Method for measuring dynamic characteristics of concrete engineering apparent diseases

Abstract

The invention belongs to the technical field of concrete engineering nondestructive testing and metering, and discloses a concrete engineering apparent disease dynamic characteristic measuring method, which comprises the following steps: carrying out dry state thermal response pre-scanning on a to-be-measured surface of a structure to generate pollution compensation parameters, and then applying volatile liquid to the to-be-measured surface of the structure; the method comprises the following steps: synchronously applying acoustic excitation, collecting a thermal image time sequence, performing mode switching according to environment humidity, analyzing a compensated and corrected cooling rate, temperature fluctuation and acoustic-thermal correlation in low humidity, or analyzing acoustic-induced local temperature rise in high humidity, and finally generating a quantitative index representing dynamic activity of diseases. According to the method, a dynamic thermal physical response test analysis path under controlled physical excitation is constructed, the diagnosis dimension is improved from a traditional static geometrical shape to direct quantification of concrete structure microcosmic dynamic mechanical behaviors, and direct physical causal association between observable thermal signals and internal structure risks is established; and a decision basis is provided for structural safety assessment.

Description

Technical Field

[0001] This invention relates to a method for measuring the dynamic characteristics of apparent defects in concrete engineering, belonging to the field of non-destructive testing and metrology technology for concrete engineering. Background Technology

[0002] Currently, identifying and measuring structural defects, especially cracks, is a fundamental task to ensure structural safety. Existing technical practices typically involve using optical imaging equipment to acquire high-resolution images of the surface to be measured, and then using image processing algorithms to automatically identify and quantify geometric parameters such as the length, width, and direction of the cracks. This method is more efficient than traditional manual contact measurement.

[0003] However, when technologies based on precise geometric parameter measurement are applied to long-term health monitoring of large-scale concrete engineering facilities, the system outputs a large amount of precise data on the geometric dimensions of defects in field engineering practice. However, these data lack a direct physical correlation with the local mechanical state of the concrete structure where the defects are located. This makes it difficult for technicians to effectively classify and rank the risk levels when faced with massive amounts of geometric data. This is because the entire technology is based on the premise that the geometric shape of cracks is directly related to their mechanical hazards. However, this premise does not hold true in engineering scenarios. Therefore, even if we follow the original path, such as using more complex image algorithms to pursue higher geometric measurement accuracy, we cannot solve the above-mentioned technical problems. Instead, we will generate more data that is difficult to interpret effectively. This is because static geometric shape is not a direct representation of the actual mechanical state of the structure. The potential risks of concrete structures depend more on their dynamic mechanical behavior under actual loads than on their apparent damage morphology at a certain static moment.

[0004] Therefore, a long-standing technical bias exists in this field: focusing on obtaining more precise geometric parameters by improving image resolution and algorithm accuracy to indirectly assess structural risk. The inventors recognize that this approach has inherent limitations, as static geometry is not a direct representation of the actual mechanical state of a structure. The potential risk of a concrete structure depends more on its dynamic mechanical behavior under actual loads. Specifically, existing technologies suffer from the following shortcomings: 1. They lack a testing method that can transform the dynamic mechanical behavior of microscopic deformation caused by structural defects under actual loads into a physical signal that can be directly captured by on-site testing equipment; 2. Due to the inability to obtain information on the dynamic mechanical behavior of defects, existing risk assessments largely rely on the empirical correlation between geometric shape and risk level. This correlation lacks direct physical causal support, resulting in limited reliability of the conclusions. Therefore, how to establish a physical process-based testing and analysis method to directly identify and quantify the dynamic mechanical activity of concrete structural defects, thereby providing a more directly physically meaningful decision-making basis for structural safety assessment, is the technical problem this invention aims to solve. Summary of the Invention

[0005] This invention provides a method for measuring the dynamic characteristics of apparent defects in concrete engineering. Its main purpose is to solve the problem that existing technologies cannot obtain the dynamic mechanical behavior of defects through direct physical measurement methods, and can only conduct indirect assessments by analyzing static geometric morphology, resulting in a lack of direct physical basis for risk assessment.

[0006] To achieve the above objectives, the present invention provides a method for measuring the dynamic characteristics of apparent defects in concrete engineering. The method for measuring the dynamic characteristics of apparent defects in concrete engineering includes the following steps: S1. Before applying a layer of volatile liquid to the test surface of the concrete structure, a dry surface thermal response pre-scan is performed. A compensation matrix characterizing the contamination distribution on the test surface is generated by applying a sequence of transient thermal excitations to the test surface and acquiring its transient thermal response. S2, apply a volatile liquid to the surface to be tested with a preset unit area dose, and during the evaporation of the volatile liquid, use an infrared thermal imaging device to acquire a series of thermal images of the surface to be tested in a continuous time period to form a thermal image time series, and simultaneously apply an acoustic excitation signal with a frequency varying within a preset range to the surface to be tested. S3. Based on the comparison results between ambient humidity and preset threshold, select to perform either the first mode analysis or the second mode analysis. Specifically, when the ambient humidity is detected to be lower than the preset threshold, apply a compensation matrix to normalize the temperature value of each pixel in the thermal image time series, and perform the first mode analysis on the dynamic thermal characteristics of the temperature change of the target area over time in the processed thermal image time series. The first mode analysis includes: calculating the slope and variance of the temperature change curve of the target area, and determining the mechanical resonance characteristics by analyzing the acoustic-thermal correlation between the dynamic temperature spectrum of the target area and the instantaneous frequency change trajectory of the acoustic excitation signal. When the ambient humidity is detected to be not lower than the preset threshold, acoustic energy pumping is performed on the target area, and a second mode analysis is performed on the thermal image time series. The second mode analysis includes analyzing the local temperature rise characteristics of the target area caused by the acoustic energy pumping. S4 generates a quantitative index characterizing the dynamic activity of apparent diseases based on the results of the first or second mode analysis.

[0007] Preferably, the first mode analysis further includes: selecting a region on the surface to be tested where the heat flux gradient is less than a preset gradient threshold as a reference region; and comparing the slope and variance of the target region with the slope and variance of the reference region calculated based on the processed thermal image time series.

[0008] Preferably, the step of generating a quantitative index characterizing the dynamic activity of apparent diseases is calculated based on the difference in slope between the target region and the reference region, the difference in variance between the target region and the reference region, and the analysis results of mechanical resonance characteristics.

[0009] Preferably, the step of determining the mechanical resonance characteristics includes: performing a short-time Fourier transform on the temperature change sequence of the target region over time to obtain a temperature dynamic spectrum; and determining whether there is an energy-enhanced response band in the temperature dynamic spectrum, and whether the acoustic-thermal correlation between the center frequency of the response band and the instantaneous frequency of the acoustic excitation signal is not lower than a preset correlation threshold. If so, the acoustic-thermal correlation is taken as part of the mechanical resonance characteristics.

[0010] Preferably, when the ambient humidity is detected to be not lower than a preset threshold, acoustic energy pumping is performed on the target area, including the following steps: if a response frequency band has been determined to exist, the target area is acoustically excited with the center frequency of the response frequency band; and if a response frequency band has not been determined to exist, the target area is acoustically excited with a wideband acoustic signal.

[0011] Preferably, the quantitative index, namely the disease activity index AI, is calculated according to the following rule: AI=w_1·|Δk|+w_2·|Δ(σ²)|+w_3·γ_max, where AI is the disease activity index, Δk is the difference in slope between the target area and the reference area, Δ(σ²) is the difference in variance between the target area and the reference area, γ_max is the maximum value of the acoustic-thermal correlation, and w_1, w_2 and w_3 are preset weighting coefficients.

[0012] Preferably, after generating the quantification index, the method further includes the following steps: applying a transient thermal pulse to one side of the identified apparent disease; collecting and analyzing the time delay of the temperature response caused by the transient thermal pulse on the other side of the apparent disease; and determining the depth information of the apparent disease based on the time delay.

[0013] Preferably, the method further includes the following steps: analyzing the temperature distribution uniformity of the reference area in the processed thermal image time series to determine the material degradation state at the location of the apparent defects; the analysis of temperature distribution uniformity includes calculating the information entropy of each frame of the thermal image of the reference area in a continuous time period, generating a curve of information entropy changing with time, and determining the material degradation state based on the morphological characteristics of the curve of information entropy changing with time.

[0014] Preferably, the volatile liquid is water.

[0015] Preferably, it further includes: after identifying and quantifying the apparent defects, using a non-contact detection method based on electrical principles to measure the electrical conductivity or impedance characteristics of the target area; and verifying or arbitrating the quantification index based on the measurement results.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. By applying volatile liquid to the surface under test and continuously acquiring time series of thermal images during its evaporation process, a test path directly correlates the micro-dynamic behavior of concrete structures with changes in surface thermophysical characteristics was established. Under the influence of structural micro-dynamics, the liquid evaporation process in the diseased area is disturbed, which alters the rate and fluctuation characteristics of temperature change over time in that area. By analyzing this dynamic thermal characteristic, the dynamic activity of the disease that cannot be distinguished by geometric morphology observation can be identified, providing a physical process-based analytical method for assessing structural condition.

[0017] 2. While acquiring the time series of thermal images, an acoustic excitation signal with varying frequency is applied to the surface under test. By establishing the physical index of acoustic-thermal correlation, this method can not only identify whether the disease is active, but also identify its sensitivity to specific frequency excitations. This provides direct evidence that traditional methods cannot provide for judging the cause of the disease (such as resonant fatigue), achieving an improvement from judging the presence or absence to mechanism analysis. This testing method not only identifies the dynamic activity of the disease, but also further obtains its mechanical resonance characteristics, so that the judgment of the cause and sensitivity of the disease has a specific physical basis.

[0018] 3. In the process of analyzing the dynamic thermal characteristics of the diseased area, the same set of thermal image time series can also be used to analyze the background area around the disease. Differences in the physical state of the material, such as different densities and porosities, will lead to different uniformities of surface temperature distribution during liquid evaporation. By analyzing the temperature distribution characteristics of the background area thermal images, the material degradation information of the diseased location can be obtained simultaneously. Thus, in one test operation, a dual assessment of the mechanical behavior of the disease and the state of the matrix material can be completed at the same time. Attached Figure Description

[0019] Figure 1 This is a flowchart of the dual-mode analysis method based on environmental humidity determination of the present invention; Figure 2 This is a comparison diagram of the acoustic-thermal correlation response characteristics of dynamic and static diseases according to the present invention. Figure 3 This is an interactive diagram illustrating the application scenarios for the implementation and decision support of the technology of this invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. However, it should be understood that the specific embodiments described below are only illustrative and are not intended to limit the scope of protection of the present invention.

[0021] A method for measuring the dynamic characteristics of apparent defects in concrete engineering is proposed, which establishes a dynamic thermophysical response testing and analysis process under controlled physical excitation. This process mainly includes three stages: pre-scanning of dry surface thermophysical characteristics, excitation and acquisition of wet dynamic response, and dual-mode analysis and quantification based on environmental conditions. Through this process, the dynamic mechanical behavior of the structure under external loads, which is difficult to observe directly, is converted into a set of objectively measurable and quantifiable dynamic thermal characteristic parameters, thereby generating a quantitative index characterizing the dynamic activity of the defects. In engineering sites, the surface of the concrete structure under test is often exposed for a long time, and its thermal emissivity and hydrophilicity are often uneven due to the random distribution of contaminants such as dust and oil. Uniformity, or non-uniformity, can generate background thermal signals unrelated to the actual state of the disease in subsequent tests, thus affecting the accuracy of the analysis results. To avoid this physical interference, this method performs a pre-scanning step of dry surface thermal response before applying liquid for the main testing process. This step uses a directional light source, such as a laser diode or a high-power light-emitting diode (LED), to scan the surface under test along a preset gridded path. A standardized, millisecond-level transient thermal excitation is applied to each grid point. Simultaneously, an industrial infrared thermal imager with a temperature measurement accuracy of no less than 0.1℃ and a sampling frequency of no less than 50Hz records the temperature of each point after the micro-thermal pulse. Transient temperature response curves, i.e., the heating and cooling processes; given the differences in thermal inertia and thermal conductivity of surface contaminants of different properties, identifiable thermal texture features will be generated accordingly. The system analyzes the peak temperature or cooling rate of these transient thermal response curves to generate a compensation matrix that corresponds one-to-one with the pixels of the main thermal image. Each value in this matrix quantifies the expected thermal signal deviation of that pixel due to surface contamination, providing a pixel-by-pixel correction basis for subsequent wet measurement data. After completing the dry pre-scan and generating the compensation matrix, the wet dynamic response excitation and acquisition stage begins. This stage uses standard spraying equipment with a preset unit area dosage, such as 50 per square meter. mL, a layer of volatile liquid is uniformly applied to the pre-scanned surface to be tested. In one specific embodiment, the volatile liquid is water. During a preset time period after spraying, such as 3 to 5 minutes, the infrared thermal imaging device continuously records video of the target area to acquire a series of thermal images that are continuous in time, forming a thermal image time series. During the acquisition process of the infrared thermal imaging device, a directional acoustic generator simultaneously applies an acoustic excitation signal with a frequency within a preset range, such as from 10 Hz to 500 Hz, that changes linearly with time to the surface to be tested, actively using a controllable external excitation source to stimulate the dynamic response that may exist in the structural defects.

[0022] After acquiring the thermal image time series, the system enters the analysis and quantification stage. This stage employs an environmentally aware adaptive analysis mode switching mechanism to address changes in the dominant physical effects under different humidity conditions. The system monitors on-site environmental parameters in real time using a temperature and humidity sensor, and uses a preset environmental humidity threshold, such as 95% relative humidity, as the criterion for mode switching. When the detected environmental humidity is below this preset threshold, liquid evaporation cooling becomes the dominant thermophysical process, and the system executes the first mode analysis. This mode first calls the compensation matrix generated in the pre-scanning stage to normalize the original temperature value of each pixel in the acquired thermal image time series frame by frame to filter out background noise caused by surface contamination. Subsequently, in the processed thermal image time series, the system selects a region with a heat flux gradient less than a preset gradient threshold as a reference region, and extracts the temperature change curve of each pixel or pixel region over time for the target region to be analyzed. The first mode analysis specifically includes: calculating the target... The system calculates the slope of the temperature change curve in the target area, which represents the average cooling rate of the area; it also calculates the variance of the temperature change curve in the target area, which represents the fluctuation range of the temperature change in the area; and it determines the mechanical resonance characteristics by analyzing the acoustic-thermal correlation between the dynamic temperature spectrum of the target area and the instantaneous frequency trajectory of the acoustic excitation signal. The acoustic-thermal correlation is determined by calculating the Pearson correlation coefficient between the center frequency trajectory of the response band and the instantaneous frequency trajectory of the acoustic excitation signal. This analysis is implemented by performing a short-time Fourier transform (STFT) on the temperature change sequence of the target area over time to obtain its dynamic temperature spectrum, and then determining whether there is an energy-enhanced response band in the dynamic temperature spectrum, and whether the acoustic-thermal correlation between the center frequency of the response band and the instantaneous frequency trajectory of the acoustic excitation signal is not lower than a preset correlation threshold, such as 0.8. Finally, the system compares the slope and variance of the target area with the slope and variance of the benchmark area calculated from the same data source to obtain the difference.

[0023] When the ambient humidity is detected to be not lower than a preset threshold, the liquid evaporation process is suppressed, and the system automatically switches to the second analysis mode. In this mode, the acoustic generator changes its function from frequency sweep detection to energy pumping. The system will continuously excite the target area with a wideband acoustic signal based on the mechanical resonance characteristics that may have been measured in the first mode, using the resonance frequency corresponding to that characteristic, or, if the resonance characteristics are not determined, a resonance frequency corresponding to that characteristic. Correspondingly, the analysis target of the infrared thermal imager changes from observing the cooling process to capturing the local temperature rise characteristics caused by sound energy dissipation. A dynamically active disease, whose internal interface is affected by sound waves... The excitation generates more intense friction, causing more acoustic energy to be converted into heat energy, which appears as a localized temperature rise area on the differential thermal image. The second mode analysis uses lock-in amplification technology to extract this temperature rise signal, which originates from the excitation signal, from the background thermal noise and quantifies its intensity and range. Finally, based on the results of the first or second mode analysis, the system generates a quantitative index characterizing the dynamic activity of the apparent disease. This index is a comprehensive technical indicator whose value reflects the degree of dynamic response of the disease under external load excitation. In the first mode, this index, namely the disease activity index AI, can be determined by a... The calculation is performed using a weighted summation formula: AI = w_1·|Δk| + w_2·|Δ(σ²)| + w_3·γ_max. In this formula, AI is the disease activity index; Δk is the difference in slope between the target region and the baseline region, in °C / s; Δ(σ²) is the difference in variance between the target region and the baseline region, in °C²; γ_max is the maximum value of the acoustic-thermal correlation, a dimensionless parameter; w_1, w_2, and w_3 are a set of preset weighting coefficients calibrated based on experimental data. This method may also include analysis of disease depth information and material degradation status. After generating the quantified index, the analysis is performed on the identified... For other diseases, a precisely directional light source can be used to apply a transient thermal pulse to one side of the disease. A high frame rate infrared thermal imager can be used to collect and analyze the time delay of the temperature response caused by the thermal pulse passing through the water medium inside the disease to the other side. The depth information of the disease can be determined based on this time delay. At the same time, the uniformity of temperature distribution in the reference area around the disease can be analyzed using the acquired thermal image time series. Specifically, the information entropy of each frame of the thermal image in the area within a continuous time period can be calculated, and a curve of information entropy changing with time can be generated. The material degradation state at the location of the disease can be determined based on the shape characteristics of the curve.To further verify the analysis results, this method may also include an independent verification or arbitration step. After completing the analysis based on thermophysical principles, a handheld reinforced concrete conductivity meter or similar equipment can be used to measure the conductivity characteristics of the target area. The specific verification or arbitration rules are as follows: When the measured conductivity value of the target area is higher than a preset threshold, it indicates that the defect has penetrated deep into the structure and may have reached the reinforcing steel, forming an electrical pathway. In this case, if the previously generated defect activity index (AI) value is already high, then this electrical measurement result constitutes verification of high risk. If the AI ​​value is not high, then this electrical result will serve as the arbitration basis, raising the risk level of the defect. Although its dynamic activity is not strong, it provides a direct path for media intrusion. Conversely, if the measured conductivity value is lower than the preset threshold, it indicates that it is a surface defect, and its risk level is mainly determined by the AI ​​value. In this way, independent verification or arbitration of the quantitative index is achieved.

[0024] Example 1: To illustrate the application of this invention in the field of precision measurement in concrete engineering, in an application of structural energy measurement of a large cross-sea bridge's T-shaped concrete main beam, the bridge surface had hundreds of surface defects with similar widths and lengths. Using optical image analysis, a list containing the geometric parameters of each defect was obtained. However, the maintenance team, faced with this list, could not determine which defects were surface material shrinkage textures, or which were dynamic active defects with potential risks caused by structural fatigue or overloading, making it difficult to formulate a targeted maintenance plan. To solve this problem, the method claimed in this invention was adopted. In the field test, firstly, under an ambient humidity of 65%, the integrated system mounted on the UAV was used to perform a dry surface thermal response pre-scan on a key test area of ​​the bridge's main beam. The system used a laser diode as the thermal excitation source to perform a gridded scan of the test area surface, and an infrared thermal imager simultaneously collected the transient thermal response at each point to generate a compensation matrix characterizing the surface contamination distribution of the test area. Next, the system uniformly sprayed water onto the test area at a dose of 50 mL per square meter. Simultaneously, while acquiring the thermal image time series, a directional acoustic generator applied an acoustic excitation signal with a linear frequency variation from 10 Hz to 500 Hz to the test area.

[0025] Among the collected data, two defects, each approximately 0.2 mm wide, were identified in the optical image and labeled as target area A and target area B, respectively. The system detected that the current ambient humidity was below a preset threshold of 95%, thus triggering the first mode of analysis. The system first applied a compensation matrix generated from a dry pre-scan to normalize the original thermal image time series pixel by pixel. This step provided a data foundation for subsequent dynamic thermal feature analysis, eliminating interference from surface thermophysical property inhomogeneities. Subsequently, the system extracted the temperature change over time curves of target areas A and B after processing. The temperature change curve of target region A was compared with that of a reference region extracted from a perfectly flat surface within the test area. The analysis showed that the slope and variance of the temperature change curve of target region A were very similar to those of the reference region. Furthermore, when performing acoustic-thermal correlation analysis on its temperature dynamic spectrum and the instantaneous frequency of the acoustic excitation signal, no peak value of acoustic-thermal correlation was found, indicating that its thermophysical behavior was basically consistent with that of the perfectly flat surface. In contrast, the absolute value of the slope of the temperature change curve of target region B was greater than that of the reference region, and its variance was also higher, indicating that its cooling process was faster and more unstable. Specifically, by performing a short-time Fourier transform on the temperature sequence of target region B, an energy-enhanced response band was found in its temperature dynamic spectrum. The center frequency of this band consistently changes around 88 Hz, following the instantaneous frequency of the acoustic excitation signal, with a maximum acoustic-thermal correlation coefficient γ_max reaching 0.92. This highly locked relationship between the acoustic excitation frequency and the thermal signal fluctuation frequency, along with the synchronous anomaly of the cooling rate and temperature fluctuation amplitude, constitutes a chain of physical evidence: target region B not only exhibits dynamic activity, but this activity is also sensitive to external mechanical vibration excitation at 88 Hz. Ultimately, the system... Based on the calculation rules of the disease activity index AI disclosed in the aforementioned specific implementation method, the quantitative indexes of target area A and target area B are calculated respectively. The AI ​​value of target area A is close to zero, while the AI ​​value of target area B is higher than the preset attention threshold. In this way, two diseases that could not be distinguished in terms of risk level based on static geometric dimensions can be quantitatively distinguished with clear physical meaning through a single test. This allows the maintenance team to focus the resources used for in-depth exploration directly on disease points with high potential risk, such as target area B, which have been identified as having high dynamic activity and specific resonance frequencies.

[0026] Example 2: To objectively verify the effectiveness of the method claimed in this invention in distinguishing between different types of apparent defects, this example constructs a verification experiment including a control group and an experimental group. The aim is to demonstrate, through quantitative data, that the method can identify defects with different mechanical properties but similar geometric shapes. The experimental platform consists of standard concrete specimens, loading and environmental control equipment, and a data acquisition and analysis system conforming to the functional specifications of the aforementioned specific embodiments. The specimens used in the experiment are two sets of C40 concrete beams, each 400mm × 100mm × 100mm in size. For the first set of specimens, early water loss was controlled by pre-fabricating a 0.2mm wide strip on its surface. The first group of specimens, with static shrinkage cracks, was defined as static defect samples. The second group of specimens had an unbonded prestressed steel bar embedded during casting. By applying a small tensile stress later and subjecting it to low-cycle reciprocating loading under stress, a crack with a width of 0.2 mm was pre-fabricated on its surface. A servo hydraulic actuator applied a continuous reciprocating displacement of ±0.005 mm at a frequency of 1 Hz to simulate dynamic active defects that continuously open and close under real loads. This group of specimens was defined as dynamic defect samples. The entire test was conducted in an environmental chamber with a constant temperature of 25℃ and a constant relative humidity of 70% to eliminate the interference of environmental temperature and humidity fluctuations.

[0027] The experiment consisted of two treatment methods. The first method was the control group, which used a technique that included only thermal imaging analysis and not acoustic excitation to test the two disease samples. In this method, the system only collected and analyzed the cooling process of the specimen after wetting, and calculated the slope and variance of its temperature change curve. The second method was the sample group of the present invention, which used the complete technical solution disclosed in the aforementioned specific embodiments, namely, the complete process including dry pre-scanning, wet thermal response acquisition, synchronous acoustic excitation, and dual-mode analysis, to test the two groups of disease samples. The frequency of the acoustic excitation signal was set to a linear scan from 10Hz to 500Hz, and the weighting coefficients of the disease activity index AI were set to w_1=0.4, w_2=0.4, and w_3=0.2. During the experiment, the above two treatment methods were performed on the static and dynamic disease samples respectively, the key physical quantities in each group of experiments were collected, and the final disease activity index AI was calculated. The results are recorded in Table 1.

[0028] Table 1: Comparison of test data for static and dynamic disease samples under different testing methods.

[0029]

[0030] Experimental data show that by introducing a controlled, frequency-varying acoustic excitation signal and analyzing the correlation between this excitation and the thermophysical response at the defect site over time, the dynamic mechanical behavior of the defect can be amplified. The method claimed in this invention, through this synergistic analysis of thermal and acoustic features, can distinguish dynamically active defects in concrete structures compared to a single passive thermal imaging analysis method.

[0031] To further highlight the necessity and substantial contribution of the technique of simultaneously applying acoustic excitation and analyzing the acoustic-thermal correlation in the method of the present invention, the following comparative examples are provided.

[0032] Comparative Example 1: To further verify the effectiveness of the key technical steps in the method of this invention, this comparative example adopts a simplified technical solution that omits the acoustic excitation step. A sample with weaker dynamic characteristics is tested. Except for the absence of a synchronous acoustic excitation signal applied to the specimen, the test platform, C40 concrete specimens (including a static defect sample with a 0.2mm wide static shrinkage crack and a 0.2mm wide dynamic active defect sample with a weaker ±0.002mm continuous reciprocating downward movement), environmental control conditions (temperature 25℃, relative humidity 70%), dry pre-scanning procedure, application method and dosage of volatile liquid (water), and acquisition parameters of the infrared thermal imaging equipment are all completely consistent with the conditions used in the test group using the method of this invention in Example 2. Under this technical solution, the data analysis process can only analyze the cooling process after wetting, and the final test data is recorded in Table 2.

[0033] Table 2: Test data table of weak dynamic disease samples under simplified technical solution.

[0034]

[0035] Experimental results show that, in the absence of synchronous acoustic excitation and subsequent acoustic-thermal correlation analysis, when faced with disease samples with weaker dynamic characteristics (±0.002mm displacement), the slope difference and variance difference are almost indistinguishable from those of static disease samples. The disease activity index AI values ​​calculated by both are almost identical (0.001 and 0.003, respectively), completely losing their distinguishing ability. This result confirms that relying solely on passive thermal imaging analysis is fundamentally unable to separate the microscopic dynamic mechanical behavior of the disease from the background thermal noise, especially when the disease activity is weak, this simplified method completely fails.

[0036] Example 3: This example combines Figures 1 to 3 The method for measuring the dynamic characteristics of surface defects in concrete engineering is explained, such as... Figure 1As shown, the process first performs a dry surface thermal response pre-scan to generate a compensation matrix characterizing surface contamination. Then, it enters the wet dynamic response excitation and acquisition stage. In this stage, volatile liquid and acoustic excitation are applied to the surface under test, and a thermal image time series is acquired simultaneously. Based on this series, the depth of the damage can be determined in parallel by analyzing the transient thermal pulse response time delay, and the material degradation state can be analyzed by analyzing the temperature distribution uniformity of the reference area. The core of this process lies in an adaptive analysis switching based on environmental humidity. When the environmental humidity is below a preset threshold, the first mode analysis is performed, which analyzes the cooling rate, temperature fluctuation, and acoustic-thermal correlation. When the environmental humidity is not below the preset threshold, the process switches to the second mode analysis, which analyzes the acoustic-induced local temperature rise characteristics. The output results of both analysis modes are combined in the step of generating a quantitative index, ultimately outputting a damage activity index AI and forming a quantitative assessment report.

[0037] like Figure 2 As shown in the figure, this graph represents the relationship between acoustic-thermal correlation and acoustic excitation frequency (Hz). The static defect area, indicated by the dashed line, maintains a low acoustic-thermal correlation of no more than 0.2 across the entire acoustic excitation frequency range. In contrast, the dynamic defect area, indicated by the solid line, exhibits a significant peak in acoustic-thermal correlation at the acoustic excitation frequency of 88 Hz, reaching a maximum value of 0.92. Figure 3 As shown in the diagram, the technicians, acting as users, interact with the system to perform multiple operations, including performing disease scanning and data acquisition, calibrating system parameters, assessing the material deterioration status, and analyzing disease depth information. Among these, the core operation of performing disease scanning and data acquisition directly outputs a disease activity index, which is the core basis for generating the final quantitative assessment report. This report is then submitted to the structural maintenance team to provide decision support for developing a maintenance plan.

[0038] Example 4: To ensure that the disease activity index AI output by the method claimed in this invention has a consistent and traceable physical meaning when applied to concrete structures with different material properties or load environments, a standardized offline calibration procedure needs to be implemented to determine the weighting coefficients w_1, w_2, and w_3 in its core calculation formula, as well as the final risk discrimination threshold. The implementation of this procedure begins with the preparation of a calibration sample set, which consists of at least four specimens with the same concrete grade and reinforcement ratio as the target structure. All specimens are 400mm×100mm×100mm in size. Specimen S0 is a static defect sample, with a static crack formed on its surface through natural drying shrinkage. Specimens S1, S2, and S3 are dynamic defect samples, and continuous reciprocating displacements of ±0.002mm, ±0.005mm, and ±0.010mm are applied to them respectively using a servo hydraulic actuator to simulate three different levels of dynamic activity. A dimensionless baseline activity label is assigned to each of them, with values ​​of 0.0, 0.3, 0.6, and 1.0, respectively.

[0039] In a controlled laboratory environment, the complete test procedure disclosed in the aforementioned specific implementation method was performed on each specimen in the calibration sample set. Thermal image time series under wetting and acoustic excitation were acquired, and corresponding physical characteristic parameters were extracted. Specifically, to determine the calculation path for the maximum acoustic-thermal correlation γ_max, after obtaining the temperature dynamic spectrum through short-time Fourier transform, the system first identifies the time-frequency trajectory of the response band with the strongest energy. Then, this trajectory is processed as a two-dimensional vector with the known linear frequency scan trajectory of the input acoustic excitation signal, and the Pearson correlation coefficient between the two is calculated. This coefficient value is used as the value of the maximum acoustic-thermal correlation γ_max. Through measurements of four specimens, a calibration dataset containing a baseline activity label and corresponding physical characteristic parameters is obtained. Next, to determine the weighting coefficients w_1, w_2, and w_3, the system uses a multiple linear regression analysis method to fit the calibration dataset. The goal of this method is to find a set of coefficients such that, according to the formula AI=w_1·|Δk|+w_2·|Δ(σ²)|+w The AI-predicted value of the disease activity index calculated by _3·γ_max has the smallest mean square error between it and the baseline activity label of the sample. In a calibration test for a certain type of bridge T-beam, the data obtained was input into the regression model, and a set of weighting coefficients was obtained: w_1=0.38, w_2=0.41, w_3=0.21. After the weighting coefficients are determined, the AI ​​value can be calculated for any disease to be tested. When setting a threshold of concern for risk screening for this AI value, in order to balance the sensitivity and specificity of detection, the system further adopts receiver operation. The characteristic curve (ROC) analysis is used to determine the threshold. This analysis requires a set of multiple known static and dynamic validation samples. The system uses pre-calibrated weighting coefficients to calculate the AI ​​value of all validation samples. By traversing all possible thresholds, the true positive rate and false positive rate at each threshold are calculated and plotted as ROC curves. Finally, the AI ​​value corresponding to the point closest to the top left corner of the curve is selected as the attention threshold for this application scenario, such as 0.25. This threshold represents an operating point where the risk of missed detection and the risk of false positive are balanced under this calibration system.

[0040] Example 5: In an application to detect concrete wharf pile foundations eroded by marine salt spray, the initial test was conducted in the morning when the ambient humidity was 78%, which was lower than the preset threshold of 95%. The system used the first mode analysis disclosed in the aforementioned specific implementation to work. By applying an acoustic excitation signal with a frequency varying from 10Hz to 500Hz, a dynamic active defect was identified, and its mechanical resonance characteristics were determined to correspond to a frequency of 75Hz.

[0041] By the afternoon, due to fog on the sea surface, the ambient humidity rose rapidly. When the system's built-in temperature and humidity sensor detected that the ambient humidity reached 95%, the system automatically triggered a switch in its operating mode. At this time, the system's acoustic generator function changed from frequency sweep detection to energy pumping, continuously injecting acoustic energy into the affected area at the determined 75Hz resonant frequency. Simultaneously, the system's analysis software switched its analysis target from observing the cooling process to capturing local temperature rises. By locking and amplifying the acquired thermal image time series, a weak local temperature rise signal generated by interface friction dissipation at the same frequency as the 75Hz excitation signal was extracted from the background thermal noise, confirming the dynamic activity of the affected area. Through this adaptive switching of operating modes, the method claimed in this invention can still maintain the ability to identify the dynamic activity of the affected area through the physical mechanism of acoustic dissipation heat generation even when the evaporative cooling effect is suppressed.

[0042] Example 6: Before quantitatively detecting the depth of defects in a structure using concrete of a specific grade, an offline calibration procedure needs to be performed to establish the mapping relationship between the thermal conduction delay Δt of the cross-crack under this specific material and the depth of defects D. The procedure first prepares a set of calibration specimens using the same batch and mix proportion of concrete as the structure to be tested. On the surface of these specimens, thin metal sheets of different depths (e.g., 5mm, 10mm, 15mm, 20mm and 25mm) are pre-embedded and then removed after the concrete has set. This creates a series of standard artificial cracks with known geometric depths.

[0043] Subsequently, all calibration specimens were wetted, and for each standard artificial crack with a known depth D, a depth information determination step was performed. This involved applying a standardized transient thermal pulse to one side of the crack and using a high-frame-rate infrared thermal imager to record and calculate the time delay Δt for the thermal response to travel to the other side of the crack. Through systematic testing of all calibration specimens, a (D, Δt) dataset was obtained. The data points in this dataset were plotted in a two-dimensional coordinate system, and the least squares method was used to fit a function to establish a quantitative mathematical relationship that describes the change of the defect depth D with the cross-crack heat conduction delay Δt under this specific concrete material. For example, a mathematical model of the form D equals a multiplied by (Δt raised to the power of negative b) plus c, where a, b, and c are calibration coefficients obtained through fitting that are applicable to this material. This mathematical relationship is used in the on-site detection system to directly calculate and output the corresponding defect depth estimate from any newly measured time delay.

[0044] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for measuring the dynamic characteristics of apparent defects in concrete engineering, characterized in that, Includes the following steps: S1. Before applying a layer of volatile liquid to the surface of the concrete structure to be tested, perform a dry surface thermal response pre-scan. By applying a series of transient thermal excitations to the surface to be tested and collecting its transient thermal response, a compensation matrix characterizing the contamination distribution on the surface to be tested is generated. S2. Apply a volatile liquid to the surface to be tested at a preset unit area dose, and during the evaporation of the volatile liquid, use an infrared thermal imaging device to acquire a series of thermal images of the surface to be tested over a continuous time period to form a thermal image time series, and simultaneously apply an acoustic excitation signal with a frequency varying within a preset range to the surface to be tested. S3. Based on the comparison results between ambient humidity and preset threshold, select to perform either the first mode analysis or the second mode analysis, specifically including: When the ambient humidity is detected to be lower than the preset threshold, the temperature value of each pixel in the thermal image time series is normalized by applying a compensation matrix, and the dynamic thermal characteristics of the target area temperature change over time in the processed thermal image time series are analyzed by a first mode. The first mode analysis includes: calculating the slope and variance of the temperature change curve of the target area, and determining the mechanical resonance characteristics by analyzing the acoustic-thermal correlation between the dynamic temperature spectrum of the target area and the instantaneous frequency change trajectory of the acoustic excitation signal. When the ambient humidity is detected to be not lower than the preset threshold, acoustic energy pumping is performed on the target area, and a second mode analysis is performed on the thermal image time series. The second mode analysis includes analyzing the local temperature rise characteristics of the target area caused by the acoustic energy pumping. S4. Based on the results of the first or second mode analysis, generate a quantitative index that characterizes the dynamic activity of the apparent disease.

2. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 1, characterized in that, The first mode analysis also includes: selecting a region on the surface to be tested where the heat flux gradient is less than a preset gradient threshold as a reference region; and comparing the slope and variance of the target region with the slope and variance of the reference region calculated based on the processed thermal image time series.

3. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 2, characterized in that, The step of generating a quantitative index that characterizes the dynamic activity of apparent diseases is based on the difference in slope between the target region and the reference region, the difference in variance between the target region and the reference region, and the analysis results of mechanical resonance characteristics.

4. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 1, characterized in that, The steps for determining the mechanical resonance characteristics include: performing a short-time Fourier transform on the temperature change sequence of the target region over time to obtain the temperature dynamic spectrum; and determining whether there is an energy-enhanced response band in the temperature dynamic spectrum, and whether the acoustic-thermal correlation between the center frequency of the response band and the instantaneous frequency of the acoustic excitation signal is not lower than a preset correlation threshold. If such a band exists, the acoustic-thermal correlation is taken as part of the mechanical resonance characteristics.

5. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 1, characterized in that, When the ambient humidity is detected to be not lower than a preset threshold, acoustic energy pumping is performed on the target area, including the following steps: if a response frequency band has been determined to exist, the target area is acoustically excited with the center frequency of the response frequency band; and if a response frequency band has not been determined to exist, the target area is acoustically excited with a wideband acoustic signal.

6. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 3, characterized in that, The quantitative index, also known as the disease activity index AI, is calculated using the following rule: AI = w_1·|Δk| + w_2·|Δ(σ²)| + w_3·γ_max, where AI is the disease activity index, Δk is the difference in slope between the target area and the baseline area, Δ(σ²) is the difference in variance between the target area and the baseline area, γ_max is the maximum value of the acoustic-thermal correlation, and w_1, w_2, and w_3 are preset weighting coefficients.

7. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 1, characterized in that, After generating the quantitative index, the following steps are also included: applying a transient thermal pulse to one side of the identified apparent disease; collecting and analyzing the time delay of the temperature response caused by the transient thermal pulse on the other side of the apparent disease, and determining the depth information of the apparent disease based on the time delay.

8. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 2, characterized in that, The method also includes the following steps: analyzing the temperature distribution uniformity of the reference area in the processed thermal image time series to determine the material degradation state at the location of the apparent defects; the analysis of temperature distribution uniformity includes calculating the information entropy of each frame of the thermal image of the reference area in a continuous time period, generating a curve of information entropy changing with time, and determining the material degradation state based on the morphological characteristics of the curve of information entropy changing with time.

9. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 1, characterized in that, The volatile liquid is water.

10. The method for measuring the dynamic characteristics of apparent defects in concrete engineering according to claim 1, characterized in that, Also includes: After identifying and quantifying the apparent defects, a non-contact detection method based on electrical principles is used to measure the conductivity or impedance characteristics of the target area. And based on the measurement results, the quantitative index is verified or arbitrated.

Images