Intelligent electric monitoring method and system for bridge crack

Abstract

The application relates to the technical field of crack monitoring, and discloses a bridge crack intelligent electric monitoring method and system, which comprises the following steps: collecting, in real time, electric signals of a bridge structure through preset conductive fiber bundles to obtain a first signal set; decoupling the first signal set according to the first signal set and combining pre-acquired environmental data to analyze environmental interference signals, to obtain a second signal set; analyzing signal growth and step mutation of the second signal set, extracting corresponding signal features, identifying crack edges, and obtaining crack edge identification results; and evolving and predicting crack expansion through a preset material crack coupling model according to the crack edge identification results, to obtain crack parameter prediction results. The application can improve signal collection efficiency, improve crack identification and monitoring efficiency and accuracy, evolve and predict crack expansion, provide a safety warning for cracks, and help prevent further expansion of cracks.

Description

Technical Field

[0001] This application relates to the field of crack monitoring technology, and more specifically to a smart electrical monitoring method and system for bridge cracks. Background Technology

[0002] During the service life of bridges, cracks will inevitably develop in the structure due to the long-term effects of traffic loads and the natural environment. Monitoring these cracks is crucial for ensuring the structural safety of bridges. Currently, with the rapid development of transportation infrastructure and the introduction of numerous bridges, the safe operation of bridges ensures smooth traffic flow. Timely detection of potential bridge cracks allows for real-time monitoring and prediction of their spread, thus guaranteeing the structural safety of the bridge.

[0003] Existing technologies suffer from the following problems: monitoring efficiency is low due to the reliance on a single sensor, and the methods are highly susceptible to environmental factors and the experience of the inspectors. They may miss or misjudge some hidden areas or minute cracks. Furthermore, the limited monitoring range of the sensors makes it impossible to identify the comprehensive distribution of cracks in the bridge structure. Directly acquired signal data is susceptible to environmental interference, leading to reduced accuracy and stability of the monitoring data and an inability to accurately identify parameters such as crack width and depth. Single-sensor monitoring of cracks lacks effective prediction of crack propagation, hindering bridge safety early warning and proactive maintenance. To address at least one of these problems, this application proposes an intelligent electrical monitoring method and system for bridge cracks. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the purpose of this application is to provide an intelligent electrical monitoring method and system for bridge cracks, which can effectively solve the problems in the background technology. The specific technical solution of this application is as follows:

[0005] A smart electrical monitoring method for bridge cracks includes:

[0006] The electrical signals of the bridge structure are collected in real time by a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and a micro sensor is embedded at the interlacing point of the conductive fiber bundle.

[0007] Based on the first signal set, environmental interference signals are analyzed in conjunction with pre-acquired environmental data. The first signal set is then decoupled using the environmental interference signals to obtain a second signal set.

[0008] The signal growth and step change of the second signal set are analyzed, the corresponding signal features are extracted, and the crack edges are identified to obtain the crack edge identification results.

[0009] Based on the crack edge identification results, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain crack parameter prediction results, thereby enabling intelligent monitoring of bridge cracks.

[0010] Specifically, the method involves real-time acquisition of electrical signals from the bridge structure using a pre-set conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layered structure with micro-sensors embedded at the interlacing points of the fiber bundles.

[0011] Signal acquisition is performed by setting up a pre-defined conductive fiber bundle on the bridge. The conductive fiber bundle includes a graphene-coated fiber layer, a polyurethane matrix layer, and a carbon fiber bundle layer. A micro-sensor is embedded at the interlacing point of the conductive fiber bundle.

[0012] By simultaneously exciting multiple conductive fiber bundles, the electrical signals are synchronously acquired in real time to obtain the first signal set.

[0013] Specifically, the method involves parallel excitation of multiple conductive fiber bundles to synchronously acquire electrical signals in real time, thereby obtaining a first signal set, including:

[0014] Multiple conductive fiber bundles are assigned orthogonal carrier frequencies and corresponding AC excitation signals are applied to synchronously acquire the response signals of each conductive fiber bundle;

[0015] The response signals of each conductive fiber bundle are separated by Fourier transform to obtain the first signal set.

[0016] Specifically, based on the first signal set, environmental interference signals are analyzed in conjunction with pre-acquired environmental data. The first signal set is then decoupled using the environmental interference signals to obtain a second signal set, including:

[0017] The reference value of each signal is calculated based on the first signal set to obtain the environmental reference signal set;

[0018] By combining the aforementioned environmental reference signal set and the pre-acquired environmental data, the signal residuals are analyzed to obtain the environmental interference signal;

[0019] The environmental interference signal is decoupled and separated from the first signal set to obtain the second signal set.

[0020] Specifically, by combining the aforementioned environmental reference signal set and pre-acquired environmental data, the signal residuals are analyzed to obtain environmental interference signals, including:

[0021] The first residual set is obtained by calculating the difference between the environmental reference signal set and the corresponding signal in the first signal set;

[0022] By combining the first residual set and the pre-acquired environmental data, the signal residual of environmental interference is calculated using a preset residual analysis model to obtain the environmental interference signal.

[0023] Specifically, the signal growth and step abrupt changes of the second signal set are analyzed, the corresponding signal features are extracted, and the crack edges are identified to obtain crack edge identification results, including:

[0024] The signal growth and step change of the second signal set were analyzed, and the corresponding signal features were extracted to identify the crack region.

[0025] In the crack region, edge pixels are fitted, crack angle and crack width are analyzed, and crack length is calculated based on crack region size;

[0026] By combining the crack angle, crack width, and crack length, the crack edge identification result is obtained.

[0027] Specifically, the analysis of signal growth and step abrupt changes in the second signal set, and the extraction of corresponding signal features to identify the crack region, includes:

[0028] Analyze the resistance change rate and capacitance step change point in the second signal set to extract the change range and step change point set where the resistance change rate is greater than the preset resistance change threshold.

[0029] According to the structure of the bridge, the set of change intervals and step change points are mapped to the corresponding positions of the bridge;

[0030] A resistance gradient map is generated based on the change range, and a capacitance mutation map is generated based on the set of step mutation points.

[0031] The resistance gradient map and the capacitance abrupt change map are weighted and fused to obtain a fused gradient map;

[0032] The crack region is identified from the fused gradient map using a preset crack edge recognition model.

[0033] Specifically, based on the crack edge identification results, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain crack parameter prediction results, thereby enabling intelligent monitoring of bridge cracks, including:

[0034] Crack parameter information is extracted from the crack edge identification results and combined with environmental data to construct a spatiotemporal feature vector;

[0035] Based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain a first prediction result;

[0036] The first prediction result is corrected by adjusting the crack width in real time to obtain the crack parameter prediction result.

[0037] Specifically, based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain a first prediction result, including:

[0038] Based on the spatiotemporal feature vector, the coupling relationship between material stress intensity and crack propagation is analyzed through a pre-defined material crack coupling model to obtain the crack propagation rate.

[0039] According to a preset time period, the evolution of crack propagation is predicted based on the crack propagation rate to obtain a first prediction result.

[0040] A smart electrical monitoring system for bridge cracks, used to implement the aforementioned smart electrical monitoring method for bridge cracks, includes:

[0041] The signal acquisition module acquires electrical signals of the bridge structure in real time through a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and embeds micro sensors at the interlacing points of the conductive fiber bundle.

[0042] The signal processing module analyzes environmental interference signals based on the first signal set and pre-acquired environmental data, and decouples the first signal set through the environmental interference signals to obtain a second signal set.

[0043] The crack identification module analyzes the signal growth and step change of the second signal set, extracts the corresponding signal features, and identifies the crack edge to obtain the crack edge identification result.

[0044] The crack prediction module, based on the crack edge identification results, uses a preset material crack coupling model to predict the evolution of crack propagation and obtain crack parameter prediction results for intelligent monitoring of bridge cracks.

[0045] The beneficial effects of this application are as follows: Signal acquisition is based on conductive fiber bundles with multi-layered structures and embedded micro-sensors at the interlacing points. By parallel excitation and synchronous acquisition of multiple conductive fiber bundles, signal data is acquired. Environmental interference signals in the signal data are decoupled to obtain accurate signal data. By analyzing the signal growth and step change, signal features are extracted. By combining the weighted fusion of resistance gradient map and capacitance change map, the location and parameter information of cracks are identified. By analyzing the coupling correlation between material stress intensity and crack propagation, the evolution of crack conditions is predicted, resulting in accurate crack identification results. Parallel excitation and synchronous acquisition of conductive fiber bundles can improve signal acquisition efficiency, remove environmental interference signals from the signal data, effectively reduce the influence of interference factors, and enable the signal data to effectively reflect the crack condition. By predicting the evolution of crack parameter information and propagation, a basis for early maintenance and safety warning of bridges can be provided, which helps to take timely measures to prevent further crack propagation. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating the process of an intelligent electrical monitoring method for bridge cracks in an embodiment of this application.

[0047] Figure 2 This is a schematic diagram of the resistance gradient diagram and capacitance change diagram in the embodiments of this application;

[0048] Figure 3 This is a schematic diagram of the crack propagation process in an embodiment of this application;

[0049] Figure 4 This is a schematic diagram of the structure of a smart electrical monitoring system for bridge cracks according to an embodiment of this application. Detailed Implementation

[0050] The present application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0051] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0052] Hereinafter, the terms "first," "second," and other generic terms are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0053] refer to Figure 1 The image shows a specific implementation of a smart electrical monitoring method for bridge cracks according to this application, comprising:

[0054] S101. The electrical signals of the bridge structure are collected in real time through a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and a micro sensor is embedded at the interlacing point of the conductive fiber bundle.

[0055] S102. Based on the first signal set, analyze the environmental interference signal in conjunction with the pre-acquired environmental data, and decouple the first signal set through the environmental interference signal to obtain the second signal set.

[0056] S103. Analyze the signal growth and step change of the second signal set, extract the corresponding signal features, and identify the crack edge to obtain the crack edge identification result.

[0057] S104. Based on the crack edge identification results, the evolution prediction of crack propagation is performed using a preset material crack coupling model to obtain crack parameter prediction results, so as to intelligently monitor bridge cracks.

[0058] Current methods for monitoring bridge cracks suffer from low signal acquisition efficiency, poor accuracy, limited monitoring range, and a lack of crack prediction, failing to meet the needs of bridge safety monitoring. This embodiment addresses this issue by leveraging the characteristic that changes in the electrical signals of conductive materials reflect structural damage. It utilizes multilayer conductive fiber bundles and micro-sensors to acquire bridge electrical signals in real time. By combining environmental data with decoupling from environmental interference signals, accurate signal data reflecting crack information is obtained. Analysis of signal growth and abrupt changes identifies crack edges, and a material crack coupling model is used to predict crack propagation, resulting in accurate crack identification. The use of conductive materials for signal acquisition improves efficiency, enabling real-time automatic monitoring of crack data. Decoupling from environmental interference data and crack feature analysis reduces environmental interference in the signal acquisition process, improving signal data accuracy. This allows for accurate crack identification, and predicting crack propagation trends provides early warnings, facilitating timely bridge maintenance and preventing structural safety issues caused by crack propagation.

[0059] In this embodiment, the electrical signals of the bridge structure are collected in real time using a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and embeds micro sensors at the interlacing points of the conductive fiber bundle. The conductive fiber bundle is composed of multiple interlaced conductive fibers. The conductivity of the conductive fiber bundle is related to the structural integrity of the fiber bundle. When cracks appear in the bridge structure, the cracks will cause the fiber bundle to be locally stressed and stretched or broken, causing the resistance of the fiber bundle to change, and the corresponding electrical signal will change accordingly. The micro sensors embedded at the interlacing points can collect the changes in electrical signals in real time. The multi-layered interlaced conductive fiber bundle can fully cover the bridge structure, avoid monitoring blind spots, and improve the integrity and accuracy of signal collection. The micro sensors have high sensitivity and can collect minute changes in electrical signals. The collected first signal set can be used to analyze the structural cracks of the bridge in real time and monitor the bridge cracks in real time.

[0060] Specifically, based on the first signal set, environmental interference signals are analyzed in conjunction with pre-acquired environmental data. The first signal set is then decoupled using these environmental interference signals to obtain a second signal set. Environmental factors, including temperature, humidity, and wind, can affect the electrical properties of conductive fiber bundles. Increased temperature increases fiber resistance, leading to signal changes in the first signal set caused by factors other than cracks. The correlation between environmental data and signals in the first signal set is analyzed, and environmental interference signals are separated and filtered out through decoupling. Removing environmental interference signals eliminates the interference of environmental factors on monitoring signals, improves signal accuracy, accurately identifies signal changes caused by crack changes, reduces misjudgments, and improves the accuracy of crack identification results.

[0061] The signal growth and step change of the second signal set are analyzed to extract the corresponding signal features and identify the crack edges, thus obtaining crack edge identification results. When cracks exist in the bridge structure, the conductive fiber bundles will be damaged, and the electrical signal at the corresponding location will continue to grow or undergo step changes. The signal changes of the second signal set are analyzed to identify the corresponding crack edge positions and shapes, thus obtaining crack edge identification results. Through signal feature analysis, the crack edges can be accurately located, improving the efficiency and accuracy of crack edge identification results.

[0062] Meanwhile, based on the crack edge identification results, the evolution of crack propagation is predicted using a pre-set material crack coupling model, resulting in crack parameter predictions. The material crack coupling model predicts crack evolution based on the mechanical properties of the bridge material and the initial state of the crack. Mechanical properties include tensile strength and elastic modulus, while the initial state of the crack includes crack width and depth. By analyzing the evolution of crack propagation using the material crack coupling model, the changes in crack parameters can be predicted. By predicting crack propagation trends, early warnings can be given regarding the structural safety of the bridge, allowing for timely structural maintenance and ensuring the bridge's safety and stability.

[0063] This application utilizes conductive fiber bundles with multi-layered structures and embedded micro-sensors at their interlacing points for signal acquisition. By simultaneously exciting and acquiring multiple conductive fiber bundles, signal data is collected. Environmental interference signals in the signal data are decoupled to obtain accurate signal data. By analyzing signal growth and step abrupt changes, signal features are extracted. Combined with weighted fusion of resistance gradient maps and capacitance abrupt change maps, the location and parameter information of cracks are identified. By analyzing the coupling correlation between material stress intensity and crack propagation, the evolution of crack conditions is predicted, resulting in accurate crack identification results. The parallel excitation and synchronous acquisition of conductive fiber bundles improves signal acquisition efficiency, removes environmental interference signals from the signal data, and effectively reduces the influence of interference factors, enabling the signal data to effectively reflect the crack condition. By predicting the evolution of crack parameter information and propagation, a basis for early maintenance and safety warning of bridges is provided, facilitating timely measures to prevent further crack propagation.

[0064] Furthermore, the electrical signals of the bridge structure are acquired in real time through a pre-set conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layered structure and embeds miniature sensors at the interlacing points of the conductive fiber bundle, including:

[0065] S201. Set up a preset conductive fiber bundle on the bridge for signal acquisition. The conductive fiber bundle includes a graphene-coated fiber layer, a polyurethane matrix layer, and a carbon fiber bundle layer. Embed a micro sensor at the interlacing point of the conductive fiber bundle.

[0066] S202. By simultaneously exciting multiple conductive fiber bundles, the electrical signals are synchronously acquired in real time to obtain the first signal set.

[0067] This embodiment utilizes the synergistic conductivity of multilayer conductive materials, interweaving graphene-coated fiber layers, polyurethane matrix layers, and carbon fiber bundle layers. Micro-sensors are embedded at the interweaving points to obtain conductive fiber bundles. These conductive fiber bundles are deployed to key parts of the bridge, converting the mechanical deformation caused by bridge cracks into detectable electrical signal changes. By parallel excitation and synchronous acquisition of multiple conductive fiber bundles, the electrical signals of each bundle are collected in real time, resulting in a first signal set. The combination of the multilayer structure and parallel excitation allows for simultaneous acquisition of bridge structural signal data. Compared to the traditional method of time-division signal data acquisition for each fiber bundle separately, this scheme shortens signal acquisition time and improves data acquisition efficiency. Furthermore, even if one layer of the conductive fiber bundle is damaged, the other layers can continue to function normally, improving the stability of signal acquisition.

[0068] In this embodiment, a pre-set conductive fiber bundle is placed on the bridge for signal acquisition. The conductive fiber bundle includes a graphene-coated fiber layer, a polyurethane matrix layer, and a carbon fiber bundle layer. Miniature sensors are embedded at the interlacing points of the conductive fiber bundle. The multi-layered structure of the conductive fiber bundle works synergistically. The graphene-coated fiber layer utilizes the high conductivity and sensitivity of graphene to enhance the response to electrical signals from minute deformations. The polyurethane matrix layer has good flexibility and adhesion, which can uniformly transfer the deformation of the bridge structure to the conductive fibers. The carbon fiber bundle layer serves as the core load-bearing structure, ensuring that the conductive fiber bundle is not easily broken during long-term use. Simultaneously providing basic conductivity; graphene-coated fiber layers, polyurethane matrix layers, and carbon fiber bundle layers are composited through a hot-pressing process, and miniature piezoresistive sensors are embedded at the interlacing points of the conductive fiber bundles to obtain conductive fiber bundles; according to the bridge structure, conductive fiber bundles are deployed in key parts such as beams and piers to collect signals; the multi-layer structure of conductive fiber bundles can improve the accuracy and stability of the signal acquisition process, and placing sensors at the interlacing points can directly contact different fiber layers, enabling the simultaneous acquisition of multi-dimensional electrical signals, avoiding misjudgments caused by signal distortion from a single fiber layer, and improving the accuracy of the acquired signal data.

[0069] Specifically, by simultaneously exciting multiple conductive fiber bundles, the electrical signals are synchronously acquired in real time to obtain a first signal set. Parallel excitation refers to applying a specific frequency excitation voltage to different conductive fiber bundles simultaneously using a multi-channel signal generator. Utilizing the independent circuit loops of each conductive fiber bundle, multi-channel electrical signals are synchronously acquired. Bridge cracks cause local resistance changes in the conductive fiber bundles, and cracks at different locations will cause corresponding changes in the response signals of the corresponding fiber bundles. Synchronous acquisition can distinguish the crack states in different areas, improving signal acquisition efficiency. Parallel excitation allows multiple sets of fiber bundles to work simultaneously, improving signal acquisition efficiency while ensuring the temporal consistency of signals acquired from different areas, thus guaranteeing the spatial distribution of data during crack analysis.

[0070] Furthermore, by simultaneously exciting multiple conductive fiber bundles and synchronously acquiring electrical signals in real time, a first signal set is obtained, including:

[0071] S301. Assign orthogonal carrier frequencies to multiple conductive fiber bundles and apply corresponding AC excitation signals to synchronously acquire the response signals of each conductive fiber bundle;

[0072] S302. The response signal of each conductive fiber bundle is separated by Fourier transform to obtain the first signal set.

[0073] In this embodiment, orthogonal carrier frequencies are assigned to multiple conductive fiber bundles, and corresponding AC excitation signals are applied. The response signals of each conductive fiber bundle are acquired synchronously. Orthogonal carrier frequencies refer to carrier signals that are orthogonal to each other. After assigning a corresponding orthogonal carrier frequency to each conductive fiber bundle, a corresponding AC excitation signal is applied. The carrier frequency interval is set according to the Nyquist sampling theorem and signal bandwidth requirements to determine the corresponding orthogonal carrier frequency for each conductive fiber bundle. A multi-channel signal generator is used to set the carrier frequency of each channel according to the corresponding orthogonal carrier frequencies, and an AC excitation signal is applied to each conductive fiber bundle using the multi-channel signal generator. A summarizing signal acquisition unit is connected to the output end of all conductive fiber bundles to acquire the response signals of multiple conductive fiber bundles. Orthogonal carrier frequency allocation avoids crosstalk between signals from multiple fiber bundles. Synchronous application of excitation and acquisition signals allows for the simultaneous acquisition of signal data from multiple conductive fiber bundles, ensuring the temporal consistency of signals from different conductive fiber bundles and improving signal acquisition efficiency.

[0074] Specifically, the response signals of each conductive fiber bundle are separated using Fourier transform to obtain a first signal set. Fourier transform converts the composite response signal in the time domain to the frequency domain. In the frequency domain, each orthogonal carrier frequency corresponds to an independent signal peak. By identifying the peak amplitude or phase at different frequencies, the response signal corresponding to each conductive fiber bundle can be separated. Based on the pre-assigned orthogonal carrier frequencies for each conductive fiber bundle, the amplitude and phase information corresponding to each frequency are found. The amplitude data of each conductive fiber bundle at the corresponding carrier frequency is extracted, and the separated response signal sequences of each conductive fiber bundle are arranged in chronological order to obtain the first signal set. Frequency domain separation of the signals using Fourier transform can eliminate interference between different conductive fiber bundles and improve the quality of the crack signal.

[0075] Furthermore, based on the first signal set, and combined with pre-acquired environmental data, environmental interference signals are analyzed. The first signal set is then decoupled using these environmental interference signals to obtain a second signal set, which includes:

[0076] S401. Calculate the reference value of each signal based on the first signal set to obtain the environmental reference signal set;

[0077] S402. By combining the environmental reference signal set and the pre-acquired environmental data, the signal residual is analyzed to obtain the environmental interference signal;

[0078] S403. Decouple and separate the environmental interference signal from the first signal set to obtain the second signal set.

[0079] In this embodiment, data in the crack-free state are selected from the first signal set, the baseline value of each signal is calculated, and an environmental baseline signal set is constructed. By analyzing the correlation between the signal residual and the environmental data, the signal residual is calculated to obtain the environmental interference signal. The environmental interference signal is removed from the first signal set, and the second signal set is obtained by decoupling. After decoupling, the signal fluctuations caused by environmental factors can be removed, which can more accurately reflect the crack information and improve the accuracy of crack identification results.

[0080] In this embodiment, a reference value for each signal is calculated based on the first signal set to obtain an environmental reference signal set. The data of the first signal set when the bridge has just completed the installation of conductive fiber bundles and is free of cracks is selected. The average value of each signal of each conductive fiber bundle in the first signal set data is calculated. The environmental reference signal set is constructed by integrating each conductive fiber bundle. The environmental reference signal set can reflect the signal characteristics in the crack-free state, providing a reference for distinguishing between signal changes caused by environmental interference and cracks, and reducing misjudgment.

[0081] Specifically, by combining the environmental reference signal set and the pre-acquired environmental data, the signal residual is analyzed to obtain the environmental interference signal. The signal residual refers to the difference between the signal value in the first signal set and the corresponding reference value in the environmental reference signal set. There is a correlation between the environmental data and the signal residual. By analyzing the correlation, the signal change caused by environmental factors is calculated to obtain the environmental interference signal. By screening out environmental factors through correlation analysis, the efficiency and accuracy of environmental interference signal calculation can be improved.

[0082] The environmental interference signal is decoupled and separated from the first signal set to obtain the second signal set. The corresponding environmental interference signal value is subtracted from the first signal set according to the position of the conductive fiber bundle to eliminate the influence of environmental factors. The resulting second signal set eliminates environmental interference and only reflects the changes in electrical signals caused by bridge cracks, thereby improving the accuracy of bridge crack identification and analysis.

[0083] Furthermore, by combining the environmental reference signal set and the pre-acquired environmental data, the signal residuals are analyzed to obtain the environmental interference signals, including:

[0084] S501. Obtain the first residual set by calculating the difference between the environmental reference signal set and the corresponding signal in the first signal set;

[0085] S502. Combining the first residual set and the pre-acquired environmental data, the signal residual of environmental interference is calculated through a preset residual analysis model to obtain the environmental interference signal.

[0086] In this embodiment, a first residual set is obtained by calculating the difference between the environmental reference signal set and the corresponding signal in the first signal set. The first residual set is the set of differences between the environmental reference signal set and the corresponding signal in the first signal set. The environmental reference signal set represents the electrical signal reference corresponding to different environmental conditions under crack-free conditions, while the first signal set includes signal changes caused by environmental interference and cracks. By calculating the difference between the two, a comprehensive residual including the effects of environmental interference and cracks is obtained. Information of each signal is extracted from the first signal set, and the corresponding reference value is matched in the environmental reference signal set. The residual value of the signal is calculated according to the corresponding position to obtain the first residual set. The residual data reflects the degree of signal deviation from the reference, providing data support for distinguishing between normal environmental fluctuations and abnormal changes caused by cracks.

[0087] Specifically, by combining the first residual set and the pre-acquired environmental data, the signal residual of environmental interference is calculated using a preset residual analysis model to obtain the environmental interference signal. The residual analysis model includes, but is not limited to, a multivariate regression model. The multivariate regression model is trained using a large amount of environmental data to obtain a pre-trained multivariate regression model. The pre-acquired environmental data is input into the pre-trained multivariate regression model, and the model calculates the signal residual of environmental interference. The signal residual corresponding to each conductive fiber bundle is calculated using the pre-trained multivariate regression model. The calculated signal residuals are integrated with the first residual set according to the conductive fiber bundles to obtain the environmental interference signal. The residual analysis model can accurately calculate the residual value of the environmental interference signal, improving the accuracy of the analyzed environmental interference signal.

[0088] Furthermore, the signal growth and step abrupt changes in the second signal set are analyzed to extract the corresponding signal features and identify the crack edges, yielding crack edge identification results, including:

[0089] S601. Analyze the signal growth and step change of the second signal set, extract the corresponding signal features, and identify the crack region.

[0090] S602. In the crack region, fit the edge pixels, analyze the crack angle and crack width, and calculate the crack length based on the crack region size.

[0091] S603. Combining the crack angle, crack width, and crack length, the crack edge identification result is obtained.

[0092] This embodiment locates the crack region based on the electrical signal characteristics caused by the crack. It analyzes the signal growth and step change of the second signal set, extracts the corresponding features to identify the crack region, fits the edge pixels within the crack region, and converts the electrical signal data into crack angle, width and length. It integrates crack parameter information to obtain crack edge identification results. Through signal feature analysis and fitting, accurate crack edge identification results can be obtained. It can not only identify whether a crack exists, but also calculate detailed crack parameter information to carry out corresponding maintenance measures for the crack.

[0093] In this embodiment, the signal growth and step change of the second signal set are analyzed to extract the corresponding signal features and identify the crack region. Environmental interference has been removed from the second signal set, which only includes the electrical signal changes caused by the crack. The expansion of the crack will cause damage to the conductive fiber bundle, and the corresponding signal will continue to grow. The sudden expansion of the crack will cause the conductive fiber bundle to break or stretch instantly, resulting in a step change in the signal. By analyzing the signal growth and step change, the region with abnormal signal can be located and the crack region can be identified. By analyzing the signal growth and step change, different types of cracks can be screened out, avoiding missed detections and improving the efficiency and accuracy of crack identification.

[0094] Specifically, within the crack region, edge pixels are fitted to analyze the crack angle and width, and the crack length is calculated based on the crack region size. Within the identified crack region, signal changes in conductive fiber bundles at different locations reflect the crack's morphological characteristics. The fiber bundle signal changes are more pronounced at the crack edges. Within the crack region, the coordinates of the fiber bundle with the largest signal change are extracted as the edge pixel positions. The edge pixel positions and angles are fitted using the least squares method to obtain the fitted edges. The crack width and length are calculated based on the area formed by the fitted edges. Accurate crack parameter information can be identified through edge fitting, and accurate crack parameter data can provide accurate data references for the maintenance of bridge cracks.

[0095] Specifically, by combining crack angle, crack width, and crack length, crack edge identification results are obtained. Crack angle reflects the direction of crack, while width and length reflect the area of ​​crack. Combining the three can reflect the specific shape and distribution of cracks on the bridge structure. By integrating the identification results of multiple parameters, crack characteristics can be comprehensively reflected, avoiding the limitations of single-parameter description, and providing accurate data support for assessing the degree of damage caused by cracks and formulating maintenance plans.

[0096] Furthermore, the signal growth and step abrupt changes in the second signal set were analyzed to extract corresponding signal features and identify the crack region, including:

[0097] S701. Analyze the resistance change rate and capacitance step change point in the second signal set, and extract the change range and step change point set where the resistance change rate is greater than the preset resistance change threshold.

[0098] S702. According to the structure of the bridge, map the change range and the set of step change points to the corresponding positions of the bridge.

[0099] S703. Generate a resistance gradient map based on the variation range, and generate a capacitance change map based on the set of step change points.

[0100] S704. Weighted fusion of the resistance gradient map and the capacitance abrupt change map to obtain the fused gradient map;

[0101] S705. Identify the crack region from the fused gradient map using a preset crack edge recognition model.

[0102] In this embodiment, the resistance change rate and capacitance step change points in the second signal set are analyzed to extract the change range where the resistance change rate is greater than a preset resistance change threshold and the set of step change points. The resistance and capacitance characteristics of the conductive fiber bundle change accordingly with the generation and expansion of bridge cracks. When the crack expands slowly, the contact area of ​​the conductive fiber bundle gradually decreases, and the corresponding resistance continues to increase, causing the resistance change rate to exceed the normal range. When the crack suddenly expands, the structure of the conductive fiber bundle undergoes abrupt changes, and the capacitance exhibits a step change. The resistance change rate is obtained by calculating the resistance difference between two adjacent time points in chronological order and dividing it by the time interval. A resistance change threshold is set based on the material characteristics of the conductive fiber bundle and the maximum change rate under stable bridge conditions. Regions with resistance change rates greater than the resistance change threshold are selected as change ranges. The capacitance difference between adjacent time points is calculated, and a step change threshold is set based on the capacitance fluctuation range. Points where the capacitance difference is greater than the step change threshold are selected as the set of step change points. Combining resistance and capacitance characteristics for analysis allows for the identification of cracks in different modes, reducing misjudgments and omissions from single-parameter analysis.

[0103] Specifically, according to the bridge's structure, the variation range and the set of abrupt change points are mapped to the corresponding locations on the bridge. The placement of each conductive fiber bundle on the bridge is predetermined, and the resistance variation range and the set of abrupt change points are associated with the corresponding conductive fiber bundles. Based on the mapping relationship between each conductive fiber bundle and the bridge location, the selected variation range and the set of abrupt change points are mapped to the corresponding physical locations on the bridge, achieving a precise correspondence between electrical signal anomalies and the physical locations of the bridge, providing a spatial coordinate basis for subsequent image generation and crack area identification.

[0104] like Figure 2As shown, a resistance gradient map is generated based on the variation range, and a capacitance mutation map is generated based on the set of step abrupt change points. The resistance gradient map uses color intensity to show the magnitude of the resistance change rate at different locations on the bridge; the darker the color, the greater the resistance change rate, and the greater the likelihood of a crack. The thickness of the lines in the map indicates the color intensity; thicker lines indicate darker colors, and thinner lines indicate lighter colors. The capacitance mutation map uses special triangle markers to mark the locations of capacitance step abrupt change points, and the size of the markers reflects the magnitude of the change. By constructing the resistance gradient map and the capacitance mutation map, areas with concentrated abnormal signals can be quickly and intuitively identified, providing a reference for the identification of crack areas.

[0105] Specifically, the resistance gradient map and capacitance abrupt change map are weighted and fused to obtain a fused gradient map. While the resistance gradient map and capacitance abrupt change map reflect crack characteristics from different perspectives, a single image has limitations. The resistance gradient map lags behind in responding to suddenly appearing cracks, and the capacitance abrupt change map struggles to reflect slowly expanding cracks. Based on historical data statistics, the accuracy of crack identification using the resistance change rate and capacitance abrupt change points is analyzed. Appropriate weights are set based on these accuracy rates. The resistance gradient map and capacitance abrupt change map are converted into numerical matrices according to pixel values. The pixel values ​​at corresponding positions are weighted and summed to obtain a fusion matrix. The fused gradient map is constructed based on the corresponding element values ​​of the fusion matrix. This weighted fusion combines the advantages of both the resistance gradient map and the capacitance abrupt change map. The fused gradient map highlights the comprehensive characteristics of signal anomalies, improving the accuracy of crack region identification.

[0106] Based on the fused gradient map, a pre-defined crack edge recognition model is used to identify crack regions from the fused gradient map. The crack edge recognition model includes, but is not limited to, the U-Net model based on deep learning. The U-Net model is trained using a large number of annotated crack images, including the fused gradient map and actual crack regions, to obtain a pre-trained U-Net model. The fused gradient map is then input into the pre-trained U-Net model, which identifies crack edges and identifies the regions formed by the crack edges as crack regions. The deep learning model can quickly identify crack regions, improving the efficiency and accuracy of crack region recognition.

[0107] Furthermore, based on the crack edge identification results, the evolution of crack propagation is predicted using a pre-defined material crack coupling model to obtain crack parameter prediction results, enabling intelligent monitoring of bridge cracks, including:

[0108] S801. Extract crack parameter information from the crack edge identification results and construct a spatiotemporal feature vector by combining it with environmental data;

[0109] S802. Based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain the first prediction result.

[0110] S803. Correct the first prediction result by the real-time crack width to obtain the crack parameter prediction result.

[0111] like Figure 3 As shown, cracks extend and expand over time. This embodiment predicts crack propagation trends based on the initial state of the crack, historical environmental influences, and spatial structural characteristics through a material crack coupling model. Crack parameters are extracted from crack edge identification results, and a spatiotemporal feature vector is constructed by combining environmental data. The spatiotemporal feature vector is then input into a preset material crack coupling model to predict the evolution of crack propagation, resulting in a first prediction result. This first prediction result is then corrected by combining the real-time crack width to obtain the crack parameter prediction result. By predicting the evolution of crack propagation, the time when cracks exceed safety thresholds can be analyzed, providing a safety warning for the bridge structure and enabling timely maintenance of the bridge structure.

[0112] In this embodiment, crack parameter information is extracted from the crack edge identification results and combined with environmental data to construct a spatiotemporal feature vector. The expansion of a crack is influenced by both its own parameters and environmental factors. The crack parameters include, but are not limited to, initial width, length, and angle, while the environmental factors include, but are not limited to, temperature, humidity, and load. Corresponding parameter information is extracted from the crack edge identification results and environmental data, respectively, and the extracted parameters are arranged in chronological order to construct a spatiotemporal feature vector. By constructing a spatiotemporal feature vector, the crack's own parameters, historical environmental influences, and spatial structural characteristics are integrated, providing comprehensive feature information for the material crack coupling model and improving the accuracy of the prediction results.

[0113] Specifically, based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a pre-defined material crack coupling model to obtain the first prediction result. The material crack coupling model is a multi-physics coupling model based on fracture mechanics and materials science. It can simulate the crack propagation process under the action of internal material stress, environmental erosion, and external loads. The spatiotemporal feature vector is input into the material crack coupling model, and the model predicts the evolution of crack propagation and the changes in crack parameters to obtain the first prediction result. The material crack coupling model is based on physical mechanisms and can reflect the law of crack propagation, obtaining accurate crack evolution prediction results, which provides a basis for safety early warning of bridge structures.

[0114] Specifically, the first prediction result is corrected by real-time crack width to obtain crack parameter prediction results. Since the first prediction result may deviate from the actual situation, the crack width is monitored in real-time using the electrical signal of the conductive fiber bundle. The difference between the real-time crack width and the predicted crack width in the first prediction result is calculated. A difference threshold is set according to the crack prediction accuracy requirements. When the difference exceeds the threshold, the first prediction result is corrected to obtain crack parameter prediction results. Correcting the first prediction result based on the real-time crack width can eliminate the deviation between the prediction result and the actual situation, improve the accuracy of the prediction result, and enable the prediction result to reflect the real-time crack state, thus providing timely and accurate early warning of bridge cracks.

[0115] Furthermore, based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a pre-defined material crack coupling model, yielding a first prediction result, including:

[0116] S901. Based on the spatiotemporal feature vector, the coupling relationship between material stress intensity and crack propagation is analyzed through a preset material crack coupling model to obtain the crack propagation rate.

[0117] S902. According to the preset time period, the evolution of crack propagation is predicted by the crack propagation rate to obtain the first prediction result.

[0118] In this embodiment, based on the spatiotemporal feature vector, the coupling relationship between material stress intensity and crack propagation is analyzed using a preset material crack coupling model to obtain the crack propagation rate. There is a coupling relationship between material stress intensity and crack propagation; when the stress intensity factor at the crack tip exceeds the fracture toughness of the material, the crack begins to propagate, and the larger the stress intensity factor value, the faster the crack propagation rate. The preset material crack coupling model calculates the stress intensity factor at different times using structural parameters, environmental data, and load data from the spatiotemporal feature vector, and then calculates the crack propagation rate by combining this with material properties. Crack parameters are extracted from the spatiotemporal feature vector, and the stress intensity factor is calculated according to the fracture mechanics formula, which is:

[0119] ;

[0120] In the formula, Stress intensity factor For shape factor, For load, The cross-sectional area where the crack is located. The crack depth is defined as follows: The stress intensity factor is calculated using fracture mechanics formulas. The fracture toughness of the material is determined based on the concrete strength grade in the spatiotemporal characteristic vector. When the stress intensity factor is less than the fracture toughness, the crack does not propagate; when the stress intensity factor is greater than or equal to the fracture toughness, the crack propagates. The crack propagation rate is obtained by dividing the stress intensity by the fracture toughness and then weighting the results. By combining the coupling relationship between stress intensity and crack propagation, the crack propagation rate can be accurately calculated, thus predicting the accurate crack propagation situation. This allows for the analysis and prediction of crack development trends, yielding accurate crack prediction results.

[0121] Specifically, according to a preset time period, the evolution of crack propagation is predicted by the crack propagation rate to obtain a first prediction result. Crack propagation accumulates over time. The crack propagation amount for each time period is calculated within the preset time period, and the crack propagation amount for each time period is accumulated to obtain the bridge crack prediction result for the corresponding time period. By accumulating the propagation over time periods, cracks can be gradually accumulated, the accurate crack propagation situation can be analyzed, and the accuracy of the first prediction result can be improved.

[0122] like Figure 4 As shown, a smart electrical monitoring system for bridge cracks is used to implement a smart electrical monitoring method for bridge cracks, comprising:

[0123] The signal acquisition module acquires electrical signals of the bridge structure in real time through a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and embeds micro sensors at the interlacing points of the conductive fiber bundle.

[0124] The signal processing module analyzes environmental interference signals based on the first signal set and pre-acquired environmental data, and decouples the first signal set through the environmental interference signals to obtain the second signal set.

[0125] The crack identification module analyzes the signal growth and step change of the second signal set, extracts the corresponding signal features, and identifies the crack edges to obtain crack edge identification results.

[0126] The crack prediction module, based on the crack edge identification results, uses a preset material crack coupling model to predict the evolution of crack propagation and obtain crack parameter prediction results for intelligent monitoring of bridge cracks.

[0127] In this embodiment, the signal acquisition module collects electrical signals from the bridge structure in real time using conductive fiber bundles containing multi-layered structures and embedded micro-sensors at the interlacing points, forming a first signal set. Through the combination of the multi-layered structure and micro-sensors, signals can be collected comprehensively and accurately, covering key areas of the bridge and capturing electrical signal changes caused by tiny cracks, providing accurate signal data for crack monitoring and analysis. The signal processing module analyzes environmental interference signals based on the first signal set and pre-acquired environmental data, and decouples the first signal set using the environmental interference signals to obtain a second signal set. This effectively eliminates the interference of environmental factors such as temperature and humidity on the electrical signals, making the second signal set more accurately reflect the signal changes caused by cracks, improving the accuracy of crack identification results and reducing misjudgments caused by environmental interference.

[0128] Specifically, the crack identification module analyzes the growth and abrupt changes of signals in the second signal set, extracts signal features, and identifies crack edges, thus obtaining crack edge identification results. Through precise analysis of signal features, it can accurately locate crack edges, distinguish crack propagation patterns, and improve the efficiency and accuracy of crack identification. Based on the crack edge identification results, the crack prediction module uses a preset material crack coupling model to predict the evolution of crack propagation, obtaining crack parameter prediction results. By predicting the changing trends of parameters such as crack width and depth in advance, it provides trend references for bridge maintenance, avoids accidents caused by crack expansion, and improves the timeliness and effectiveness of bridge monitoring.

[0129] The above description is merely a preferred embodiment of this application. The scope of protection of this application is not limited to the above embodiments. All technical solutions falling within the scope of this application's concept are within the scope of protection of this application. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of this application should also be considered within the scope of protection of this application.

Claims

1. A method for intelligent electrical monitoring of bridge cracks, characterized in that, include: The electrical signals of the bridge structure are collected in real time by a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and a micro sensor is embedded at the interlacing point of the conductive fiber bundle. Based on the first signal set, environmental interference signals are analyzed in conjunction with pre-acquired environmental data. The first signal set is then decoupled using the environmental interference signals to obtain a second signal set. The signal growth and step change of the second signal set are analyzed, the corresponding signal features are extracted, and the crack edges are identified to obtain the crack edge identification results. Based on the crack edge identification results, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain crack parameter prediction results, thereby enabling intelligent monitoring of bridge cracks.

2. The intelligent electrical monitoring method for bridge cracks according to claim 1, characterized in that, The method involves real-time acquisition of electrical signals from the bridge structure via a pre-set conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layered structure with miniature sensors embedded at the interlacing points. Signal acquisition is performed by setting up a pre-defined conductive fiber bundle on the bridge. The conductive fiber bundle includes a graphene-coated fiber layer, a polyurethane matrix layer, and a carbon fiber bundle layer. A micro-sensor is embedded at the interlacing point of the conductive fiber bundle. By simultaneously exciting multiple conductive fiber bundles, the electrical signals are synchronously acquired in real time to obtain the first signal set.

3. The intelligent electrical monitoring method for bridge cracks according to claim 2, characterized in that, The first signal set is obtained by simultaneously exciting multiple conductive fiber bundles in parallel and synchronously acquiring electrical signals in real time, including: Multiple conductive fiber bundles are assigned orthogonal carrier frequencies and corresponding AC excitation signals are applied to synchronously acquire the response signals of each conductive fiber bundle; The response signals of each conductive fiber bundle are separated by Fourier transform to obtain the first signal set.

4. The intelligent electrical monitoring method for bridge cracks according to claim 1, characterized in that, Based on the first signal set, and combined with pre-acquired environmental data, environmental interference signals are analyzed. The first signal set is then decoupled using these environmental interference signals to obtain a second signal set, including: The reference value of each signal is calculated based on the first signal set to obtain the environmental reference signal set; By combining the aforementioned environmental reference signal set and the pre-acquired environmental data, the signal residuals are analyzed to obtain the environmental interference signal; The environmental interference signal is decoupled and separated from the first signal set to obtain the second signal set.

5. The intelligent electrical monitoring method for bridge cracks according to claim 4, characterized in that, By combining the aforementioned environmental reference signal set and pre-acquired environmental data, the signal residuals are analyzed to obtain environmental interference signals, including: The first residual set is obtained by calculating the difference between the environmental reference signal set and the corresponding signal in the first signal set; By combining the first residual set and the pre-acquired environmental data, the signal residual of environmental interference is calculated using a preset residual analysis model to obtain the environmental interference signal.

6. The intelligent electrical monitoring method for bridge cracks according to claim 1, characterized in that, The signal growth and step abrupt changes of the second signal set are analyzed, the corresponding signal features are extracted, and the crack edges are identified to obtain the crack edge identification results, including: The signal growth and step change of the second signal set were analyzed, and the corresponding signal features were extracted to identify the crack region. In the crack region, edge pixels are fitted, crack angle and crack width are analyzed, and crack length is calculated based on crack region size; By combining the crack angle, crack width, and crack length, the crack edge identification result is obtained.

7. The intelligent electrical monitoring method for bridge cracks according to claim 6, characterized in that, The analysis of signal growth and step abrupt changes in the second signal set, extracting corresponding signal features to identify the crack region, includes: Analyze the resistance change rate and capacitance step change point in the second signal set to extract the change range and step change point set where the resistance change rate is greater than the preset resistance change threshold. According to the structure of the bridge, the set of change intervals and step change points are mapped to the corresponding positions of the bridge; A resistance gradient map is generated based on the change range, and a capacitance mutation map is generated based on the set of step mutation points. The resistance gradient map and the capacitance abrupt change map are weighted and fused to obtain a fused gradient map; The crack region is identified from the fused gradient map using a preset crack edge recognition model.

8. The intelligent electrical monitoring method for bridge cracks according to claim 1, characterized in that, Based on the crack edge identification results, the propagation of the crack is predicted using a preset material crack coupling model to obtain crack parameter prediction results, thereby enabling intelligent monitoring of bridge cracks, including: Crack parameter information is extracted from the crack edge identification results and combined with environmental data to construct a spatiotemporal feature vector; Based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain a first prediction result; The first prediction result is corrected by adjusting the crack width in real time to obtain the crack parameter prediction result.

9. The intelligent electrical monitoring method for bridge cracks according to claim 8, characterized in that, Based on the spatiotemporal feature vector, the evolution of crack propagation is predicted using a preset material crack coupling model to obtain a first prediction result, including: Based on the spatiotemporal feature vector, the coupling relationship between material stress intensity and crack propagation is analyzed through a pre-defined material crack coupling model to obtain the crack propagation rate. According to a preset time period, the evolution of crack propagation is predicted based on the crack propagation rate to obtain a first prediction result.

10. A smart electrical monitoring system for bridge cracks, characterized in that, A method for implementing a smart electrical monitoring method for bridge cracks as described in any one of claims 1 to 9 includes: The signal acquisition module acquires electrical signals of the bridge structure in real time through a preset conductive fiber bundle to obtain a first signal set. The conductive fiber bundle includes a multi-layer structure and embeds micro sensors at the interlacing points of the conductive fiber bundle. The signal processing module analyzes environmental interference signals based on the first signal set and pre-acquired environmental data, and decouples the first signal set through the environmental interference signals to obtain a second signal set. The crack identification module analyzes the signal growth and step change of the second signal set, extracts the corresponding signal features, and identifies the crack edge to obtain the crack edge identification result. The crack prediction module, based on the crack edge identification results, uses a preset material crack coupling model to predict the evolution of crack propagation and obtain crack parameter prediction results for intelligent monitoring of bridge cracks.

Images