Method and system for mass detection inside slab track

Abstract

The application discloses a mass detection method and system for the internal part of a slab track, and relates to the field of railway track engineering. The method comprises the following steps: constructing a multi-point excitation and multi-source sensing cooperative collection system in a target track section, applying a preset excitation load to the track slab, and synchronously collecting structural response data; combining the layered structure characteristics of the track, establishing an interlayer response transmission relationship, and mapping the response data into response characteristic parameters; performing disturbance sensitive feature extraction to obtain multi-dimensional response features; calling reference response features to compare and analyze the multi-dimensional response features to generate difference response characteristics; inverting and positioning the abnormal areas in the track structure to obtain defect position and degree data, and outputting the results. The application solves the problems of weak deep defect identification ability, inability to continuously quantitatively evaluate and easy to cause structural damage in the existing track internal quality detection, and achieves the effects of accurately identifying deep hidden defects, realizing global continuous quantitative detection and full-process non-damage.

Description

Technical Field

[0001] This application relates to the field of railway track engineering, and in particular to a quality inspection method and system for the interior of slab track. Background Technology

[0002] The detection of hidden defects inside slab track is a crucial step in ensuring the long-term operational safety and structural stability of rail transit lines, directly impacting train ride smoothness and track maintenance management. Currently, the industry commonly uses traditional methods such as manual visual inspection, borehole sampling, and conventional ultrasonic single-point flaw detection and impact echo simple testing to screen for internal track quality. Existing conventional detection methods are easily affected by the on-site environment, lack sufficient accuracy in identifying deep and subtle defects, and also have practical shortcomings such as limited detection coverage, difficulty in achieving continuous quantitative assessment of the entire track area.

[0003] At present, the detection of internal quality defects in slab track has technical problems such as weak ability to identify deep defects, inability to continuously quantify and evaluate defects, and easy to cause structural damage. Summary of the Invention

[0004] This application provides a quality inspection method and system for the interior of slab track. It employs a multi-point excitation and multi-source sensing collaborative acquisition system built in the test section of the slab track. A preset excitation load is applied, and vibration and wave propagation structural response data are collected simultaneously. The interlayer response transmission relationship is established based on the track's layered structure. The raw response data is transformed into quantitative parameters characterizing the interlayer connection state. Multi-dimensional sensitive features reflecting local stiffness, damping, and interface contact changes are extracted from these parameters. Reference features of a defect-free, healthy track are retrieved for comparison, generating a difference characterization of structural state deviation. Based on the difference characterization and the interlayer response transmission law, abnormal areas are located, defect locations and severity are obtained, and inspection results are output. These technical means solve the technical problems of weak deep defect identification capability, inability to continuously quantify and evaluate defects, and easy structural damage in existing slab track internal quality defect detection methods. This achieves the technical effects of accurately identifying deep hidden defects, realizing continuous quantitative detection across the entire area, and ensuring the integrity of the track structure without damage throughout the entire process.

[0005] This application provides a quality inspection method for the interior of slab track, comprising: constructing a multi-point excitation and multi-source sensing collaborative acquisition system within a target slab track section; applying a preset excitation load to the track slab and simultaneously acquiring corresponding structural response data, wherein the structural response data includes vibration response signals and wave propagation response signals; based on the structural response data and combined with the layered structural characteristics of the slab track, establishing the interlayer response transmission relationship under excitation, and mapping the structural response data into response characterization parameters that characterize changes in the interlayer connection state; performing disturbance-sensitive feature extraction on the response characterization parameters to obtain multidimensional response features that characterize changes in local structural stiffness, damping, and interface contact state; calling a pre-constructed reference response feature of a defect-free state, comparing and analyzing the multidimensional response features to generate a difference response characterization reflecting the degree of deviation of the current structural state; inverting and locating abnormal areas inside the track structure according to the difference response characterization and the interlayer response transmission relationship to obtain corresponding defect location and defect severity data, and outputting quality inspection results.

[0006] In a possible implementation, the abnormal region inside the track structure is inverted and located based on the differential response characterization and inter-layer response transmission relationship, and the following processing is performed: the differential response characterization is reconstructed according to the spatial correspondence between the excitation location and the response acquisition location to form a differential response spatial mapping field distributed along the track structure; based on the inter-layer response transmission relationship, the theoretical propagation path of the differential response characterization between the track slab layer, the filling layer, and the base layer and the corresponding response change constraints are determined, the response change constraints including amplitude attenuation continuity constraints and phase evolution consistency constraints along the propagation path; in the differential response spatial mapping... In the field, the differential response evolution sequence along the theoretical propagation path is extracted, and the deviation between the differential response evolution sequence and the response change constraint is calculated to generate the corresponding path anomaly matching degree index. Based on the path anomaly matching degree index, multiple candidate anomaly regions are screened, and the dominant layer of the anomaly response is determined by decomposing the contribution of differential responses in different layer structures. Based on the dominant layer and the cumulative characteristics of differential responses on the corresponding path, an anomaly degree evaluation function is constructed. By minimizing the deviation between the anomaly degree evaluation function and the actual differential response, the spatial location and defect degree of the anomaly region are inverted.

[0007] In a possible implementation, based on the path anomaly matching degree index, multiple candidate anomaly regions are screened, and the dominant layer of the anomaly response is determined by decomposing the contribution of the differential response in different layer structures. The following processing is performed: Based on the path anomaly matching degree index, the anomaly matching degree of each spatial unit in the differential response spatial mapping field along different propagation paths is aggregated to construct the path anomaly aggregation intensity corresponding to each spatial unit, and candidate anomaly regions are screened according to the path anomaly aggregation intensity; for each candidate anomaly region, the differential response corresponding to the region is projected layer by layer along the propagation path of the track slab layer, filling layer, and base layer in combination with the inter-layer response transmission relationship to obtain the differential response distribution corresponding to each layer structure; according to the attenuation gradient change and phase shift accumulation characteristics of the differential response distribution on the propagation path, the contribution weight of each layer structure to the overall differential response is calculated, wherein the contribution weight is used to characterize the degree of influence of each layer structure on the path anomaly matching degree index; the contribution weight is normalized, and contribution screening is performed according to the normalization result to establish the dominant layer.

[0008] In a possible implementation, disturbance-sensitive feature extraction is performed on the response characterization parameters to obtain multidimensional response features that characterize changes in local structural stiffness, damping, and interface contact state. The following processing is then performed: based on the joint distribution characteristics of the response characterization parameters in the time and frequency domains, the structural response signal is decomposed to obtain local response components corresponding to different frequency bands and time periods; a first feature parameter reflecting structural stiffness changes is extracted from the local response components using the resonant frequency offset and the change in frequency band energy concentration; a second feature parameter reflecting structural damping changes is extracted using the response decay rate and the change in energy dissipation ratio; a third feature parameter reflecting interface contact state changes is extracted using the degree of phase abrupt change and the propagation delay offset during wave propagation; and the first, second, and third feature parameters are normalized and coupled together to establish multidimensional response features.

[0009] In a possible implementation, the following processing is performed: the preset excitation load is a multi-band scanning excitation load, which applies excitation signals in different frequency ranges to excite the response characteristics of the track structure at different depth levels in order to establish structural response data.

[0010] In a possible implementation, the following processing is performed: the multi-source sensing collaborative acquisition system adopts a distributed array deployment along the track extension direction, and sets at least two rows of sensing units in the vertical track direction to form a multi-channel response acquisition structure covering different propagation paths.

[0011] In a possible implementation, a pre-built reference response feature of a defect-free state is invoked, and the following processing is performed: historical detection data of the same type of track structure in a healthy state is read; adaptive correction of the target slab track section is performed on the historical detection data to establish the reference response feature.

[0012] In a possible implementation, the following process is performed: before generating a differential response representation that reflects the degree of deviation from the current structural state, spatial continuity constraint processing of multidimensional response features is performed, and consistency correction is performed through the response features of adjacent spatial units to suppress the interference of local noise on the differential representation.

[0013] In a possible implementation, the quality inspection results are output, and the following processing is performed: spatial location identification of quality anomalies in the quality inspection results is performed to generate a quality anomaly sequence; after anomaly verification of the quality anomaly sequence, encrypted storage is performed.

[0014] This application also provides a quality inspection system for the interior of slab track, comprising: a structural response data acquisition module, used to construct a multi-point excitation and multi-source sensing collaborative acquisition system within the target slab track section, apply a preset excitation load to the track slab, and simultaneously acquire corresponding structural response data, the structural response data including vibration response signals and wave propagation response signals; a response characterization parameter mapping module, used to establish the inter-layer response transmission relationship under excitation based on the structural response data and combined with the layered structural characteristics of the slab track, and map the structural response data into response characterization parameters characterizing changes in the inter-layer connection state; and a disturbance... The dynamic-sensitive feature extraction module is used to extract disturbance-sensitive features from the response characterization parameters to obtain multi-dimensional response features that characterize changes in local structural stiffness, damping, and interface contact state. The differential response characterization generation module is used to call pre-constructed reference response features of a defect-free state, compare and analyze the multi-dimensional response features, and generate differential response characterizations that reflect the degree of deviation of the current structural state. The anomaly inversion and localization module is used to invert and locate the abnormal areas inside the track structure based on the differential response characterizations and the inter-layer response transmission relationship, obtain the corresponding defect location and defect severity data, and output the quality inspection results.

[0015] The proposed method and system for quality inspection of the interior of slab track, as described in this application, firstly constructs a multi-point excitation and multi-source sensing collaborative acquisition system within the target slab track section. A preset excitation load is applied to the track slab, and corresponding structural response data is simultaneously acquired. This structural response data includes vibration response signals and wave propagation response signals. Then, based on this structural response data and the layered structural characteristics of the slab track, an inter-layer response transmission relationship under excitation is established. The structural response data is then mapped to response characterization parameters representing changes in inter-layer connection states. Next, disturbance-sensitive feature extraction is performed on these response characterization parameters to obtain multi-dimensional response features characterizing changes in local structural stiffness, damping, and interface contact states. Then, a pre-constructed reference response feature of a defect-free state is invoked, and the multi-dimensional response features are compared and analyzed to generate a difference response characterization reflecting the degree of deviation from the current structural state. Finally, based on the difference response characterization and the inter-layer response transmission relationship, abnormal areas within the track structure are inverted and located to obtain the corresponding defect location and defect severity data, and the quality inspection results are output. Through the above process, the method and system proposed in this application achieve the technical effects of accurately identifying deep hidden defects, realizing continuous quantitative detection across the entire domain, and ensuring the integrity of the track structure without damage throughout the entire process. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a flowchart illustrating a quality inspection method for the interior of a slab track provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the structure of a quality inspection system for the interior of a slab track, provided in an embodiment of this application.

[0019] Figure labeling: 10 Structural response data acquisition module, 20 Response characterization parameter mapping module, 30 Disturbance sensitive feature extraction module, 40 Differential response characterization generation module, 50 Anomaly inversion and localization module. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This application provides a quality inspection method for the interior of slab track, applicable to the detection of internal defects in various types of CRTSⅠ, CRTSⅡ, and CRTSⅢ slab track structures in railways, passenger dedicated lines, and urban rail transit. It can accurately identify hidden quality anomalies such as interlayer separation, voids, cracks, and stiffness reduction between track slab layers, filling layers, and foundation layers. The inspection process does not require damage to the track structure itself, constituting a non-destructive, precise inspection method. Figure 1 As shown, the method includes: Step S100: A multi-point excitation and multi-source sensing collaborative acquisition system is constructed within the target slab track section. A preset excitation load is applied to the track slab, and corresponding structural response data is acquired simultaneously. The structural response data includes vibration response signals and wave propagation response signals. The preset excitation load is a multi-frequency scanning excitation load, which applies excitation signals in different frequency ranges to stimulate the response characteristics of the track structure at different depth levels, thereby establishing structural response data. The multi-source sensing collaborative acquisition system adopts a distributed array deployment along the track extension direction, and at least two rows of sensing units are set in the direction perpendicular to the track to form a multi-channel response acquisition structure covering different propagation paths.

[0022] Specifically, the target detection section is determined. For example, a standard slab track unit with a length of 20 meters is selected as the detection object. Debris on the track surface is cleaned to ensure that the excitation application position and the sensor unit placement position are flat and free of foreign objects. A multi-point excitation and multi-source sensing collaborative acquisition system is built. Portable electromagnetic vibrators are used as excitation equipment. For example, one is deployed every 2 meters along the track extension direction, and a total of 10 are deployed in the 20-meter section to form a multi-point synchronous excitation array, ensuring that the excitation covers the entire detection section. The sensing units are a combination of piezoelectric accelerometer and ultrasonic sensor units. One set is deployed every 1 meter along the track extension direction, and two rows are deployed perpendicular to the track direction. Each row corresponds to 20 sets of sensing units, forming a full-coverage distributed multi-channel sensing array. All sensing units are stably connected to the synchronous data acquisition instrument through shielded signal cables to ensure lossless signal transmission. A multi-band scanning excitation load was set, dividing the excitation frequency into three intervals: low frequency, medium frequency, and high frequency. The low-frequency interval (10Hz to 100Hz) specifically excites the response characteristics at the depth level of the track foundation layer; the medium-frequency interval (100Hz to 500Hz) specifically excites the response characteristics at the depth level of the cement asphalt mortar filling layer; and the high-frequency interval (500Hz to 2000Hz) specifically excites the response characteristics at the depth level of the track slab layer. A linear continuous scanning method was used to apply the excitation signal within each frequency interval, with a preferred sweep rate of 1Hz / s to 5Hz / s to ensure the continuity and stability of signal acquisition. Signals from all sensing units were simultaneously acquired using a synchronous data acquisition system. A piezoelectric accelerometer was responsible for acquiring vibration response signals, and an ultrasonic sensor was responsible for acquiring wave propagation response signals. These two types of signals were synchronously integrated, classified, and stored as structural response data in a standard format.

[0023] Step S200: Based on the structural response data and combined with the layered structural characteristics of the slab track, establish the interlayer response transmission relationship under excitation, and map the structural response data into response characterization parameters that characterize the changes in the interlayer connection state.

[0024] Specifically, the layered structure of the slab track is characterized as follows: the track structure consists of three continuous layers from top to bottom: a track slab layer, a cement asphalt mortar filling layer, and a concrete foundation layer. The interfaces between each layer are key structural connections and stress-bearing components. The core of the interlayer response transmission relationship is reflected in the transmission attenuation and reflection / transmission law of the excitation signal after it is input from the track slab surface, sequentially penetrating the track slab layer, the cement asphalt mortar filling layer, and the concrete foundation layer. The structural response data output in step S100 is retrieved, and the multi-channel response data is classified and categorized according to the one-to-one spatial correspondence between the excitation position and the sensor unit position. Abnormal discrete data caused by equipment interference or sudden external vibrations are removed. The unified criterion for judging abnormal data is: discrete data points whose signal amplitude exceeds three times the average amplitude of the same channel at the same time period are directly removed.

[0025] After removing outlier data, a suitable excitation-response transfer mathematical model was constructed by combining the inherent material properties, standard thickness parameters, and interface contact mechanical characteristics of the three-layer slab track structure. This model clarifies the propagation speed, fixed attenuation coefficient, and interface reflection and transmission quantification laws of excitation signals in different frequency bands within each structural layer, and refines the interlayer response transfer relationship. In a healthy, defect-free track structure, the excitation signal is transmitted smoothly between layers, with uniform amplitude attenuation, regular phase change, and stable propagation delay. If defects such as interlayer separation, voids, or internal cracks exist, the signal will undergo abnormal reflection, scattering, and diffraction at the defect interface, exhibiting abnormal transmission characteristics such as a sudden drop in amplitude, abrupt phase change, and a significant increase in propagation delay. Based on this, an interlayer response transfer discrimination criterion was established.

[0026] Based on this, the original vibration response signal and wave propagation response signal are standardized and quantized. External interference factors such as environmental noise, aftershocks from train passage, and vibrations from surrounding construction are eliminated through filtering. Five types of response characterization parameters that directly reflect the interlayer connection state are extracted, specifically including vibration amplitude attenuation rate, ultrasonic wave propagation delay, interface signal reflection coefficient, interlayer stiffness response coefficient, and damping response coefficient. All parameters are directly linked to the interlayer contact state and internal density of the track structure. In a defect-free state, the parameter values ​​are stable within a preset standard threshold range. After defects appear, the parameters will deviate significantly, and the magnitude of the deviation is positively correlated with the severity of the defect. Through this mapping process, the original response data is transformed into intuitive, quantifiable, and horizontally comparable response characterization parameters.

[0027] Step S300: Perform disturbance-sensitive feature extraction on the response characterization parameters to obtain multidimensional response features for characterizing changes in local stiffness, damping, and interface contact state of the structure.

[0028] Specifically, hidden defects within track structures are often located between layers or deep within the structure. Conventional overall response parameters are insufficient to capture the local mechanical state changes caused by minute defects, only reflecting the overall macroscopic performance of the structure and failing to pinpoint local hidden defects. These hidden defects directly cause local stiffness degradation, abrupt changes in damping characteristics, and abnormal interlayer interface contact states in the corresponding areas. These subtle disturbances are transmitted synchronously with the excitation response signal. Therefore, a multi-domain joint analysis method is used to specifically extract characteristic parameters highly sensitive to defect disturbances, amplify the differences between defect signals and normal signals, and eliminate irrelevant interference features, providing accurate characteristic basis for defect identification.

[0029] In one possible implementation, disturbance-sensitive feature extraction is performed on the response characterization parameters to obtain multidimensional response features that characterize changes in local structural stiffness, damping, and interface contact states. Step S300 further includes step S310, which decomposes the structural response signal based on the joint distribution characteristics of the response characterization parameters in the time domain, frequency domain, and time and frequency domain to obtain local response components corresponding to different frequency bands and time periods. Specifically, a combination of variational mode decomposition algorithm and wavelet packet transform is used for decomposition. The wavelet packet transform is preferably decomposed into 3 to 5 levels, and the variational mode decomposition presets the number of modal components to be 3 to 6, adapting to the layered signal stripping requirements of a three-layer structure. For the response signal corresponding to multi-frequency band excitation, it is decomposed according to the low-frequency, mid-frequency, and high-frequency bands respectively, stripping the response components corresponding to different structural layers, eliminating the globally stationary response part, and retaining the non-stationary local response components caused only by local structural anomalies, avoiding the global normal response from masking local defect features and improving the feature specificity and sensitivity.

[0030] Step S320: Extract a first characteristic parameter reflecting structural stiffness changes from the local response components using the resonant frequency offset and the change in frequency band energy concentration. Specifically, track structure stiffness is positively correlated with resonant frequency. Defects such as local stiffness attenuation, cracks, and voids will cause the resonant frequency in the corresponding region to shift to lower frequencies, and the frequency band energy concentration will decrease significantly. By calculating the offset difference between the resonant frequency reference value and the actual detected value of each local response component, and statistically analyzing the rate of change of the concentration coefficient of energy distribution within the corresponding frequency band, the two types of indicators are weighted and fused to obtain the first characteristic parameter. This parameter is used to characterize the degree of local stiffness degradation of the track slab, filling layer, and foundation layer, and to distinguish between uniform stiffness attenuation and local abrupt defects.

[0031] Step S330 involves extracting a second characteristic parameter reflecting changes in structural damping by analyzing the changes in response decay rate and energy dissipation ratio. Specifically, in a normal track structure, the interlayer bonding is tight, the damping characteristics are stable, and the excitation response decay rate is uniform. Interlayer delamination and internal voids lead to an abnormal increase in structural damping, a faster response signal decay rate, and a significantly increased energy dissipation ratio. For each local response component, the signal amplitude decay slope per unit time is calculated, the dissipation ratio of excitation energy input to response energy output is statistically analyzed, and the change amplitude is calculated by comparing it with the baseline value of the healthy state. These parameters are then fused to form the second characteristic parameter, which identifies interlayer bonding failure and internal loosening defects, compensating for the inability of simple stiffness characteristics to identify damping defects.

[0032] Step S340 involves extracting a third characteristic parameter reflecting changes in the interface contact state by analyzing the degree of phase abrupt change and the propagation delay offset during wave propagation. Specifically, when ultrasonic waves and vibration waves propagate through a complete interlayer interface, the phase is continuous and the propagation delay is fixed. If interlayer separation or gaps occur, the wave signal is reflected and scattered at the interface, resulting in abrupt phase changes and prolonged propagation delays. By calculating the phase difference abrupt change value and the offset of the propagation delay from the reference value of the signals received by each sensing unit, the tightness of the interface contact is quantified, and the third characteristic parameter is obtained through fusion. This parameter is specifically designed to accurately capture hidden interface defects such as interlayer separation and interface gaps.

[0033] Step S350 involves normalizing and coupling the first, second, and third feature parameters to establish a multidimensional response feature. Specifically, since the dimensions and numerical ranges of the three types of feature parameters differ, direct comparison can easily lead to an imbalance in feature weights. Therefore, the range standardization method is used to normalize the three types of parameters within the 0-1 interval to eliminate the influence of dimensions. Then, based on the characteristics of different types of track structures and combined with engineering measurement data, reasonable weights are assigned to each type of feature to complete the coupling combination, forming a multidimensional response feature vector that includes stiffness, damping, and interface contact state. This comprehensively covers the response performance of various hidden defects in the track structure, avoiding the problems of missed or false detections by single features.

[0034] Step S400: Invoke the pre-constructed reference response features of the defect-free state, perform comparative analysis on the multidimensional response features, and generate a difference response characterization that reflects the degree of deviation of the current structural state.

[0035] Specifically, a single test data point cannot directly determine whether a structure has defects. It is necessary to rely on the standard response characteristics of a healthy, defect-free track as a reference benchmark. The reference response characteristics are the standard response templates of the same type of healthy track. By comparing the multi-dimensional response characteristics obtained from the actual test with the standard templates dimension by dimension and point by point, the degree of difference between the two is quantified, which intuitively reflects the extent to which the current test section's structural state deviates from the healthy state. The magnitude of the difference directly corresponds to the severity of the defect.

[0036] In one possible implementation, a pre-built reference response feature of a defect-free state is invoked. Step S400 further includes step S410, which reads historical inspection data of the same type of track structure under healthy conditions. Specifically, CRTS series slab track with the same line, model, and construction process is selected, and the initial health inspection data after completion and acceptance, as well as the periodic inspection data of defect-free sections during the operation period, are retrieved. This ensures that the historical data is highly consistent with the structural parameters, material properties, and service environment of the target inspection section, avoiding distortion of the reference benchmark due to differences in track type. The data includes complete response parameters and multi-dimensional feature values ​​under multi-frequency excitation, forming a basic health feature database.

[0037] Step S420 involves performing adaptive correction on the historical inspection data for the target slab track section, establishing reference response characteristics. Specifically, there are slight differences in environmental temperature, foundation settlement, and service life among different inspection sections. Directly applying historical data would result in benchmark deviations. Therefore, by combining the on-site environmental parameters of the target section, track service duration, and foundation settlement monitoring data, temperature compensation, time correction, and spatial adaptation adjustments are made to the historical health characteristics to eliminate systematic errors caused by environmental and service conditions. This generates a unique, defect-free reference response characteristic adapted to the target inspection section, ensuring the accuracy of the comparative analysis.

[0038] In one possible implementation, before generating a differential response characterization reflecting the degree of deviation from the current structural state, step S400 further includes step S430, which performs spatial continuity constraint processing on the multidimensional response features. Consistency correction is performed using the response features of adjacent spatial units to suppress interference from local noise on the differential characterization. Specifically, the slab track structure is continuously distributed, and the response features of healthy sections are continuous and stable along the track extension direction, with only defective areas exhibiting local feature abrupt changes. Equipment noise and environmental interference are mostly isolated random anomalies and lack spatial continuity. Therefore, smoothing correction is performed on the multidimensional response features of adjacent detection units to eliminate isolated random anomalies and retain continuously distributed feature abrupt change regions, further improving the reliability of the differential response characterization and avoiding misjudging noise as structural defects.

[0039] After completing spatial continuity correction, the multidimensional response features obtained from actual detection of the target section are precisely compared with the adapted reference response features, dimension by dimension and spatial point by spatial point. The absolute deviation value and relative deviation rate of each dimension feature are calculated, and then a weighted calculation is performed to obtain the comprehensive deviation degree, which is integrated to form a standardized differential response characterization. This characterization adopts a combination of numerical quantization and spatial mapping. The numerical value directly reflects the degree to which the structural state deviates from the healthy state; the larger the value, the more severe the defect. The spatial distribution corresponds to the actual detection points on the track, marking the distribution range and dense areas of abnormal responses.

[0040] Step S500: Based on the differential response characterization and interlayer response transmission relationship, the abnormal area inside the track structure is inverted and located to obtain the corresponding defect location and defect degree data, and the quality inspection result is output.

[0041] Specifically, differential response characterization can only reflect the location and degree of deviation of abnormal points, and cannot directly determine the structural layer, precise spatial location and severity level of defects. It is necessary to combine the inter-layer response transmission relationship established in the early stage, utilize the signal propagation law and differential response distribution characteristics, and achieve layered positioning, precise positioning and severity quantification of defects through mathematical inversion and spatial analysis. Finally, complete quality inspection results that can be directly used for engineering maintenance are output, clarifying specific defect information and treatment suggestions.

[0042] In one possible implementation, the abnormal region inside the track structure is inverted and located based on the differential response characterization and interlayer response transmission relationship. Step S500 further includes step S510, which reconstructs the differential response characterization according to the spatial correspondence between the excitation location and the response acquisition location to form a differential response spatial mapping field distributed along the track structure. Specifically, a three-dimensional spatial mapping field is constructed with the track extension direction as the horizontal axis, the perpendicular track direction as the vertical axis, and the differential response deviation as the vertical axis. The differential response values ​​of each detection unit are mapped to the actual spatial coordinates of the track to form a visualized differential distribution cloud map, presenting the high-value core area, distribution range, and gradient change of the differential response, so as to intuitively locate the suspected defect core area.

[0043] Step S520: Based on the interlayer response transmission relationship, determine the theoretical propagation path and corresponding response change constraints of the differential response characterization among the track slab layer, filling layer, and base layer. These constraints include amplitude attenuation continuity constraints and phase evolution consistency constraints along the propagation path. Specifically, based on the standard thickness of the three-layer structure, the inherent propagation characteristics of the materials, and the signal propagation parameters obtained from previous modeling, calculate the theoretical propagation paths of excitation signals in different frequency bands within each structural layer. Determine the hard constraint rules for continuous amplitude attenuation and uniform phase evolution of the signal along the path under healthy conditions. If defects exist within the structure, the signal propagation path will be distorted, exceeding the above constraints and exhibiting phenomena such as sudden amplitude changes, phase jumps, and abnormal propagation delays. This serves as the core constraint basis for defect identification.

[0044] Step S530: In the differential response spatial mapping field, extract the differential response evolution sequence along the theoretical propagation path, and calculate the deviation degree between the differential response evolution sequence and the response change constraint conditions to generate a corresponding path anomaly matching degree index. Specifically, for each theoretical propagation path, extract differential response values ​​point by point to construct a complete differential response evolution sequence. Compare the deviation score of a single point with the comprehensive deviation score of the entire path by comparing it with the response change constraint conditions to obtain the path anomaly matching degree index. The higher the index value, the greater the probability that there is a structural defect in the area corresponding to the path. This is used for preliminary screening of suspected defective paths and filtering out low-probability anomaly points.

[0045] Step S540 involves filtering multiple candidate anomaly regions based on the path anomaly matching degree index, and determining the dominant layer of the anomaly response by decomposing the contribution of differential responses in different structural layers. Specifically, the initially screened suspected defect areas contain multiple overlapping and low-confidence anomaly points, requiring further precise filtering using the path anomaly matching degree index to eliminate false and weak anomaly regions and lock in high-confidence candidate anomaly regions. Simultaneously, the slab track has a three-layer structure, and a single anomaly response may be formed by the superposition of responses from multiple layers. It is necessary to decompose the contribution of each layer to the differential response, identify the core structural layer causing the anomaly (i.e., the dominant layer), avoid misjudging the defect layer due to the superposition of multiple responses, achieve precise defect layering and location, and provide precise targeting for maintenance.

[0046] Step S550: Based on the cumulative characteristics of the differential response on the dominant layer and its corresponding path, an anomaly severity evaluation function is constructed. By minimizing the deviation between the anomaly severity evaluation function and the actual differential response, the spatial location and defect severity of the anomaly area are inverted. Specifically, a standardized anomaly severity evaluation function is constructed by combining the dominant layer type, the cumulative value of the differential response on the corresponding propagation path, spatial coordinate parameters, and inter-layer propagation characteristics. The actual differential response is used as the target value. An iterative optimization algorithm minimizes the deviation between the evaluation function output value and the actual differential response, accurately inverting the lateral mileage location, vertical track direction location, and vertical burial depth of the defect. At the same time, based on the cumulative value of the differential response and the magnitude of the deviation, the defect is uniformly divided into three levels: minor, moderate, and severe, achieving dual accurate quantification of defect location and defect severity.

[0047] In one possible implementation, based on the path anomaly matching degree index, multiple candidate anomaly regions are screened, and the dominant layer of the anomaly response is determined by decomposing the contribution of the differential response in different layer structures. Step S540 further includes step S541, which aggregates the anomaly matching degree of each spatial unit in the differential response spatial mapping field along different propagation paths based on the path anomaly matching degree index, constructs the path anomaly aggregation intensity corresponding to each spatial unit, and filters candidate anomaly regions based on the path anomaly aggregation intensity. Specifically, the anomaly matching degree of intersecting and overlapping propagation paths is weighted and aggregated, duplicate anomaly scores are removed, and the comprehensive anomaly intensity of each spatial unit is obtained. A standardized intensity threshold is set, weak anomalies and isolated anomalies below the threshold are eliminated, and continuous distribution areas with high aggregation intensity are retained as the final candidate anomaly regions, thereby narrowing the defect location range and improving analysis efficiency and accuracy.

[0048] Step S542: For each candidate anomaly region, the differential response corresponding to the region is projected layer by layer along the propagation path of the track slab layer, filling layer, and foundation layer, based on the inter-layer response transmission relationship, to obtain the differential response distribution corresponding to each layer structure. Specifically, according to the excitation characteristics of low-frequency, medium-frequency, and high-frequency signals corresponding to the foundation layer, cement asphalt mortar filling layer, and track slab layer, respectively, the multidimensional differential response is decomposed and mapped to the corresponding structural layer, generating differential response distribution cloud maps for each structural layer, clarifying the distribution location, magnitude, and diffusion range of the differential response in each layer.

[0049] Step S543: Based on the attenuation gradient changes and phase shift accumulation characteristics of the differential response distribution along the propagation path, calculate the contribution weight of each layer structure to the overall differential response. The contribution weight characterizes the influence of each layer structure on the path anomaly matching index. Specifically, analyze the attenuation gradient and phase shift accumulation of each layer's differential response to quantify the contribution ratio of each layer's anomaly response to the overall difference. A higher contribution weight indicates a greater likelihood of defects in that layer structure.

[0050] Step S544: The contribution weights are normalized, and contribution screening is performed based on the normalization results to establish the dominant layer. Specifically, the contribution weights are normalized in the 0-1 range to eliminate differences in numerical magnitude. Based on the normalization results, contribution screening is performed, and the structural layer with the highest contribution weight is determined as the dominant layer. This clarifies the core defect occurrence layer, distinguishes different types of defects such as surface cracks in the track slab, hollowing and detachment of the cement asphalt mortar filling layer, and settlement damage in the concrete foundation layer, and avoids misjudgment or incorrect identification of defect layers caused by the superposition of multi-layer structural responses, thus achieving accurate layered location of defects.

[0051] In one possible implementation, the quality inspection results are output, and step S500 further includes step S560, which identifies the spatial location of quality anomalies in the quality inspection results to generate a quality anomaly sequence. Specifically, according to the track mileage station, planar spatial coordinates, and vertical structural layers, all detected defect areas are numbered one by one, and the corresponding dominant layer, location coordinates, defect type, and defect level are labeled to form a quality anomaly sequence. At the same time, visual annotations are made on the track structure plan view and vertical layered profile view to present the overall distribution and severity of defects along the entire line.

[0052] Step S570: After verifying the quality anomaly sequence, the data is encrypted and stored. Specifically, on-site manual review and instrument sampling re-inspection are used to verify severe defects and suspected misjudged anomalies, correcting minor detection errors. The final quality inspection results after successful verification, including original structural response data, multidimensional response characteristics, differential response characterization, and complete defect location information, are stored in an encrypted format and classified to establish inspection files for corresponding inspection sections. These files are used for subsequent routine track maintenance, periodic re-inspection and comparison, and defect source analysis, achieving standardized and traceable management of inspection data throughout its entire lifecycle.

[0053] This application employs a multi-point excitation and multi-source sensing collaborative acquisition system to be built in the test section of slab track. A preset excitation load is applied and two types of structural response data, vibration and wave propagation, are collected simultaneously. The interlayer response transmission relationship is established in combination with the layered structure of the track. The original response data is transformed into quantitative parameters that characterize the interlayer connection state. Multi-dimensional sensitive features that can reflect local stiffness, damping and interface contact changes are extracted from the parameters. Reference features of defect-free healthy tracks are retrieved for comparison to generate a difference characterization of structural state deviation. Based on the difference characterization and the interlayer response transmission law, abnormal areas are located, the location and severity of defects are obtained and the detection results are output. These technical means solve the technical problems of weak deep defect identification ability, inability to continuously quantify and evaluate, and easy structural damage in the detection of internal quality defects of existing slab track. It achieves the technical effects of accurately identifying deep hidden defects, realizing continuous quantitative detection throughout the entire range, and ensuring the integrity of the track structure without damage throughout the process.

[0054] In the above text, refer to Figure 1 A quality inspection method for the interior of a slab track according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 A quality inspection system for the interior of a slab track is described according to an embodiment of the present invention.

[0055] The quality inspection system for the interior of slab track according to embodiments of the present invention addresses the technical problems of weak deep defect identification capability, inability to continuously quantify and evaluate defects, and easy structural damage in existing slab track internal quality defect detection methods. It achieves the technical effects of accurately identifying deep hidden defects, realizing continuous quantitative detection across the entire area, and ensuring the integrity of the track structure without damage throughout the entire process. The quality inspection system for the interior of slab track includes: a structural response data acquisition module 10, a response characterization parameter mapping module 20, a disturbance-sensitive feature extraction module 30, a difference response characterization generation module 40, and an anomaly inversion and localization module 50.

[0056] The structural response data acquisition module 10 is used to construct a multi-point excitation and multi-source sensing collaborative acquisition system within the target slab track section, apply a preset excitation load to the track slab, and simultaneously acquire corresponding structural response data, including vibration response signals and wave propagation response signals; the response characterization parameter mapping module 20 is used to establish the inter-layer response transmission relationship under excitation based on the structural response data and combined with the layered structural characteristics of the slab track, and map the structural response data into response characterization parameters characterizing changes in the inter-layer connection state; the disturbance sensitive feature extraction module 30 is used to extract... The response characterization parameters perform disturbance-sensitive feature extraction to obtain multidimensional response features for characterizing local stiffness changes, damping changes, and interface contact state changes of the structure; the difference response characterization generation module 40 is used to call the pre-constructed reference response features of the defect-free state, compare and analyze the multidimensional response features, and generate a difference response characterization reflecting the degree of deviation of the current structural state; the anomaly inversion and localization module 50 is used to invert and locate the abnormal area inside the track structure according to the difference response characterization and the inter-layer response transmission relationship, obtain the corresponding defect location and defect degree data, and output the quality inspection results.

[0057] The specific configuration of the anomaly inversion and localization module 50 is described in detail below: As mentioned above, the anomaly region inside the track structure is inverted and located based on the differential response characterization and the inter-layer response transmission relationship. The anomaly inversion and localization module 50 may further include: a differential response spatial mapping field generation unit for reconstructing the differential response characterization according to the spatial correspondence between the excitation position and the response acquisition position to form a differential response spatial mapping field distributed along the track structure; and a theoretical propagation path determination unit for determining the theoretical propagation path of the differential response characterization between the track slab layer, the filling layer, and the base layer and the corresponding response change constraints based on the inter-layer response transmission relationship. The response change constraints include amplitude attenuation continuity constraints and phase constraints along the propagation path. Evolutionary consistency constraints; the deviation degree calculation unit is used to extract the differential response evolution sequence along the theoretical propagation path in the differential response spatial mapping field, and calculate the deviation degree between the differential response evolution sequence and the response change constraint conditions, generating a corresponding path anomaly matching degree index; the screening processing unit is used to perform screening processing on multiple candidate anomaly regions based on the path anomaly matching degree index, and determine the dominant layer of the anomaly response by decomposing the contribution of differential responses in different layer structures; the inversion unit is used to construct an anomaly degree evaluation function based on the dominant layer and the cumulative characteristics of differential responses on the corresponding path, and perform the inversion of the spatial location and defect degree of the anomaly region by minimizing the deviation between the anomaly degree evaluation function and the actual differential response.

[0058] The process involves several steps. First, based on the path anomaly matching degree index, multiple candidate anomaly regions are screened. Second, by decomposing the contribution of differential responses in different layer structures, the dominant layer of the anomaly response is determined. The screening unit may further include: an aggregation subunit, which aggregates the anomaly matching degree of each spatial unit along different propagation paths in the differential response spatial mapping field based on the path anomaly matching degree index, constructs the path anomaly aggregation intensity for each spatial unit, and screens candidate anomaly regions based on the path anomaly aggregation intensity; a layered projection subunit, which, for each candidate anomaly region, performs layered projection of the corresponding differential response along the propagation paths of the track slab layer, filling layer, and base layer, based on the inter-layer response transmission relationship, to obtain the differential response distribution corresponding to each layer structure; a contribution weight calculation subunit, which calculates the contribution weight of each layer structure to the overall differential response based on the attenuation gradient change and phase shift accumulation characteristics of the differential response distribution along the propagation path, wherein the contribution weight characterizes the degree of influence of each layer structure on the path anomaly matching degree index; and a dominant layer establishment subunit, which normalizes the contribution weight, performs contribution screening based on the normalization result, and establishes the dominant layer.

[0059] The specific configuration of the disturbance-sensitive feature extraction module 30 is described in detail below: As mentioned above, disturbance-sensitive feature extraction is performed on the response characterization parameters to obtain multidimensional response features used to characterize changes in local stiffness, damping, and interface contact state of the structure. The disturbance-sensitive feature extraction module 30 may further include: a decomposition processing unit for decomposing the structural response signal based on the joint distribution characteristics of the response characterization parameters in the time domain, frequency domain, and time and frequency domain to obtain local response components corresponding to different frequency bands and different time periods; a first feature parameter extraction unit for extracting a first feature parameter reflecting changes in structural stiffness from the local response components through the resonant frequency offset and the change in frequency band energy concentration; a second feature parameter extraction unit for extracting a second feature parameter reflecting changes in structural damping through the change in response decay rate and energy dissipation ratio; a third feature parameter extraction unit for extracting a third feature parameter reflecting changes in interface contact state through the degree of phase change and the propagation delay offset during wave propagation; and a normalization coupling combination processing unit for performing normalization coupling combination processing on the first feature parameter, the second feature parameter, and the third feature parameter to establish multidimensional response features.

[0060] The specific configuration of the structural response data acquisition module 10 is described in detail below: As mentioned above, the structural response data acquisition module 10 may further include: the preset excitation load is a multi-band scanning excitation load, which applies excitation signals in different frequency ranges to excite the response characteristics of the track structure at different depth levels in order to establish structural response data.

[0061] The structural response data acquisition module 10 may further include: a multi-source sensing collaborative acquisition system that adopts a distributed array deployment along the track extension direction, and sets at least two rows of sensing units in the vertical track direction to form a multi-channel response acquisition structure covering different propagation paths.

[0062] The detailed description of the specific configuration of the difference response characterization generation module 40 is explained as follows: As mentioned above, the difference response characterization generation module 40 can further include: a historical detection data reading unit for reading historical detection data of the same type of track structure in a healthy state; and an adaptation and adaptive correction unit for performing adaptation and adaptive correction on the target slab track section of the historical detection data to establish reference response characteristics.

[0063] Before generating the difference response representation reflecting the degree of deviation of the current structural state, the difference response representation generation module 40 may further include: a consistency correction unit for performing spatial continuity constraint processing of multidimensional response features, and performing consistency correction through the response features of adjacent spatial units to suppress the interference of local noise on the difference representation.

[0064] The anomaly inversion and localization module 50, which outputs quality detection results, may further include: a spatial location identification unit for identifying the spatial location of quality anomalies in the quality detection results and generating a quality anomaly sequence; and an encrypted storage unit for performing encrypted storage after verifying the quality anomaly sequence.

[0065] The quality inspection system for the interior of slab track provided in this embodiment of the invention can execute the quality inspection method for the interior of slab track provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0066] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0067] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A quality inspection method for the interior of slab track, characterized in that, The method includes: A multi-point excitation and multi-source sensing collaborative acquisition system is constructed within the target slab track section. A preset excitation load is applied to the track slab, and the corresponding structural response data is acquired simultaneously. The structural response data includes vibration response signals and wave propagation response signals. Based on the structural response data and combined with the layered structural characteristics of slab track, the interlayer response transmission relationship under excitation is established, and the structural response data is mapped into response characterization parameters that characterize the changes in interlayer connection state. Perturbation-sensitive feature extraction is performed on the response characterization parameters to obtain multidimensional response features for characterizing changes in local stiffness, damping, and interface contact state of the structure. The reference response features of the pre-constructed defect-free state are invoked, and the multi-dimensional response features are compared and analyzed to generate a difference response characterization that reflects the degree of deviation of the current structural state. Based on the differential response characterization and interlayer response transmission relationship, the abnormal regions inside the track structure are inverted and located to obtain the corresponding defect location and defect severity data, and the quality inspection results are output.

2. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, Based on the aforementioned differential response characterization and inter-layer response transfer relationship, the anomalous region within the orbital structure is inverted and located, including: The differential response characterization is reconstructed according to the spatial correspondence between the excitation location and the response acquisition location to form a differential response spatial mapping field distributed along the track structure; Based on the interlayer response transmission relationship, the theoretical propagation path of the differential response characterization between the track slab layer, the filling layer and the base layer and the corresponding response change constraints are determined. The response change constraints include the amplitude attenuation continuity constraint and the phase evolution consistency constraint along the propagation path. In the differential response space mapping field, the differential response evolution sequence along the theoretical propagation path is extracted, and the degree of deviation between the differential response evolution sequence and the response change constraint is calculated to generate the corresponding path anomaly matching degree index. Based on the path anomaly matching degree index, multiple candidate anomaly regions are screened, and the dominant layer of the anomaly response is determined by decomposing the contribution of the difference response in different layer structures. Based on the cumulative characteristics of the differential responses on the dominant layer and the corresponding path, an anomaly degree evaluation function is constructed. By minimizing the deviation between the anomaly degree evaluation function and the actual differential response, the spatial location and defect degree of the anomaly region are inverted.

3. The quality inspection method for the interior of slab track as described in claim 2, characterized in that, Based on the path anomaly matching degree index, multiple candidate anomaly regions are screened, and the dominant layer of the anomaly response is determined by decomposing the contribution of the difference response in different layer structures, including: Based on the path anomaly matching degree index, the anomaly matching degree of each spatial unit in the differential response spatial mapping field along different propagation paths is aggregated to construct the path anomaly aggregation intensity of each spatial unit, and candidate anomaly regions are selected according to the path anomaly aggregation intensity. For each candidate anomaly region, the differential response corresponding to the region is projected along the propagation path of the track slab layer, filling layer, and base layer in combination with the inter-layer response transmission relationship to obtain the differential response distribution corresponding to each layer structure. Based on the attenuation gradient change and phase shift accumulation characteristics of the differential response distribution along the propagation path, the contribution weight of each layer structure to the overall differential response is calculated, wherein the contribution weight is used to characterize the degree of influence of each layer structure on the path anomaly matching degree index. The contribution weights are normalized, and contributions are screened based on the normalization results to establish a dominant layer.

4. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, Perturbation-sensitive feature extraction is performed on the response characterization parameters to obtain multidimensional response features for characterizing changes in local structural stiffness, damping, and interface contact states, including: Based on the distribution characteristics of the response characterization parameters in the time domain, frequency domain, and joint time and frequency domain, the structural response signal is decomposed to obtain local response components corresponding to different frequency bands and different time periods. The first characteristic parameter reflecting the change in structural stiffness is extracted from the local response components by means of the resonant frequency offset and the change in frequency band energy concentration. A second characteristic parameter reflecting the change in structural damping is extracted by the change in response decay rate and energy dissipation ratio. The third characteristic parameter reflecting the change in interface contact state is extracted by the degree of phase change and the propagation time delay offset during wave propagation. The first feature parameter, the second feature parameter, and the third feature parameter are normalized and coupled together to establish a multidimensional response feature.

5. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, The preset excitation load is a multi-band scanning excitation load, which applies excitation signals in different frequency ranges to stimulate the response characteristics of the track structure at different depth levels, so as to establish structural response data.

6. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, The multi-source sensing collaborative acquisition system adopts a distributed array deployment along the track extension direction, and sets at least two rows of sensing units in the vertical track direction to form a multi-channel response acquisition structure covering different propagation paths.

7. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, Invoke the reference response features of the pre-built defect-free state, including: Read historical test data of similar track structures under healthy conditions; Adaptive correction is performed on the target slab track section based on the historical detection data to establish reference response characteristics.

8. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, Before generating a differential response representation that reflects the degree of deviation from the current structural state, spatial continuity constraint processing of multidimensional response features is performed. Consistency correction is performed through the response features of adjacent spatial units to suppress the interference of local noise on the differential representation.

9. The quality inspection method for the interior of slab track as described in claim 1, characterized in that, Output quality inspection results, including: The spatial location of quality anomalies in the quality inspection results is identified, and a quality anomaly sequence is generated. After verifying the quality anomaly sequence, it is then encrypted and stored.

10. A quality inspection system for the interior of slab track, characterized in that, The system is used to implement the quality inspection method for the interior of slab track as described in any one of claims 1-9, the system comprising: The structural response data acquisition module is used to construct a multi-point excitation and multi-source sensing collaborative acquisition system within the target slab track section, apply a preset excitation load to the track slab, and simultaneously acquire the corresponding structural response data, which includes vibration response signals and wave propagation response signals. The response characterization parameter mapping module is used to establish the inter-layer response transmission relationship under excitation based on the structural response data and combined with the layered structural characteristics of the slab track, and to map the structural response data into response characterization parameters that characterize the changes in the inter-layer connection state. The disturbance-sensitive feature extraction module is used to perform disturbance-sensitive feature extraction on the response characterization parameters to obtain multidimensional response features that characterize changes in local stiffness, damping, and interface contact state of the structure. The differential response characterization generation module is used to call the pre-built reference response features of the defect-free state, compare and analyze the multi-dimensional response features, and generate a differential response characterization that reflects the degree of deviation of the current structural state. The anomaly inversion and localization module is used to invert and locate the abnormal areas inside the track structure based on the differential response characterization and inter-layer response transmission relationship, obtain the corresponding defect location and defect severity data, and output the quality inspection results.

Images