acquisition computer (1). S2. According to the real-time acoustic emission data, plot the curve of the cumulative impact number and energy of acoustic emission versus the number of loading cycles. Select multiple DIC image data during the test, perform grayscale processing, screen the areas with larger gradients by setting thresholds, and preliminarily judge the spatio-temporal information of cracks. S3. Based on the curve of the cumulative impact number and energy of acoustic emission versus the number of loading cycles, fit the acoustic emission parameter curve. Based on the changes in the impact number and acoustic emission energy within the window, divide the fatigue damage stage of the asphalt mixture. S4. Align the time of the acoustic emission signal obtained in S3 with the time of the crack spatio-temporal information obtained in S2 to obtain combined characteristic information. Determine the precursor information of asphalt mixture fatigue instability according to the acoustic emission parameter characteristics before the specimen instability failure, and determine the starting position of the specimen instability failure.

2. A method for quickly identifying fatigue damage stages of asphalt mixture according to claim 1, characterized in that: The multiple acoustic emission sensors (5) are arranged on the surface of the asphalt mixture specimen according to the layout principle of alternating in the same plane, offsetting in the relative plane, and global coverage, for receiving the acoustic emission signals on the surface of the asphalt mixture specimen.

3. The method for quickly identifying fatigue damage stage of asphalt mixture according to claim 1 is characterized in that: An acoustic emission main chassis (6) is installed on one side of the asphalt mixture specimen, for sending acoustic emission signals to the asphalt mixture specimen; collecting and converting the acoustic emission signals received by the acoustic emission sensors (5), and finally storing them in the acoustic emission acquisition computer (7).

4. The method for quickly identifying fatigue damage stage of asphalt mixture according to claim 1 is characterized in that: In S1, collect the calibration plate image data through the high-speed industrial CCD camera (2). After processing and analyzing by digital image processing software, confirm the calibration data, establish a data processing coordinate system, and calibrate the position of the specimen.

5. The method for quickly identifying fatigue damage stage of asphalt mixture according to claim 1 is characterized in that: In S2, in the grayscale image of the DIC data, I(x, y) represents the grayscale value of the image. The amplitude G and direction θ of the gradient are calculated by the following formula: Among them, G x , G y Respectively represent the components of the gradient magnitude G in the x and y directions, I x and I y It is the gradient component of the gray value I(x,y) of the image in the x and y directions, and * represents the convolution operation.

6. The method for quickly identifying fatigue damage stage of asphalt mixture according to claim 1 is characterized in that: In S4, the detection method for the mutation points of the impact number and acoustic emission energy within the window: Select the impact number per second and the acoustic emission energy as the indicators of the change in the damage state. Use Python to calculate the impact number and the acoustic emission energy per second using a sliding window. If the slope change in a certain window exceeds the threshold compared with the previous window, the point corresponding to this window is a mutation point.

7. A method for quickly identifying fatigue damage stages of asphalt mixture according to claim 6, characterized in that: The cumulative number of impacts and acoustic emission energy are cumulative values ​​that vary with the number of loadings / time. Δt represents the time interval between two loadings, N represents the number of loadings, and I represents the total number of impacts. N Indicates the number of impacts during the Nth loading, I N-1 Indicates the number of impacts during the N-1th loading; E N represents the acoustic emission energy at the Nth loading, E N-1 represents the acoustic emission energy at the N-1th loading; The number of impacts I within the time Δt s and acoustic emission energy E s It is expressed as: For the mutation points of the cumulative curve, determine them by the change rate of the impact number and acoustic emission energy within the Δt time; the number of loading cycles is N, and the mutation point is M, 1 < M < N. Then the change rate of the impact number and acoustic emission energy within the Δt time before and after the mutation point is expressed as: Where: Δt is the time interval between two loadings, H is the rate of change of the number of impacts within Δt before the mutation point, and H M represents the number of impacts during the Mth loading, H M-1 represents the number of impacts during the M-1th loading; H' is the rate of change of the number of impacts within Δt after the mutation point, H N represents the number of impacts during the Nth loading; E is the rate of change of acoustic emission energy in the Δt time before the mutation point, E M represents the acoustic emission energy at the Mth loading, E M-1 represents the acoustic emission energy during the M-1th loading; E' is the rate of change of acoustic emission energy within Δt after the mutation point, E N Represents the acoustic emission energy at the Nth loading.

8. The method for quickly identifying fatigue damage stage of asphalt mixture according to claim 1 is characterized in that: The high-speed industrial CCD camera (2) is fixed at a certain distance from the specimen at an angle of 15 to 60°, and the high-speed industrial CCD camera (2) is fixed by a tripod.

9. The method for quickly identifying fatigue damage stage of asphalt mixture according to claim 1, characterized in that: A supplementary light source (3) is installed between the high-speed industrial CCD camera (2) and the asphalt mixture specimen; the acoustic emission main chassis (6) and the acoustic emission sensors (5) are connected through an amplifier.

10. A monitoring system for fatigue damage of asphalt mixture, characterized in that: A method for quickly identifying fatigue damage stages of asphalt mixture as claimed in claim 1, comprising: A loading system (4), used for fixing the asphalt mixture specimen and applying fatigue loading to the specimen; The DIC image acquisition module is used to grayscale multiple DIC image data during the test, screen the area with larger gradient by setting the threshold, and preliminarily judge the spatiotemporal information of the crack; The acoustic emission acquisition module is used to input the collected acoustic emission data into the software analysis terminal, and draw a curve chart of the cumulative number of acoustic emission impacts and energy changes with the number of loading times based on the real-time acoustic emission data; fit the acoustic emission parameter curve based on the curve chart of the cumulative number of acoustic emission impacts and energy changes with the number of loading times; and divide the fatigue damage stage of asphalt mixture based on the changes in the number of impacts and acoustic emission energy within the window; The data fusion module is used to align the time of the obtained acoustic emission signal and the crack space-time information, determine the fatigue instability precursor information of the asphalt mixture according to the acoustic emission parameter characteristics before the specimen instability and failure, and determine the starting position of the instability and failure of the corresponding specimen.

Images