Method and system for detecting impact damage of bamboo fiber concrete open-web floor

Abstract

The application provides a kind of bamboo fiber concrete open-web floor impact damage detection method and system, effective ultrasonic signal is extracted from the ultrasonic signal of bamboo fiber concrete open-web floor impact damage area;Damage echo signal of bamboo fiber concrete open-web floor is extracted from effective ultrasonic signal, damage echo feature and nonlinear damage feature of extraction damage echo signal;Damage echo feature and nonlinear damage feature are fused into the damage fusion feature of bamboo fiber concrete open-web floor based on the historical impact damage feature of the historical impact damage data of bamboo fiber concrete open-web floor;Impact damage type result of bamboo fiber concrete open-web floor is determined based on the historical labeled sample set of bamboo fiber concrete open-web floor composite damage and damage fusion feature.The application can reduce the influence of non-damage interference of bamboo fiber concrete open-web structure on the detection of bamboo fiber concrete open-web floor impact damage.

Description

Technical Field

[0001] This application relates to the field of damage detection technology, and more specifically, to a method and system for detecting impact damage in bamboo fiber concrete hollow floor slabs. Background Technology

[0002] Damage detection is a widely used technology in the engineering field. Its core purpose is to identify, locate, and assess the damage state of an object or structure, such as defects, cracks, and corrosion, without destroying it. By conducting regular or real-time inspections of materials or components, potential safety hazards can be detected in a timely manner, preventing sudden failures or catastrophic accidents caused by the accumulation of damage.

[0003] In the field of modern construction engineering, bamboo fiber concrete, with its lightweight, high strength, and environmentally friendly properties, is often used to manufacture floor slabs with internal cavities to achieve structural lightweighting. These components are prone to damage from impact loads during service, and accurate detection of damage is crucial for structural safety. Traditional ultrasonic testing methods face significant bottlenecks in the detection of these components: the internal cavity structure of the component can cause interference such as reflection and diffraction of the detection signal, resulting in the masking of effective information in the original detection signal and making it difficult to directly distinguish between cavity interference and real damage signals; at the same time, existing detection methods mostly rely on real-time signal analysis alone, failing to fully utilize the experience and rules accumulated from historical detection data and damage cases, resulting in insufficient accuracy in identifying single and complex damage, and easy to make misjudgments or omissions. In addition, the bonding characteristics of bamboo fiber and concrete matrix further affect the signal propagation law, and traditional detection methods are not specifically adapted to the signal characteristics of this material, resulting in limited accuracy in damage feature extraction. Therefore, how to reduce the impact of non-destructive interference from the hollow structure of bamboo fiber concrete on the impact damage detection of hollow bamboo fiber concrete floor slabs has become a problem facing the industry. Summary of the Invention

[0004] This application provides a method and system for detecting impact damage in hollow bamboo fiber concrete slabs, which can reduce the impact of non-destructive interference on the impact damage detection of hollow bamboo fiber concrete slabs.

[0005] In a first aspect, this application provides a method for detecting impact damage in bamboo fiber reinforced concrete hollow floor slabs, wherein an ultrasonic flaw detector is used to perform ultrasonic testing on the impact-damaged areas of the bamboo fiber reinforced concrete hollow floor slab using the single-probe pulse-echo method. The method includes the following steps:

[0006] Ultrasonic signals were collected from the impact-damaged area of ​​a bamboo fiber concrete hollow floor slab.

[0007] Based on the hollow information of the bamboo fiber concrete hollow floor slab, interference suppression is performed on the ultrasonic signal to obtain an effective ultrasonic signal;

[0008] The effective ultrasonic signal is subjected to damage signal feature extraction to obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals, and the damage echo features and nonlinear damage features of the damage echo signal are extracted.

[0009] Historical impact damage data of the bamboo fiber concrete hollow floor slab is obtained. Based on the historical impact damage characteristics of the historical impact damage data, the damage echo characteristics and the nonlinear damage characteristics are fused to obtain the damage fusion characteristics of the bamboo fiber concrete hollow floor slab.

[0010] Based on the historical labeled sample set of composite damage of the bamboo fiber concrete hollow floor slab and the damage fusion features, the impact damage type of the bamboo fiber concrete hollow floor slab is identified, and the impact damage type result of the bamboo fiber concrete hollow floor slab is obtained.

[0011] In some embodiments, the ultrasonic testing of the impact-damaged area of ​​a bamboo fiber concrete hollow floor slab using an ultrasonic flaw detector with a single-probe pulse echo method further includes: applying a coupling agent to the probe and then tightly attaching it to the surface of the damaged area and the surrounding normal area; alternately transmitting and receiving ultrasonic signals using a single probe; moving the probe at a constant speed along a preset path to perform a full-coverage scan; and collecting echo signals formed by the reflection of ultrasonic waves through the damaged interface, the hollow structure, and the concrete matrix as the ultrasonic waves propagate inside the floor slab in real time.

[0012] In some embodiments, the ultrasonic signal includes a reference signal component of ultrasonic waves propagating in a normal concrete matrix, an abnormal reflection / scattering signal component generated by the impact-damaged area, and a characteristic signal component formed by the interference of the hollow structure on the ultrasonic wave propagation path.

[0013] In some embodiments, the interference suppression of the ultrasonic signal based on the hollow information of the bamboo fiber concrete hollow floor slab to obtain an effective ultrasonic signal specifically includes:

[0014] Obtain the hollow information of the bamboo fiber concrete hollow floor slab;

[0015] Spatial coordinate matching is performed between the fasting information and the probe position of the ultrasound detection to establish the correspondence between the signal acquisition position and the fasting structure;

[0016] Based on the correspondence and the emptying information, interference signals from the emptying structure in the ultrasonic signal are removed to obtain an effective ultrasonic signal.

[0017] In some embodiments, the damage signal feature extraction of the effective ultrasonic signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals specifically includes:

[0018] The effective ultrasonic signal is denoised to obtain the denoised effective ultrasonic signal.

[0019] The effective ultrasonic signal of the normal area of ​​the bamboo fiber concrete hollow floor slab is used as the reference signal.

[0020] The denoised effective ultrasonic signal is decomposed into multiple signal components;

[0021] Each signal component is compared with the reference signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals.

[0022] In some embodiments, extracting the damage echo features and nonlinear damage features of the damage echo signal specifically includes:

[0023] The damaged echo signal is preprocessed to obtain the damaged echo signal after feature processing.

[0024] The damage echo features and nonlinear damage features of the damage echo signal are extracted from the damage echo signal after feature processing.

[0025] In some embodiments, the damage echo characteristics and the nonlinear damage characteristics are fused based on the historical impact damage characteristics of the historical impact damage data to obtain the damage fusion characteristics of the bamboo fiber concrete hollow floor slab, specifically including:

[0026] Determine the historical impact damage characteristics of the historical impact damage data;

[0027] From the historical impact damage characteristics, similar historical impact damage characteristics that are similar to the damage of the bamboo fiber concrete hollow floor slab are selected.

[0028] Based on the historical similar impact damage characteristics, corresponding fusion weights are assigned to the damage echo characteristics and the nonlinear damage characteristics;

[0029] The damage echo features and the nonlinear damage features are fused according to the respective fusion weights to obtain the damage fusion features of the bamboo fiber concrete hollow floor slab.

[0030] In some embodiments, the impact damage type of the bamboo fiber concrete hollow floor slab is identified based on the historical labeled sample set of composite damage of the bamboo fiber concrete hollow floor slab and the damage fusion features, and the impact damage type result of the bamboo fiber concrete hollow floor slab specifically includes:

[0031] Obtain a historical labeled sample set of composite damage to the bamboo fiber concrete hollow floor slab;

[0032] Determine the similarity data between the historical damage fusion features and the damage fusion features in the historical labeled sample set;

[0033] Determine the similarity threshold for damage fusion features;

[0034] The similarity data are judged one by one by using the similarity threshold to obtain high similarity data;

[0035] Based on the high similarity data, various impact damage types of the bamboo fiber concrete hollow floor slab are extracted from the historical labeled sample set to obtain the impact damage type results of the bamboo fiber concrete hollow floor slab.

[0036] In some embodiments, the historical labeled sample set includes historical damage echo features, nonlinear damage features and auxiliary parameters, and damage type labels.

[0037] Secondly, this application provides an impact damage detection system for bamboo fiber reinforced concrete hollow floor slabs, comprising:

[0038] The acquisition module is used to acquire ultrasonic signals from the impact-damaged area of ​​bamboo fiber concrete hollow floor slabs.

[0039] The processing module is used to suppress interference in the ultrasonic signal based on the hollow information of the bamboo fiber concrete hollow floor slab to obtain an effective ultrasonic signal.

[0040] The processing module is also used to extract the features of the damage signal from the effective ultrasonic signal, obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals, and extract the damage echo features and nonlinear damage features of the damage echo signal.

[0041] The processing module is also used to acquire historical impact damage data of the bamboo fiber concrete hollow floor slab, and to perform feature fusion of the damage echo features and the nonlinear damage features based on the historical impact damage features of the historical impact damage data to obtain the damage fusion features of the bamboo fiber concrete hollow floor slab.

[0042] The execution module is used to identify the impact damage type of the bamboo fiber concrete hollow floor slab based on the historical labeled sample set of composite damage and the damage fusion features, and to obtain the impact damage type result of the bamboo fiber concrete hollow floor slab.

[0043] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0044] The method and system for detecting impact damage to hollow bamboo fiber concrete slabs provided in this application first acquires ultrasonic signals from the impact-damaged area of ​​the hollow bamboo fiber concrete slab; based on the hollow information of the hollow bamboo fiber concrete slab, interference suppression is applied to the ultrasonic signals to obtain effective ultrasonic signals; feature extraction of the damage signal is performed on the effective ultrasonic signals to obtain the damage echo signal of the hollow bamboo fiber concrete slab after removing irrelevant interference signals, and damage echo features and nonlinear damage features of the damage echo signal are extracted; historical impact damage data of the hollow bamboo fiber concrete slab is acquired, and feature fusion of the damage echo features and nonlinear damage features is performed based on the historical impact damage features of the historical impact damage data to obtain the damage fusion features of the hollow bamboo fiber concrete slab; based on the historical labeled sample set of composite damage of the hollow bamboo fiber concrete slab and the damage fusion features, the impact damage type of the hollow bamboo fiber concrete slab is identified to obtain the impact damage type result of the hollow bamboo fiber concrete slab.

[0045] Therefore, in the impact damage detection of bamboo fiber reinforced concrete hollow floor slabs, this application first collects ultrasonic signals from the impact damage area; based on the hollow structure information, it suppresses hollow interference in the ultrasonic signals, effectively extracting the effective signals reflecting the true damage and significantly reducing misjudgments caused by the hollow structure; further, it separates damage echo features and nonlinear damage features from the effective signals, eliminating other irrelevant signal interference and accurately characterizing the damage state; by fusing historical impact damage data features, it enhances the robustness and representativeness of the damage fusion features, reducing identification fluctuations caused by sample differences; finally, it identifies the damage type based on the labeled sample set and damage fusion features, improving classification accuracy and making the detection results more consistent with the actual damage situation, thus achieving reliable and accurate identification of impact damage in bamboo fiber reinforced concrete hollow floor slabs. Using the above scheme, the impact of non-damage interference from the bamboo fiber reinforced concrete hollow structure on the impact damage detection of bamboo fiber reinforced concrete hollow floor slabs can be reduced. Attached Figure Description

[0046] Figure 1 This is an exemplary flowchart of an impact damage detection method for bamboo fiber concrete hollow floor slabs according to some embodiments of this application;

[0047] Figure 2 This is an exemplary flowchart illustrating the determination of damage echo signals according to some embodiments of this application;

[0048] Figure 3 This is an exemplary flowchart illustrating the determination of damage fusion features according to some embodiments of this application;

[0049] Figure 4This is a structural schematic diagram of a bamboo fiber concrete hollow floor slab impact damage detection system according to some embodiments of this application;

[0050] Figure 5 This is a structural schematic diagram of a computer device for implementing an impact damage detection method for bamboo fiber concrete hollow floor slabs, according to some embodiments of this application. Detailed Implementation

[0051] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] refer to Figure 1 The figure is an exemplary flowchart of an impact damage detection method for bamboo fiber reinforced concrete hollow floor slabs according to some embodiments of this application. The impact damage detection method for bamboo fiber reinforced concrete hollow floor slabs mainly includes the following steps:

[0053] In some embodiments, an ultrasonic flaw detector is used to perform ultrasonic testing on the impact-damaged area of ​​a bamboo fiber reinforced concrete hollow floor slab using a single-probe pulse-echo method. Specifically: first, the suspected impact-damaged area of ​​the bamboo fiber reinforced concrete hollow floor slab is cleaned and leveled to remove dust, loose debris, and oil to ensure good coupling between the testing surface and the probe. Then, a single-probe pulse-echo ultrasonic flaw detector suitable for bamboo fiber reinforced concrete is selected. Pre-set testing parameters, such as ultrasonic frequency, transmission power, gain, and sampling rate, are based on the floor slab's design thickness, bamboo fiber content, and hollow structure distribution. After applying a coupling agent, the probe is tightly attached to the damaged area and the surrounding normal surface. The single probe alternately transmits and receives ultrasonic signals, moving at a constant speed along a preset path to perform a full-coverage scan of the damaged area. Real-time acquisition of echo signals formed by reflections from the damaged interface, hollow structure, and concrete matrix as ultrasonic waves propagate within the floor slab is performed, and the probe's detection position and corresponding signal data are recorded simultaneously, completing the ultrasonic testing of the impact-damaged area.

[0054] In step 101, ultrasonic signals are collected from the impact-damaged area of ​​the bamboo fiber concrete hollow floor slab.

[0055] It should be noted that the ultrasonic signal in this application is an electrical signal that is ultimately received and converted by the probe after the ultrasonic flaw detector emits ultrasonic waves into the floor slab and the waves are reflected, refracted, scattered or attenuated when they propagate in the bamboo fiber concrete matrix, the impact damage interface, the cavity wall of the hollow structure and the bamboo fiber-concrete bonding surface. It reflects the bonding state between the bamboo fiber and the concrete matrix inside the floor slab, the location and shape of the hollow structure, and also directly reflects the presence, type and severity of impact damage. The ultrasonic signal includes the reference signal component of ultrasonic waves propagating in the normal concrete matrix, the abnormal reflection / scattering signal component generated through the impact damage area, and the characteristic signal component formed by the interference of the hollow structure on the ultrasonic wave propagation path.

[0056] In step 102, the ultrasonic signal is suppressed by interference based on the hollow information of the bamboo fiber concrete hollow floor slab to obtain an effective ultrasonic signal.

[0057] In some embodiments, the interference suppression of the ultrasonic signal based on the hollow information of the bamboo fiber concrete hollow floor slab to obtain an effective ultrasonic signal can be achieved by the following steps:

[0058] Obtain the hollow information of the bamboo fiber concrete hollow floor slab;

[0059] Spatial coordinate matching is performed between the fasting information and the probe position of the ultrasound detection to establish the correspondence between the signal acquisition position and the fasting structure;

[0060] Based on the correspondence and the emptying information, interference signals from the emptying structure in the ultrasonic signal are removed to obtain an effective ultrasonic signal.

[0061] In practice, the process begins by retrieving the design and construction drawings and as-built data of the bamboo fiber concrete hollow floor slab, and combining this with the radar detection or ultrasonic scanning results of the floor slab in the early stages. This allows for the extraction of precise information about the hollow structure, namely, the hollow information. This hollow information includes the three-dimensional spatial coordinates, chamber dimensions, chamber shape, and distribution spacing of the hollow structure within the bamboo fiber concrete hollow floor slab. The hollow information is a set of data that characterizes the core attributes of the hollow structure inside the bamboo fiber concrete hollow floor slab. It reflects not only the spatial layout, morphological characteristics, and geometric parameter rationality of the hollow structure within the floor slab, but also the relative positional relationship between the hollow structure, the concrete matrix, and the distribution of bamboo fibers.

[0062] In addition, in specific implementation, the spatial coordinate matching of the emptying information and the position of the ultrasonic probe to establish the correspondence between the signal acquisition position and the emptying structure can be achieved in the following way: a displacement sensor or laser positioning device is installed on the ultrasonic probe, and a global coordinate system consistent with the digital model of the emptying structure is established with the lower left corner of the floor slab as the origin. During the detection process, the X, Y, and Z axis coordinate data of the probe in this coordinate system are acquired in real time. Through coordinate translation and rotation calibration operations, the real-time acquisition position of the probe is aligned with the spatial coordinates of the digital model of the emptying structure, thereby establishing a one-to-one correspondence between each ultrasonic signal acquisition position and the emptying structure, that is, clarifying whether the current acquisition position is directly above the emptying structure, around the cavity wall, or in the matrix area far away from the emptying.

[0063] Furthermore, in specific implementation, the interference signals of the hollow structure in the ultrasonic signal are eliminated based on the aforementioned correspondence and the hollow information to obtain the effective ultrasonic signal. This can be achieved in the following way: Based on the established one-to-one correspondence between the signal acquisition location and the hollow structure, firstly, according to the three-dimensional spatial coordinates, chamber size, and distribution range of the hollow structure, determine the influence area of ​​the hollow structure corresponding to each ultrasonic signal acquisition location. Extend the influence range to both sides of the cavity wall as the center, with the ultrasonic wave propagating in the concrete at a distance of 0.5 to 1.0 times the chamber thickness. Based on the ultrasonic wave propagation speed in concrete of approximately 3000 m / s, the influence range is calculated by inversely calculating the propagation time. Based on the spatial span of the affected area, ultrasonic signal segments with acquisition locations within this affected area were screened out and identified as potentially containing empty cavity interference. Subsequently, combining the cavity morphology, cavity wall thickness, and ultrasonic reflection characteristic data of the empty cavity structure obtained from previous experiments in the empty cavity information, typical characteristics of the empty cavity interference signal were extracted: In the time domain, the arrival time of the empty cavity wall reflection signal is usually earlier than that of the damage reflection signal. Because the empty cavity structure is closer to the detection surface, the peak amplitude is in the range of 1.2 to 2.5 times the amplitude of the matrix reflection signal, and the pulse width is 0.8 to 1.0 times that of the matrix signal; In the frequency domain, the center frequency is concentrated in the range of 20 to 50 kHz, and the frequency bandwidth is relatively narrow. Based on the ≤10kHz range, a characteristic template for the hollow interference signal is established. Next, the suspected interference signal segments are segmented, with each segment's length corresponding to the signal length of twice the floor slab thickness during ultrasonic wave propagation in concrete. A template matching method is used to compare each signal segment with the hollow interference characteristic template point-by-point, calculating the correlation coefficient. A correlation coefficient threshold is set, for example, 0.75 is used as the judgment threshold; signal segments with a correlation coefficient ≥0.75 are judged as containing hollow interference. Then, a signal subtraction operation is performed on the judged interference signal segments, scaling the hollow interference characteristic template proportionally, and determining the relationship between the current signal segment and the template. The amplitude difference scaling factor is adjusted to make the template amplitude consistent with the amplitude of the interference component in the signal. The scaled template signal is subtracted from the original signal segment to achieve preliminary separation of the empty interference component. Finally, the subtracted signal segment is subjected to moving average filtering with a filter window length of 5 to 10 sampling points, determined according to the signal sampling rate. When the sampling rate is 10MHz, 8 sampling points are taken to remove the weak interference noise left by the subtraction operation. At the same time, the signal segments that are not determined to contain interference are directly retained. All processed signal segments are spliced ​​and integrated in the order of acquisition time to obtain an effective ultrasonic signal that only reflects the information related to the bamboo fiber concrete matrix and impact damage.

[0064] It should be noted that the correspondence in this application represents a one-to-one mapping relationship between the three-dimensional spatial coordinates of the probe's real-time acquisition position and the three-dimensional coordinates of each hollow structure established after alignment and calibration during ultrasonic testing of bamboo fiber concrete hollow floor slabs. This reflects whether the current ultrasonic signal acquisition point is within the influence range of the hollow structure and its relative spatial position. The effective ultrasonic signal represents the electrical signal obtained after removing the reflection and diffraction interference signals generated by the hollow structure from the original ultrasonic signal. This reflects the true propagation characteristics of ultrasonic waves when they propagate inside the bamboo fiber concrete matrix and are reflected, scattered, or attenuated at the impact damage interface, as well as the bonding state between the bamboo fiber and the concrete matrix.

[0065] In step 103, the features of the damage signal are extracted from the effective ultrasonic signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals, and the damage echo features and nonlinear damage features of the damage echo signal are extracted.

[0066] In some embodiments, reference Figure 2 As shown in the figure, this is an exemplary flowchart for determining the damage echo signal in some embodiments of this application. In this embodiment, the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals by extracting the features of the damage signal from the effective ultrasonic signal can be achieved by the following steps:

[0067] In step 1031, the effective ultrasonic signal is denoised to obtain a denoised effective ultrasonic signal.

[0068] In step 1032, the effective ultrasonic signal of the normal area of ​​the bamboo fiber concrete hollow floor slab is used as a reference signal.

[0069] In step 1033, the denoised effective ultrasonic signal is decomposed into multiple signal components;

[0070] In step 1034, each signal component is compared with the reference signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals.

[0071] In specific implementation, the effective ultrasonic signal is denoised to obtain the denoised effective ultrasonic signal. This can be achieved as follows: the effective ultrasonic signal after fasting interference suppression is processed using a wavelet threshold denoising method. The db4 wavelet basis is selected as the decomposition basis function. A 5-layer wavelet decomposition is determined based on the signal sampling rate. An adaptive threshold for the high-frequency coefficients of each layer is calculated. The threshold calculation formula is: threshold = ,in The noise standard deviation is obtained by dividing the absolute value of the median of the high-frequency coefficients in the first layer by 0.6745. Given the number of signal sampling points, soft thresholding is applied to the high-frequency coefficients of each layer, i.e., coefficients greater than the threshold are subtracted from the threshold, and coefficients less than the threshold are set to 0. Then, the processed high-frequency coefficients and low-frequency coefficients are subjected to inverse wavelet transform to obtain the denoised effective ultrasonic signal after removing electronic noise and environmental vibration interference. Other processing methods can be used in other embodiments, which are not limited here.

[0072] In addition, in specific implementation, the effective ultrasonic signal of the normal area of ​​the hollow bamboo fiber concrete floor slab can be used as the reference signal in the following way: Select a normal substrate area of ​​the hollow bamboo fiber concrete floor slab that is far away from the suspected impact damage area and has been confirmed to be undamaged by previous visual inspection and radar detection. The area should be no less than 5cm×5cm. Select 5 sampling points evenly in this area. Use the same ultrasonic detection parameters as the damaged area to collect the effective ultrasonic signal of each point. The signals of these 5 points are superimposed and averaged by the arithmetic mean method to eliminate the random error of a single sampling point and obtain a stable reference signal. Other methods can be used in other embodiments, which are not limited here.

[0073] In addition, in specific implementation, the effective ultrasonic signal can be decomposed into multiple signal components in the following way: The denoised effective ultrasonic signal is decomposed using the empirical mode decomposition method. First, all local maxima and local minima in the signal are identified. The upper and lower envelopes of the signal are obtained by fitting them using cubic spline interpolation. The average value of the upper and lower envelopes is calculated and denoted as m1. The denoised effective ultrasonic signal is subtracted from m1 to obtain the first candidate intrinsic mode function component h1. The above envelope fitting and mean subtraction steps are repeated for h1 until h1 meets the determination criteria for intrinsic mode functions. That is, at the same moment, the number of local maxima and local minima of the signal are equal or differ by no more than 1, and the mean of the upper and lower envelopes is close to 0. At this time, h1 is the first intrinsic mode function component. Subtract h1 from the denoised effective ultrasonic signal to obtain the remaining signal r1. Repeat the above decomposition process for r1 to obtain multiple intrinsic mode function components such as h2, h3, ..., hn, and the final residual component rn. The residual component is a monotonic function or a constant, thus completing the multi-component decomposition of the signal. Each intrinsic mode function component is used as a signal component. In other embodiments, other methods can be used for processing, which are not limited here.

[0074] In addition, in specific implementation, the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals can be obtained by comparing each signal component with the reference signal in the following way: The time-domain characteristic difference and frequency-domain characteristic difference between each signal component and the reference signal are calculated respectively. The time-domain characteristic difference includes the amplitude peak difference, the reflected wave arrival time difference, and the pulse width difference. The frequency-domain characteristic difference includes the center frequency difference and the amplitude spectrum peak difference. Simultaneously, the Pearson correlation coefficient between each component and the reference signal is calculated, and a correlation coefficient threshold is set, for example, 0.3 is used as the judgment threshold. A correlation coefficient lower than 0.3 indicates that the component differs significantly from the normal matrix signal and is likely a damage-related signal. Signal components with correlation coefficients lower than the threshold and time-domain and frequency-domain characteristic differences exceeding a preset range are selected. These selected components are superimposed and reconstructed to obtain a damage echo signal that reflects only the reflection and scattering characteristics of the impact damage interface after removing irrelevant interference signals such as the normal matrix signal and non-damaging reflections of bamboo fiber. Other processing methods can be used in other embodiments, which are not limited here.

[0075] It should be noted that the reference signal in this application represents the stable ultrasonic signal in the normal matrix area of ​​the bamboo fiber concrete hollow floor slab, reflecting the inherent time and frequency domain characteristics of ultrasonic waves when they propagate normally in the undamaged bamboo fiber and concrete matrix; the signal components represent the components obtained by decomposing the effective ultrasonic signal, each component corresponding to a specific frequency range and physical meaning; the damage echo signal reflects the reflection, scattering and attenuation characteristics of ultrasonic waves at the impact damage interface, and can be used to provide a precise signal carrier for extracting damage features.

[0076] In some embodiments, the extraction of damage echo features and nonlinear damage features from the damage echo signal can be achieved using the following steps:

[0077] The damaged echo signal is preprocessed to obtain the damaged echo signal after feature processing.

[0078] The damage echo features and nonlinear damage features of the damage echo signal are extracted from the damage echo signal after feature processing.

[0079] In specific implementation, the damaged echo signal is preprocessed to obtain the feature-processed damaged echo signal. This can be achieved in the following way: First, cross-correlation is used for time-domain alignment, taking the moment when the amplitude of the signal reaches 5% of its peak value as the starting reference point, aligning the damaged echo signal with the starting time of the reference signal to eliminate the time offset caused by probe displacement deviation during detection. Then, a Hanning window is applied to the time-domain aligned signal to reduce spectral leakage, and a fast Fourier transform is used to convert the time-domain signal into a frequency-domain signal to obtain the amplitude spectrum and phase spectrum of the signal. Finally, a 5-point moving average filter is used to smooth the frequency-domain signal to enhance the discriminability of the feature parameters, thus obtaining the feature-processed damaged echo signal. Other processing methods can also be used in other embodiments, which are not limited here.

[0080] In addition, in specific implementation, the extraction of damage echo features and nonlinear damage features from the damaged echo signal after feature processing can be achieved in the following way: When extracting damage echo features, in the time domain, the start and end times of the echo pulse are identified by setting an amplitude threshold of 10% of the maximum amplitude of the signal, and the peak amplitude, echo arrival time, pulse width, rise slope, and amplitude decay rate are calculated. Here, the peak amplitude is the maximum amplitude between the start and end times, the echo arrival time is the interval from the start of the signal to the peak time, the pulse width is the time difference between the start and end times, the rise slope is the ratio of the amplitude change from 10% to 90% to the time change, and the amplitude decay rate is... The rate represents the amplitude decrease every 10 μs after the peak. In the frequency domain, the center frequency, peak frequency, frequency bandwidth, and harmonic component proportion are determined by calculating the power spectral density of the amplitude spectrum. The center frequency is the frequency corresponding to the centroid of the power spectral density integral, the peak frequency is the frequency corresponding to the maximum amplitude in the amplitude spectrum, the frequency bandwidth is the frequency difference between the upper and lower frequencies corresponding to the half-power point, and the harmonic component proportion is the amplitude ratio of the second harmonic to the fundamental frequency. The values ​​obtained in the time and frequency domains are used as the damage echo characteristics of the damaged echo signal. When extracting nonlinear damage characteristics, harmonic-related nonlinear characteristics are extracted based on the amplitude spectrum of the damaged echo signal after feature processing: first, the core frequency component with the most concentrated energy in the amplitude spectrum, the fundamental frequency, is identified and its amplitude is calculated. Then locate the amplitudes corresponding to the second harmonic and the third harmonic at twice the fundamental frequency. , Through formula Calculate the total harmonic distortion rate Simultaneously, the second harmonic distortion rate was calculated separately. Third harmonic distortion rate The degree of nonlinear distortion in ultrasonic wave propagation caused by damage is quantified; subsequently, a Hilbert transform is performed on the time-domain signal to obtain an analytical signal, from which the instantaneous phase is separated. The linear component in the instantaneous phase is fitted using the least squares method. ,pass Obtaining nonlinear phase deviation The peak value, root mean square value, and kurtosis of the deviation are further extracted to reflect the nonlinear disturbance caused by damage during phase propagation. Next, the signal delay time τ is determined using the mutual information method, taking the time interval corresponding to the minimum mutual information value, typically 1-5 sampling points. The phase space embedding dimension m is determined using the false nearest neighbor method, generally taking a value of 2-10. Based on τ and m, the m-dimensional phase space of the signal is reconstructed, and the average separation rate of adjacent trajectories in the phase space is calculated to obtain the maximum Lyapunov exponent (L). This exponent is positive when nonlinear damage exists, and a larger value indicates stronger nonlinearity caused by the damage. Finally, the box-counting method is used to extract fractal features: the reconstructed phase space is divided into different side lengths... ( Divide the signal amplitude range (from 0.01 to 0.1 times) into cubic boxes, and calculate the minimum number of boxes required to cover the signal trajectory. In a double logarithmic coordinate system and Linear fitting is performed, and the fitting slope is used as the fractal dimension to quantify the changes in the complexity of the signal time series caused by the damage, thereby completing the comprehensive extraction of nonlinear damage features. Other extraction methods can also be used in other embodiments, which are not limited here.

[0081] It should be noted that the damage echo characteristics in this application represent the set of key time-domain and frequency-domain parameters of the damage state of bamboo fiber reinforced concrete hollow floor slabs. They reflect the signal response characteristics of ultrasonic waves when they are reflected and scattered at the impact damage interface, and can intuitively reflect the presence, morphological characteristics, and surface roughness of the damage, while indirectly reflecting the magnitude of the damage. The nonlinear damage characteristics represent the set of parameters extracted from the damage echo signal that can quantify the nonlinear propagation characteristics of ultrasonic waves. They reflect the waveform distortion, phase disturbance, and complex changes in dynamic behavior of ultrasonic waves during propagation due to the nonlinear effect of the damage interface when impact damage occurs in bamboo fiber reinforced concrete. They also indirectly reflect the nonlinear nature and severity of the damage. The nonlinear damage characteristics include harmonic correlation characteristics, phase nonlinear characteristics, dynamic nonlinear characteristics, and fractal characteristics, comprehensively covering the nonlinear performance of the signal in the frequency domain, time domain phase, dynamic behavior, and spatial distribution.

[0082] In step 104, historical impact damage data of the bamboo fiber concrete hollow floor slab is obtained, and the damage echo characteristics and the nonlinear damage characteristics are fused based on the historical impact damage characteristics of the historical impact damage data to obtain the damage fusion characteristics of the bamboo fiber concrete hollow floor slab.

[0083] It should be noted that the historical impact damage data in this application refers to the structured / unstructured data related to impact damage accumulated in the past use, simulated impact tests, or preliminary inspections of bamboo fiber concrete hollow floor slabs. It reflects the damage generation mechanism and development law of this type of floor slab under different impact energies, impact angles, service environments, and service years, as well as the influence relationship of parameters such as bamboo fiber content, hollow structure design, and concrete strength on damage characteristics. The historical impact damage data includes the original ultrasonic detection signals and extracted historical damage echo characteristics and nonlinear damage characteristics of bamboo fiber concrete hollow floor slabs in corresponding scenarios, and also covers the content of impact condition parameters, floor slab foundation parameters, damage labeling information, and service environment data.

[0084] In some embodiments, reference Figure 3 As shown in the figure, this is an exemplary flowchart for determining damage fusion features in some embodiments of this application. In this embodiment, the damage echo features and the nonlinear damage features are fused based on the historical impact damage features of the historical impact damage data to obtain the damage fusion features of the bamboo fiber concrete hollow floor slab. This can be achieved by the following steps:

[0085] In step 1041, the historical impact damage characteristics of the historical impact damage data are determined;

[0086] In step 1042, similar historical impact damage features that are similar to the damage of the bamboo fiber concrete hollow floor slab are selected from the historical impact damage features.

[0087] In step 1043, corresponding fusion weights are assigned to the damage echo features and the nonlinear damage features based on the historical similar impact damage features;

[0088] In step 1044, the damage echo features and the nonlinear damage features are fused according to each fusion weight to obtain the damage fusion features of the bamboo fiber concrete hollow floor slab.

[0089] In specific implementation, the historical impact damage characteristics of the historical impact damage data can be determined in the following way: the historical impact damage data of bamboo fiber concrete hollow floor slabs is preprocessed with standardization, and the historical damage echo characteristics and historical nonlinear damage characteristics in the historical impact damage data are transformed to the [0,1] interval using the min-max standardization method to eliminate the differences in feature scale caused by different detection conditions and floor slab parameters. At the same time, the floor slab foundation parameters and impact condition parameters in the historical impact damage data are extracted as auxiliary features to jointly constitute complete historical impact damage characteristics. Other methods can also be used to determine the characteristics in other embodiments, which are not limited here.

[0090] In addition, in specific implementation, the historical similar impact damage features that are similar to the damage of the bamboo fiber concrete hollow floor slab can be screened from the historical impact damage features in the following way: the Euclidean distance method is used to screen historical similar impact damage features. First, the foundation parameters and impact condition parameters of the currently detected bamboo fiber concrete hollow floor slab are matched with the corresponding auxiliary features in the historical impact damage features. The subset of historical features with a matching degree of foundation parameters and condition parameters ≥ 80% is retained. Then, the Euclidean distance between the currently extracted damage echo features, nonlinear damage features and each historical feature in the subset is calculated. A distance threshold is set, for example, 0.3 is taken as the threshold. The smaller the distance, the higher the feature similarity. Historical features with an Euclidean distance ≤ 0.3 are screened, which are historical similar impact damage features that are similar to the current floor slab damage. Other methods can be used for screening in other embodiments, which are not limited here.

[0091] In addition, in specific implementation, the allocation of corresponding fusion weights to the damage echo features and the nonlinear damage features based on the historical similar impact damage features can be achieved in the following way: First, split the historical similar features into a subset of damage echo features and a subset of nonlinear damage features, and calculate the normalized frequency pi of each feature parameter in the historical similar samples, where the normalized frequency pi = the number of times the parameter value occurs / the total number of samples; Second, calculate the information entropy of each parameter in the two subsets according to the information entropy formula. Then, using the weighting formula, single-parameter weight = ,in The first step is to calculate the initial weight of each parameter as the sum of all parameters (1-H) within the subset. Finally, the average of the initial weights of all parameters within the same subset is taken to obtain the basic weights of the damage echo feature and the nonlinear damage feature in the historical similar features. The second step is to calculate the reciprocal of the Euclidean distance between the current feature and each group of historical similar features. When the distance is 0, it is taken as 1. The average of all reciprocals is taken as the similarity correction coefficient λ. The larger λ is, the higher the fit between the current feature and the historical similar features. The value range is 1~3, and the basic weights are corrected. The third step is to normalize the corrected weights to finally determine the fusion weights of the current damage echo feature and the nonlinear damage feature. Other allocation methods can also be used in other embodiments, which are not limited here.

[0092] In addition, in specific implementation, the damage echo features and the nonlinear damage features are fused according to each fusion weight to obtain the damage fusion features of the bamboo fiber concrete hollow floor slab. This can be achieved in the following way: a weighted summation method is used for feature fusion, and the damage echo features and nonlinear damage features are standardized. Specifically, for each damage echo feature parameter and nonlinear damage feature parameter extracted by the current detection, the min-max standardization formula is substituted into each feature parameter to calculate them one by one. The standardized damage echo features (x1, x2, ..., xn) are multiplied by the corresponding fusion weights, and the standardized nonlinear damage features (y1, y2, ..., ym) are multiplied by the corresponding fusion weights. Then, the two weighted feature vectors are concatenated and integrated according to their dimensions to obtain a damage fusion feature vector with dimension n+m. This vector is the damage fusion feature of the bamboo fiber concrete hollow floor slab. Other methods can also be used in other embodiments, which are not limited here.

[0093] It should be noted that the historical impact damage features in this application represent a set of features extracted from historical impact damage data accumulated from past inspections and tests of bamboo fiber concrete hollow floor slabs. These features reflect the characteristic patterns of impact damage under different scenarios and the influence of floor slab parameters and working conditions on damage features. The historical similar impact damage features represent a subset with high matching degree with the basic parameters and impact conditions of the currently inspected floor slab and significant feature similarity. These features reflect past damage experience with the same or similar formation conditions and characteristic manifestations as the current damage. The fusion weight represents the contribution of the currently detected damage echo features and nonlinear damage features to damage identification, ensuring that the fusion process highlights the role of key features. The damage fusion features represent the fusion features of the real-time signal features of the current damage in the bamboo fiber concrete hollow floor slab, combining real-time performance with empirical reference, providing comprehensive and effective feature support for the accurate identification of impact damage types.

[0094] In step 105, the impact damage type of the bamboo fiber concrete hollow floor slab is identified based on the historical labeled sample set of composite damage of the bamboo fiber concrete hollow floor slab and the damage fusion features, so as to obtain the impact damage type result of the bamboo fiber concrete hollow floor slab.

[0095] In some embodiments, the impact damage type of the bamboo fiber concrete hollow floor slab is identified based on the historical labeled sample set of composite damage and the damage fusion features. The impact damage type result of the bamboo fiber concrete hollow floor slab can be obtained by the following steps:

[0096] Obtain a historical labeled sample set of composite damage to the bamboo fiber concrete hollow floor slab;

[0097] Determine the similarity data between the historical damage fusion features and the damage fusion features in the historical labeled sample set;

[0098] Determine the similarity threshold for damage fusion features;

[0099] The similarity data are judged one by one by using the similarity threshold to obtain high similarity data;

[0100] Based on the high similarity data, various impact damage types of the bamboo fiber concrete hollow floor slab are extracted from the historical labeled sample set to obtain the impact damage type results of the bamboo fiber concrete hollow floor slab.

[0101] It should be noted that the historical labeled sample set in this application refers to a standardized sample set containing damage features and corresponding damage type labels, constructed from the dataset accumulated from past impact damage detection and simulation tests of bamboo fiber concrete hollow floor slabs. This reflects the fixed correspondence between different impact damage types and their damage fusion features, and also demonstrates the influence of floor slab foundation parameters and impact condition parameters on the matching relationship between damage types and features. The historical labeled sample set includes more than 300 sets of samples covering various typical impact damage scenarios. Each set of samples consists of two parts: first, a historical damage fusion feature vector processed by min-max standardization, integrating historical damage echo features, nonlinear damage features, and auxiliary parameters; second, damage type labels jointly confirmed by ultrasonic imaging results and core sampling analysis. In addition, it also includes a training set and a validation set divided in a 7:3 ratio, as well as valid data records after removing abnormal samples using the 3σ criterion.

[0102] In addition, in specific implementation, the similarity data between the historical damage fusion features and the damage fusion features in the historical labeled sample set can be determined in the following way: the similarity data between the historical damage fusion features and the current damage fusion features is determined by using the cosine similarity algorithm, and the similarity value is calculated by substituting the historical samples in the training set one by one to obtain the similarity data between the historical damage fusion features and the damage fusion features in the historical labeled sample set. Other methods can also be used to determine this in other embodiments, which are not limited here.

[0103] In addition, in specific implementation, the similarity threshold of damage fusion features can be determined in the following way: based on the training set samples, with the standard of similar damage recognition accuracy ≥95% and different damage misclassification rate ≤3%, the candidate threshold range of 0.6~0.9 is used for verification one by one. By statistically analyzing the recognition results of the training set under different thresholds, when the threshold is 0.78, the similar damage recognition accuracy of the training set reaches 96.2% and the different damage misclassification rate is 2.1%. The threshold is then repeatedly verified with the validation set. After confirming that it meets the standard, 0.78 is determined as the similarity threshold. Other methods can also be used to determine it in other embodiments, which are not limited here.

[0104] In addition, the similarity data can be judged one by one by the similarity threshold to obtain high similarity data. This can be achieved by judging the similarity data one by one by the similarity threshold, judging the similarity data corresponding to the historical samples with a similarity value ≥ 0.78 as high similarity data, and filtering out all historical samples corresponding to high similarity data. Other methods can also be used in other embodiments, which are not limited here.

[0105] Furthermore, in specific implementation, the impact damage types of the bamboo fiber concrete hollow floor slab are extracted from the historical labeled sample set based on the high similarity data. The impact damage type results for the bamboo fiber concrete hollow floor slab can be obtained in the following way: Extract all historical samples corresponding to the high similarity data from the historical labeled sample set; read the impact damage type labels labeled in each group of samples one by one, such as: single crack, single looseness, crack-loose composite, spalling-crack composite, etc., and construct a label statistical list; subsequently, use a one-by-one counting method to perform frequency statistics on the various damage type labels in the label statistical list. Record the frequency of each label and sort them from highest to lowest frequency. Select the label with the highest frequency as the initial candidate for identification. If two or more labels have the same frequency (tied for the highest frequency), a secondary verification mechanism is introduced: the damage location information of the currently detected bamboo fiber concrete hollow floor slab is compared with the typical feature parameter range corresponding to the tying labels. Finally, the final damage type label is determined based on the secondary verification results. If the feature matching degree of a certain tying label is higher than that of other labels (matching degree ≥ 85%), it is taken as the final identification result, and the identification criteria, such as label frequency and damage, are recorded. The location and feature parameter matching results form a clear and unique impact damage type result for bamboo fiber reinforced concrete hollow floor slabs. The 85% matching threshold is determined based on a historical labeled sample set of composite damage to bamboo fiber reinforced concrete hollow floor slabs through statistical verification and threshold optimization. First, sufficient samples covering all impact damage types are selected from the historical labeled sample set and divided into training and validation sets in a 7:3 ratio. Using a method consistent with current feature comparison, the feature matching degree between similar damage samples and between dissimilar damage samples in the training set is calculated, obtaining the statistical distribution of the two matching degrees. Then, within the 70%-90% range... Within the 0% candidate threshold range, values ​​are successively substituted, and the accuracy rate of identifying similar damage and the misclassification rate of different damage corresponding to each candidate threshold are statistically analyzed. When the threshold is set to 85%, the accuracy rate of identifying similar damage in the training set can reach over 95%, and the misclassification rate of different damage is controlled within 5%. After cross-validation on the validation set, the threshold can still maintain stable recognition performance, which avoids the increase in misclassification caused by too low a threshold and prevents the missed recognition caused by too high a threshold. Finally, through multiple rounds of statistical verification, 85% is determined to be the critical matching degree threshold that balances recognition accuracy and generalization ability. Other methods can be used to extract values ​​in other embodiments, which are not limited here.

[0106] It should be noted that the similarity data in this application represents a quantitative value of the similarity between the currently detected damage fusion feature and each historical damage fusion feature in the historical labeled sample set, reflecting the degree of fit between the current damage feature and the historical damage feature; the similarity threshold reflects the threshold of the similarity judgment standard for distinguishing whether the current damage and the historical sample damage have reference value, ensuring that only historical samples with significant correlation are selected; high similarity data reflects the high matching relationship between the current damage feature and the corresponding historical sample feature; the impact damage type result reflects the result of the actual impact damage category currently existing in the bamboo fiber concrete hollow floor slab.

[0107] In another aspect, in some embodiments, this application provides an impact damage detection system for bamboo fiber reinforced concrete hollow floor slabs, referencing... Figure 4 The figure is a schematic diagram of the impact damage detection system for bamboo fiber reinforced concrete hollow floor slabs according to some embodiments of this application. The bamboo fiber reinforced concrete hollow floor slab impact damage detection system 400 includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below:

[0108] Acquisition module 401, in this application, is mainly used to acquire ultrasonic signals from the impact-damaged area of ​​bamboo fiber concrete hollow floor slab;

[0109] Processing module 402, in this application, is used to suppress the interference of the ultrasonic signal based on the hollow information of the bamboo fiber concrete hollow floor slab, so as to obtain an effective ultrasonic signal.

[0110] It should be noted that the processing module 402 in this application is also used to extract the features of the damage signal from the effective ultrasonic signal, to obtain the damage echo signal of the bamboo fiber concrete hollow floor slab after removing irrelevant interference signals, and to extract the damage echo features and nonlinear damage features of the damage echo signal.

[0111] Additionally, it should be noted that the processing module 402 in this application is also used to obtain historical impact damage data of the bamboo fiber concrete hollow floor slab, and to perform feature fusion on the damage echo features and the nonlinear damage features based on the historical impact damage features of the historical impact damage data to obtain the damage fusion features of the bamboo fiber concrete hollow floor slab.

[0112] The execution module 403 in this application is mainly used to identify the impact damage type of the bamboo fiber concrete hollow floor slab based on the historical labeled sample set of composite damage of the bamboo fiber concrete hollow floor slab and the damage fusion features, so as to obtain the impact damage type result of the bamboo fiber concrete hollow floor slab.

[0113] In addition, this application also provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described method for detecting impact damage in hollow bamboo fiber concrete slabs.

[0114] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing an impact damage detection method for hollow bamboo fiber concrete slabs according to some embodiments of this application. The impact damage detection method for hollow bamboo fiber concrete slabs in the above embodiments can be implemented through... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0115] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0116] The communication bus 502 can be used to transmit information between the aforementioned components.

[0117] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0118] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0119] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0120] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0121] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0122] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for detecting impact damage in hollow bamboo fiber concrete floor slabs.

[0123] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0124] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for impact damage detection of a bamboo fiber concrete open-web floor, wherein, The method for detecting the impact damage area of the bamboo fiber concrete open-web floor by using the ultrasonic flaw detector with single probe pulse echo method, characterized in that the method comprises the following steps: Collecting the ultrasonic signal of the impact damage area of the bamboo fiber concrete open-web floor; Based on the open-web information of the bamboo fiber concrete open-web floor, the ultrasonic signal is subjected to interference suppression to obtain an effective ultrasonic signal; The effective ultrasonic signal is subjected to feature extraction of the damage signal to obtain the damage echo signal of the bamboo fiber concrete open-web floor after removing the irrelevant interference signal, and the damage echo feature and the nonlinear damage feature of the damage echo signal are extracted; The historical impact damage data of the bamboo fiber concrete open-web floor is acquired, the damage echo feature and the nonlinear damage feature are subjected to feature fusion according to the historical impact damage feature of the historical impact damage data, and the damage fusion feature of the bamboo fiber concrete open-web floor is obtained; Based on the historical labeled sample set of the composite damage of the bamboo fiber concrete open-web floor and the damage fusion feature, the impact damage type of the bamboo fiber concrete open-web floor is identified to obtain the impact damage type result of the bamboo fiber concrete open-web floor; The damage echo feature and the nonlinear damage feature of the damage echo signal are extracted, which specifically comprises: The damage echo signal is subjected to signal feature preprocessing to obtain a feature-processed damage echo signal; The damage echo feature and the nonlinear damage feature of the damage echo signal are extracted from the feature-processed damage echo signal; The damage fusion feature of the bamboo fiber concrete open-web floor is obtained by fusing the damage echo feature and the nonlinear damage feature according to the historical impact damage feature of the historical impact damage data, which specifically comprises: The historical impact damage feature of the historical impact damage data is determined; The historical similar impact damage feature similar to the damage of the bamboo fiber concrete open-web floor is screened from the historical impact damage feature; The corresponding fusion weight is assigned to the damage echo feature and the nonlinear damage feature based on the historical similar impact damage feature; The damage echo feature and the nonlinear damage feature are fused according to the respective fusion weights to obtain the damage fusion feature of the bamboo fiber concrete open-web floor.

2. The method of claim 1, wherein, The method for detecting the impact damage area of the bamboo fiber concrete open-web floor by using the ultrasonic flaw detector with single probe pulse echo method further comprises the following steps: after the probe is smeared with a coupling agent, the probe is tightly attached to the surface of the damage area and the surrounding normal area; the ultrasonic signal transmission and echo reception are alternately completed by the single probe; the probe is uniformly moved along the damage area at a preset path to perform full coverage scanning; the echo signal formed by the reflection of the ultrasonic wave on the damage interface, the open-web structure and the concrete matrix during the propagation of the ultrasonic wave in the floor is collected in real time.

3. The method of claim 1, wherein, The ultrasonic signal comprises a reference signal component of the ultrasonic wave propagating in the normal concrete matrix, an abnormal reflection / scattering signal component generated by the impact damage area, and a characteristic signal component formed by the interference of the open-web structure to the ultrasonic wave propagation path.

4. The method of claim 1, wherein, The interference suppression is performed on the ultrasonic signal based on the hollow information of the bamboo fiber concrete hollow floor, and effective ultrasonic signal is obtained, specifically including: Obtain the hollow information of the bamboo fiber concrete hollow floor; Match the hollow information and the probe position of the ultrasonic detection in spatial coordinates to establish the corresponding relationship between the signal collection position and the hollow structure; Based on the corresponding relationship and the hollow information, the interference signal of the hollow structure in the ultrasonic signal is removed to obtain the effective ultrasonic signal.

5. The method of claim 1, wherein, The feature extraction of damage signal is performed on the effective ultrasonic signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor after removing irrelevant interference signals, specifically including: The effective ultrasonic signal is denoised to obtain the denoised effective ultrasonic signal; The effective ultrasonic signal of the normal area of the bamboo fiber concrete hollow floor is taken as a reference signal; The denoised effective ultrasonic signal is decomposed into multiple signal components; Each signal component is compared with the reference signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor after removing irrelevant interference signals.

6. The method of claim 1, wherein, Based on the historical labeled sample set of the composite damage of the bamboo fiber concrete hollow floor and the damage fusion feature, the impact damage type of the bamboo fiber concrete hollow floor is identified to obtain the impact damage type result of the bamboo fiber concrete hollow floor, specifically including: Obtain the historical labeled sample set of the composite damage of the bamboo fiber concrete hollow floor; Determine the similarity data between the historical damage fusion feature in the historical labeled sample set and the damage fusion feature; Determine the similarity threshold of the damage fusion feature; Judge the similarity data one by one through the similarity threshold to obtain high similarity data; Based on the high similarity data, each impact damage type of the bamboo fiber concrete hollow floor is extracted from the historical labeled sample set to obtain the impact damage type result of the bamboo fiber concrete hollow floor.

7. The method of claim 1, wherein, The historical labeled sample set includes historical damage echo features, nonlinear damage features, auxiliary parameters, and damage type labels.

8. A system for impact damage detection of a bamboo fiber concrete flat slab, which is characterized by, performing impact damage detection of a bamboo fiber concrete flat slab using the method according to any one of claims 1 to 7. The system includes: The acquisition module is used to acquire the ultrasonic signal of the impact damage area of the bamboo fiber concrete hollow floor; The processing module is used to suppress the interference of the ultrasonic signal based on the hollow information of the bamboo fiber concrete hollow floor to obtain the effective ultrasonic signal; The processing module is also used to extract the features of the damage signal from the effective ultrasonic signal to obtain the damage echo signal of the bamboo fiber concrete hollow floor after removing irrelevant interference signals, and extract the damage echo features and nonlinear damage features of the damage echo signal; The processing module is also used to obtain the historical impact damage data of the bamboo fiber concrete hollow floor, and perform feature fusion on the damage echo features and the nonlinear damage features according to the historical impact damage features of the historical impact damage data to obtain the damage fusion features of the bamboo fiber concrete hollow floor; The execution module is configured to identify the impact damage type of the bamboo fiber concrete open-web floor based on the historical labeled sample set of the composite damage of the bamboo fiber concrete open-web floor and the damage fusion feature, and obtain an impact damage type result of the bamboo fiber concrete open-web floor.

Images