Bridge expansion device anchoring structure detection method, system, medium and equipment

Abstract

The invention relates to a bridge expansion device anchoring structure detection method and system, a medium and equipment. The method comprises the following steps: acquiring power sensing data of a standard component under different working conditions; training a bridge expansion device anchoring structure reference dynamic feature denoising neural network; collecting power sensing data of the anchoring structure of the to-be-detected bridge expansion device, and obtaining power characteristic data of the anchoring structure of the to-be-detected bridge expansion device through preprocessing and neural network analysis; calculating a difference coefficient; and determining the dynamic characteristic state of the anchoring structure of the to-be-detected bridge expansion device according to the difference coefficient. The method has the advantages that the representative working condition area is divided, the sub working conditions are arranged, the working condition state of the anchoring structure of the bridge expansion device is comprehensively evaluated in a multi-face mode, and the evaluation result is more accurate.

Description

Technical Field

[0001] The present invention belongs to the field of bridge detection, and particularly relates to a method, system, medium and device for detecting the anchoring structure of a bridge expansion device. Background Technique

[0002] As one of the important components of a bridge structure, the bridge expansion device is the last installation item in bridge construction, and its good performance and service status determine the safety and comfort of vehicles passing over the bridge. The bridge expansion device is arranged between two adjacent beam ends, between the beam end and the abutment, or at the hinge position of the bridge to meet the beam end deformation caused by factors such as temperature change, concrete shrinkage, various load actions, and deflection displacement during the service of the bridge, and at the same time ensure that the vehicles driving on the bridge deck pass smoothly.

[0003] However, the bridge expansion device is also one of the weak links in the bridge structure. Affected by comprehensive factors such as design, manufacturing and installation, environmental climate, and long-term heavy traffic loads, the damage problem of the bridge expansion device is very prominent, which is one of the key problems in current bridge operation and maintenance and has attracted wide attention from research scholars.

[0004] The main technical routes for research related to bridge expansion devices are as follows: 1) By studying the structure and material properties of the bridge expansion device, optimizing the structure, process design, production and installation schemes of the expansion device, and improving the construction and installation efficiency. 2) Studying the performance of the concrete interface (transition zone) directly combined with the expansion device, and by improving the concrete performance, enhancing the performance and durability of the transition zone of the bridge expansion device, and also improving the repair efficiency of the bridge expansion joint; 3) Technologies represented by SHM (structural health monitoring) have gradually become an advanced means for safety monitoring and fatigue life prediction during the service of bridges. By forming an SHM system with sensors and analysis modules, data collection, processing and analysis are carried out on the key monitoring / detection parts of the bridge, improving the dimension of data analysis and providing data support for subsequent bridge maintenance decisions.

[0005] Although there are rich research results in the field related to bridge expansion devices, there are still certain limitations:

[0006] (1) Most of the existing studies only conduct one-sided research on the expansion device or the concrete transition zone, that is, analyzing the performance of the anchoring area of the bridge expansion device through a single variable model. However, the anchoring structure of the bridge expansion device should be regarded as an integral part under the combined anchoring action of the expansion device, concrete and adjacent interfaces. The existing performance analysis methods or models are difficult to output accurate evaluation results of the anchoring structure of the bridge expansion device, thus affecting the efficiency and accuracy of subsequent bridge maintenance decisions.

[0007] (2) At present, most of the methods for evaluating the anchorage structure state of bridge expansion devices still rely on manual regular on-site visual inspections, which have the following deficiencies: First, there is a lack of in-depth detection and disease analysis of the state of the anchorage area of bridge expansion devices; second, on-site inspections of bridge expansion devices often require traffic interruption, and the efficiency of manual operations and data analysis and processing is extremely low, seriously affecting vehicle passing efficiency; third, regular manual inspections often have a lag, making it difficult to prevent sudden failures of the anchorage structure of bridge expansion devices.

[0008] (3) Although the SHM system is advanced and powerful, its price is extremely expensive, and the professional requirements for operators and data processors are very high, so its popularity still needs to be improved. Moreover, the SHM system is extremely dependent on the data collected by sensors. The entire bridge structure is complex, and in complex environments such as vehicle loads, wind forces, and vibrations, the data collected by the deployed sensors will contain a large amount of noise, seriously affecting subsequent data analysis. Existing research on bridge monitoring data processing only "refines" the specified type of data through specific algorithms, fails to effectively remove noise, and lacks applicability. Summary of the Invention

[0009] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a detection method, system, medium, and device for the anchorage structure of bridge expansion devices.

[0010] In the first aspect, a detection method for the anchorage structure of bridge expansion devices is provided, including:

[0011] S1. Prefabricate standard components for the anchorage structure of bridge expansion devices and obtain dynamic sensing data of the standard components under different working conditions;

[0012] S2. Based on the dynamic sensing data of the standard components under different working conditions, train a denoising neural network for the reference dynamic characteristics of the anchorage structure of bridge expansion devices;

[0013] S3. Collect the dynamic sensing data of the anchorage structure of the bridge expansion device to be detected, and obtain the dynamic characteristic data of the anchorage structure of the bridge expansion device to be detected through preprocessing and neural network analysis;

[0014] S4. Calculate the difference coefficient between the dynamic characteristic data of the anchorage structure of the bridge expansion device to be detected and the dynamic sensing data of the standard components under different working conditions; and determine the dynamic characteristic state of the anchorage structure of the bridge expansion device to be detected according to the difference coefficient.

[0015] Preferably, S1 includes:

[0016] S101. Prefabricate standard components for the anchorage structure of bridge expansion devices, and the standard components include: bridge expansion devices, concrete, and bridge slabs;

[0017] S102. Divide the working condition areas of the standard parts. The working condition areas include: the working condition area of the bridge expansion device, the concrete working condition area, the interface working condition area between the bridge expansion device and the concrete, and the interface working condition area between the concrete and the end of the bridge slab. Sensors are embedded in the concrete working condition area.

[0018] S103. Set multiple sub - working condition states for each working condition area.

[0019] S104. Collect the dynamic sensing data of the standard parts in each sub - working condition state through the sensors.

[0020] Preferably, in S103, setting multiple sub - working condition states for each working condition area includes:

[0021] S1031. Set the initial intact state, intact to yield state, yield state, yield to fracture state, and fracture state for the working condition area of the bridge expansion device.

[0022] S1032. Set the concrete dense state, the first cavity state, the second cavity state, the third cavity state, and the fourth cavity state for the concrete working condition area. The corresponding cavity ratios of the first cavity state, the second cavity state, the third cavity state, and the fourth cavity state increase in sequence.

[0023] S1033. Set the completely bonded state, the first detachment state, the second detachment state, the third detachment state, and the completely detached state for the interface working condition area between the bridge expansion device and the concrete. The degree of detachment of the first detachment state, the second detachment state, and the third detachment state increases in sequence.

[0024] S1034. Set the completely bonded state, the fourth detachment state, the fifth detachment state, the sixth detachment state, and the seventh detachment state for the interface working condition area between the concrete and the end of the bridge slab. The degree of detachment of the fourth detachment state, the fifth detachment state, the sixth detachment state, and the seventh detachment state increases in sequence.

[0025] Preferably, S2 includes:

[0026] S201. Pre - process the dynamic sensing data of the standard parts under different working conditions. The pre - processing includes: adding Gaussian noise.

[0027] S202. Input the dynamic sensing data with added Gaussian noise into the Transformer neural network.

[0028] S203. Calculate the loss function according to the output of the Transformer neural network.

[0029] S204. Iteratively update the neural network parameters by minimizing the loss function, and repeat S201 - S203 until the loss error converges to a preset value to obtain the target neural network.

[0030] Preferably, in S4, the working condition of the standard part corresponding to the minimum value of the difference coefficient is used as the dynamic characteristic state of the anchoring structure of the bridge expansion device to be detected.

[0031] In a second aspect, a detection system for the anchoring structure of a bridge expansion device is provided, which is used to execute the method described in any one of the first aspects, and includes:

[0032] An acquisition module, configured to prefabricate a standard part of the anchoring structure of a bridge expansion device and acquire the dynamic sensing data of the standard part under different working conditions;

[0033] A training module, configured to train a reference dynamic characteristic denoising neural network for the anchoring structure of a bridge expansion device based on the dynamic sensing data of the standard part under different working conditions;

[0034] A collection module, configured to collect the dynamic sensing data of the anchoring structure of the bridge expansion device to be detected, and obtain the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected through preprocessing and neural network analysis.

[0035] A calculation module, configured to calculate the difference coefficient between the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected and the dynamic sensing data of the standard part under different working conditions; and determine the dynamic characteristic state of the anchoring structure of the bridge expansion device to be detected according to the difference coefficient.

[0036] In a third aspect, a computer storage medium is provided, in which a computer program is stored; when the computer program runs on a computer, the computer is enabled to execute the method described in any one of the first aspects.

[0037] In a fourth aspect, an electronic device is provided, including:

[0038] A memory, configured to store a computer program;

[0039] A processor, configured to execute the computer program to implement the method described in any one of the first aspects.

[0040] The beneficial effects of the present invention are:

[0041] 1. According to the anchoring structure technology, structure and construction characteristics of the expansion device of a newly built bridge, standard components (which can be scaled to appropriate sizes) are prefabricated. The standard components are divided into representative working condition areas, and sensors are installed on the standard components and the anchoring structure of the newly built bridge expansion device synchronously to ensure that the newly built structure can collect detection data in real time during its service life. The detection data can also be compared with the different sub-working condition states set in each working condition area of the standard component, realizing real-time state detection of the anchoring structure of the bridge expansion device during its service life, being able to promptly reflect the service state, and the detection process does not require traffic interruption.

[0042] 2. Compared with most current studies that only focus on single items such as bridge expansion devices or concrete transition zones, the solution of the present invention comprehensively evaluates the working condition state of the anchoring structure of the bridge expansion device from multiple aspects by dividing representative working condition areas and setting multiple sub-working conditions, and the evaluation results are more accurate.

[0043] 3. There are very few current studies on introducing deep learning technology to evaluate the service state of the anchoring structure of bridge expansion devices. The present invention proposes a denoising neural network for the benchmark dynamic characteristics of the anchoring structure of bridge expansion devices. During the training process, by learning the deep characteristics of the dynamic data of each working condition area and each sub-working condition state, the network is enabled to have the ability to analyze the input data and obtain the corresponding characteristic data of the anchoring structure of the bridge expansion device. Secondly, real noise data is introduced into the training technical route of the denoising network, and the network is trained to generate data with the same characteristics. After the trained denoising network receives the actual detection data, it generates noise data, and then the noise-free state data of the anchoring structure of the bridge expansion device is obtained through denoising calculation, greatly reducing the analysis difficulty of bridge facility decision-making and maintenance, and solving the problems that the existing data refinement technology is prone to disturbing the original data and has poor denoising effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flowchart of a detection method for the anchoring structure of a bridge expansion device provided by the present invention;

[0045] Figure 2 It is a schematic structural diagram of a standard component of the anchoring structure of a bridge expansion device provided by the present invention;

[0046] Figure 3 It is a schematic diagram of a working condition area division of a standard component of the anchoring structure of a bridge expansion device provided by the present invention;

[0047] Figure 4 It is another schematic diagram of a working condition area division of a standard component of the anchoring structure of a bridge expansion device provided by the present invention;

[0048] Figure 5 It is a schematic structural diagram of a detection system for the anchoring structure of a bridge expansion device provided by the present invention;

[0049] Description of the reference numerals: bridge plate 1, concrete 2, bridge expansion device 3, working condition area A of the bridge expansion device, working condition area B of the concrete, working condition area C of the interface between the bridge expansion device and the concrete, and working condition area D of the interface between the concrete and the end of the bridge plate. Detailed implementation manners

[0050] The present invention will be further described below in conjunction with embodiments. The description of the following embodiments is only for helping to understand the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

[0051] Embodiment 1:

[0052] To solve the problems of the prior art, Embodiment 1 of the present application provides a detection method for the anchoring structure of a bridge expansion device to realize real-time detection of the state of the anchoring structure of the bridge expansion device under the condition of non-interrupted traffic. Specifically, as Figure 1 shown, the method includes:

[0053] S1. Fabricate a standard component for the anchoring structure of the bridge expansion device, and obtain the dynamic sensing data of the standard component under different working conditions.

[0054] S1 includes:

[0055] S101. Fabricate a standard component for the anchoring structure of the bridge expansion device according to the structural style and process of the anchoring structure of the expansion device of the bridge to be detected. Among them, as Figure 2 shown, the standard component includes: bridge expansion device 3, concrete 2, and bridge plate 1.

[0056] S102. Divide the working condition areas of the standard component. As Figure 3 and Figure 4 shown, the working condition areas include: working condition area A of the bridge expansion device, working condition area B of the concrete, working condition area C of the interface between the bridge expansion device and the concrete, and working condition area D of the interface between the concrete and the end of the bridge plate; the working condition area B of the concrete is embedded with sensors. The sensors can be paired fiber optic sensors.

[0057] S103. Set multiple sub-working condition states for each working condition area.

[0058] Exemplarily, the sub-working condition states of the working condition area are denoted as A {a} 、B {b} 、C {c} 、D {d}, respectively representing: the working condition area A of the bridge expansion device, the fatigue state of the bridge expansion device; the working condition area B of the concrete, the cavity defect state of the concrete; the working condition area C of the interface between the bridge expansion device and the concrete, the disengaged state of the interface between the bridge expansion device and the concrete; the working condition area D of the interface between the concrete and the end of the bridge slab, the disengaged state of the interface between the concrete and the end of the bridge slab.

[0059] Specifically, in S103, setting multiple sub - working condition states for each working condition area includes:

[0060] S1031. Setting the initial intact state, intact - to - yield state, yield state, yield - to - fracture state, and fracture state for the working condition area A of the bridge expansion device.

[0061] For example, the fatigue state {a} of the bridge expansion device ∈ {0, 0.25, 0.5, 0.75, 1}, respectively representing the initial intact state, intact - to - yield state, yield state, yield - to - fracture state, and fracture state of the bridge expansion device.

[0062] S1032. Setting the dense state of the concrete, the first cavity state, the second cavity state, the third cavity state, and the fourth cavity state for the working condition area B of the concrete; the cavity ratios corresponding to the first cavity state, the second cavity state, the third cavity state, and the fourth cavity state increase in sequence.

[0063] For example, the cavity defect state {b} of the concrete ∈ {0, 0.15, 0.25, 0.35, 0.5}, respectively representing the dense state of the concrete, the 15% cavity state, the 25% cavity state, the 35% cavity state, and the 50% cavity state.

[0064] S1033. Setting the fully bonded state of the interface between the bridge expansion device and the concrete, the first disengaged state, the second disengaged state, the third disengaged state, and the fully disengaged state for the working condition area C of the interface between the bridge expansion device and the concrete; the degree of disengagement of the first disengaged state, the second disengaged state, and the third disengaged state increases in sequence.

[0065] For example, the disengaged state {c} of the interface between the bridge expansion device and the concrete ∈ {0, 0.25, 0.5, 0.75, 1}, respectively representing the fully bonded state of the interface between the bridge expansion device and the concrete, the 25% disengaged state, the 50% disengaged state, the 75% disengaged state, and the fully disengaged state.

[0066] S1034. Setting the fully bonded state of the interface between the concrete and the end of the bridge slab, the fourth disengaged state, the fifth disengaged state, the sixth disengaged state, and the seventh disengaged state for the working condition area D of the interface between the concrete and the end of the bridge slab; the degree of disengagement of the fourth disengaged state, the fifth disengaged state, the sixth disengaged state, and the seventh disengaged state increases in sequence.

[0067] For example, the detachment state {d} of the interface between the concrete and the end of the bridge slab belongs to {0, 0.15, 0.25, 0.35, 0.5}, which respectively represent the fully bonded state of the interface between the concrete and the end of the bridge slab, the 15% detachment state, the 25% detachment state, the 35% detachment state, and the 50% detachment state.

[0068] In S103, according to the set working condition areas and the corresponding working condition states, permutations and combinations are carried out. Therefore, there are a total of 5 4 standard parts made in this embodiment (four working condition areas, with 5 states in each working condition area). The serial number of a single standard part is denoted as i, then i ∈ [1, 2, 3......5 4 .

[0069] S104. Collect the dynamic sensing data of the standard part in each sub - working condition state through the sensor.

[0070] Specifically, when making the standard part, a pair of sensors are set. One of them is used to generate a pulse signal with a specific frequency, and the pulse energy is transmitted inside the standard part. Then the other sensor collects the signal, that is, data acquisition. The collected data is the dynamic sensing data of the anchoring structure standard part of the bridge expansion device. It should be noted that the dynamic sensing data collected by the sensor is the first - order natural frequency of the anchoring structure of the bridge expansion device. The first - order natural frequency is an inherent property of the rigid body as a whole; the first - order natural frequency changes with the changes in the rigidity of the object itself and the constraint conditions (boundary conditions). The anchoring structure of the bridge expansion device should be regarded as an integral under the combined anchoring action of the expansion device, concrete, and adjacent interfaces. Considering the illustration Figure 4 inside and the actual installation and construction method of the structural components, the sensor actually collects the first - order natural frequency data of the concrete working condition area B. When the working condition of any working condition area changes, the collected data will also change accordingly (when A, C, D change, that is, the constraint conditions of B change, then the natural frequency data of B changes; when there are voids / damages inside B, that is, the rigidity of B changes, the natural frequency of B changes). The sub - working condition situations set in the solution of the present invention enumerate all permutation and combination situations, so they include the data situations when each sub - working condition takes effect alone and the data situations when they take effect mutually.

[0071] The dynamic sensing data collected by each standard part is denoted as: x i [A {a} 、B {b} 、C {c} 、D {d} .

[0072] Specifically, the dynamic sensing data of the first standard part is denoted as: x1[A {0} 、B {0} 、C {0} 、D {0}, representing the dynamic sensing data of the standard components of the bridge expansion device anchoring structure under the initial intact state of the bridge expansion device, the dense state of the concrete, the completely bonded state of the interface between the bridge expansion device and the concrete, and the completely bonded state of the interface between the concrete and the end of the bridge slab;... and so on by analogy, the dynamic sensing data of the last component is denoted as: x 625 [A {1} 、B {0.5} 、C {01} 、D {0.5} , representing the dynamic sensing data of the standard components of the bridge expansion device anchoring structure under the fractured state of the bridge expansion device, the 50% state of the concrete cavity, the completely separated state of the interface between the bridge expansion device and the concrete, and the 50% separated state of the interface between the concrete and the end of the bridge slab.

[0073] Specifically, the dynamic sensing data collected by the i-th standard component is X i , i ∈ [1, 2, 3......5 4 ; The set of all X i is the reference dynamic characteristic data; In this example, 5 4 standard components are manufactured.

[0074] S2. Based on the dynamic sensing data of the standard components under different working conditions, train the denoising neural network for the reference dynamic characteristics of the bridge expansion device anchoring structure.

[0075] S3. Collect the dynamic sensing data of the bridge expansion device anchoring structure to be detected, and obtain the dynamic characteristic data of the bridge expansion device anchoring structure to be detected through preprocessing and neural network analysis.

[0076] S4. Calculate the difference coefficient between the dynamic characteristic data of the bridge expansion device anchoring structure to be detected and the dynamic sensing data of the standard components under different working conditions; and determine the dynamic characteristic state of the bridge expansion device anchoring structure to be detected according to the difference coefficient.

[0077] Embodiment 2:

[0078] Based on Embodiment 1, Embodiment 2 of the present application provides a more specific detection method for the bridge expansion device anchoring structure, including:

[0079] S1. Prefabricate the standard components of the bridge expansion device anchoring structure, and obtain the dynamic sensing data of the standard components under different working conditions.

[0080] S2. Based on the dynamic sensing data of the standard components under different working conditions, train the denoising neural network for the reference dynamic characteristics of the bridge expansion device anchoring structure.

[0081] S2 includes:

[0082] S201. Preprocess the dynamic sensing data of the standard part under different working conditions (denoted as X0), and the preprocessing includes: adding Gaussian noise.

[0083] Specifically, add Gaussian noise to X0 for t times to obtain data X t , and the formula is as follows:

[0084] , Equation (1);

[0085] In Equation (1): X t represents the data obtained by adding Gaussian noise to X0 for t times; t is a random positive integer, t ∈ (1, 2, 3......T], T is a large positive integer; e is a random sampling noise value subject to the standard normal distribution N(0, 1); set the weight coefficient: 0.99 ≤ α t ≤ 0.99999; .

[0086] S202. Input the dynamic sensing data after adding Gaussian noise into the Transformer neural network.

[0087] Specifically, input (X t , t) into the Transformer neural network, and the neural network outputs ε θ (X t , t). Among them, the role of the neural network is to receive the data of the actual component to be measured in the subsequent S3, and then generate data with certain distribution characteristics (the network has this ability after training).

[0088] S203. Calculate the loss function according to the output of the Transformer neural network.

[0089] The loss function is expressed as:

[0090] loss = ||e - ε θ (X t , t)|| 2

[0091] In the formula, e is used as the true value, and ε θ (X t , t) is used as the predicted value.

[0092] S204. Iteratively update the neural network parameters by minimizing the loss function, and repeat S201 - S203 until the loss error converges to a preset value to obtain the target neural network.

[0093] Among them, the target neural network is the denoising neural network N for the benchmark dynamic characteristics of the anchor structure of the bridge expansion device.

[0094] Specifically, the preset value can be set to a small positive number, which is set to 0.0000001 in this example.

[0095] Specifically, a benchmark denoising network is obtained by training the data of a standard part, denoted as N i , i ∈ [1, 2, 3......5 4 ; Therefore, 5 denoising networks are obtained by training in this example. 4 denoising networks.

[0096] Specifically, when training a single denoising network N i , the original data X0 will first be added with noise e for t times, and the noise added each time is randomly sampled and follows a standard normal distribution. Therefore, when training a denoising network N i , first generate 1 integer t from (1, 2, 3......T], then sample t different noises e, and simultaneously generate t different α terms (0.99 ≤ α t ≤ 0.99999); The T for training each N i can be fixed or independently set. In this example, for the convenience of calculation, T is uniformly set to 2000.

[0097] Specifically, according to Equation (1) of S201, taking X0 as the initial data, after adding the first noise e1 and weight coefficient α1 for the first time (t1), the data X1 is obtained, expressed as ; Similarly, X2 is added with the second noise e2 and weight coefficient α2 for the second time (t2) to obtain X3, expressed as ;...... And so on, the data X obtained at the t-th time t is the data X of the previous time (t - 1) t-1 added with the t-th noise data e t and the corresponding weight coefficient α t obtained; When the number of times t of adding noise is large enough, the obtained data X t follows a standard normal distribution.

[0098] S3. Collect the dynamic sensing data of the anchoring structure of the bridge expansion device to be detected, and obtain the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected through preprocessing and neural network analysis.

[0099] Exemplarily, the dynamic sensing data of the anchoring structure of the bridge expansion device to be detected is denoted as G, and the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected is denoted as H. S3 includes:

[0100] S301. Perform data preprocessing on the dynamic sensing data G. The initial data G is denoted as G0, and the processed data is denoted as G t , and the preprocessing method is as follows:

[0101] , formula (2);

[0102] Determined according to the method described in S2: the number of times t, the weight coefficient term β, the random noise value E that follows the standard normal distribution N(0, 1), and G is calculated. t .

[0103] S302. Denoise and analyze G t for i rounds.

[0104] Specifically, in the i-th round, the i-th denoising network N i is used to denoise and analyze G t . The data obtained by analyzing a single N i is denoted as H i ;

[0105] Denote G t as H0, which is used as the initial data for the analysis process. Specifically, the analysis steps for each round are as follows: Input H0 into the denoising network N i , and the data output by N i is E θ (H t , t), where θ represents the neural network parameters. The number of times t and the weight coefficient term β are determined by step S31, and then H t-1 is calculated according to the following formula:

[0106] , formula (3);

[0107] Input H t-1 into N i again, and calculate t times according to formula (3) in a loop to obtain H t . Then the data obtained by analyzing a single N i is H t,i , denoted as H i , and the set of all H i is H.

[0108] S4. Calculate the difference coefficient between the dynamic characteristic data of the anchor structure of the bridge expansion device to be detected and the dynamic sensing data of the standard part under different working conditions; and determine the dynamic characteristic state of the anchor structure of the bridge expansion device to be detected according to the difference coefficient.

[0109] In S4, calculate the difference between each actual dynamic characteristic data H i and the reference dynamic characteristic data X i , and analyze the dynamic characteristic state of the actual anchor structure of the bridge expansion device to be detected according to the difference.

[0110] Specifically, S4 includes:

[0111] S401. Calculate the difference according to the following formula (4) to obtain i difference coefficients:

[0112] , formula (4);

[0113] In the formula, the difference coefficient K i represents the difference between the i-th actual dynamic characteristic data H i and the i-th reference dynamic characteristic data X i . P(X i ) and Q(H i ) respectively represent the data distributions of the reference dynamic characteristic data X i and the dynamic characteristic data H of the actual detected anchorage structure of the bridge expansion device i .

[0114] S402. Calculate min{K i}, that is, the minimum value of the i difference coefficients, and the i-th group of the standard part working conditions corresponding thereto is the actual working condition of the actual detected anchorage structure of the bridge expansion device.

[0115] For example: when the serial number i corresponding to the minimum value of the calculated difference coefficient corresponds to the first group of working conditions of the standard part, at this time there are [A {0} , B {0} , C {0} , D {0} , that is, for the anchorage structure of the bridge expansion device: the bridge expansion device is in good condition, the concrete area is in a dense state, the interface between the bridge expansion device and the concrete is completely bonded, and the interface between the concrete and the end of the bridge slab is completely bonded.

[0116] It should be noted that the parts that are the same as or similar to those in Embodiment 1 in this embodiment can be referred to each other and will not be elaborated in this application.

[0117] Embodiment 3:

[0118] Based on Embodiment 2, Embodiment 3 of the present application provides a detection system for the anchorage structure of a bridge expansion device, including:

[0119] An acquisition module, configured to prefabricate a standard part of the anchorage structure of the bridge expansion device and acquire the dynamic sensing data of the standard part under different working conditions.

[0120] Specifically, in this solution, the prefabricated standard part is provided with an optical fiber sensor. When laid, the sensor body is inside the test piece, and the sensor leaves transmission lines and other wire ends outside the test piece for connecting to an external acquisition module. Similarly, for the newly built anchorage structure of the bridge expansion device, according to this invention solution, the sensor is laid at the corresponding position during the installation construction. During the service period, the sensor wire ends can be connected to the acquisition module of this invention system.

[0121] A training module, configured to train a denoising neural network for the benchmark dynamic characteristics of the anchoring structure of the bridge expansion device based on the dynamic sensing data of the standard part under different working conditions.

[0122] A data acquisition module, configured to acquire the dynamic sensing data of the anchoring structure of the bridge expansion device to be detected, and obtain the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected through preprocessing and neural network analysis.

[0123] A calculation module, configured to calculate the difference coefficient between the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected and the dynamic sensing data of the standard part under different working conditions; and determine the dynamic characteristic state of the anchoring structure of the bridge expansion device to be detected according to the difference coefficient.

[0124] It should be noted that the system provided in this embodiment is the system corresponding to the method provided in Embodiment 2. Therefore, for the parts that are the same or similar in this embodiment and Embodiment 2, reference can be made to each other, and details will not be described again in this application.

Claims

1. A detection method for the anchoring structure of a bridge expansion device, characterized in that Including: S1. Fabricate the standard components of the anchoring structure for bridge expansion devices, and obtain the dynamic sensing data of the standard components under different working conditions; S2. Based on the dynamic sensing data of the standard components under different working conditions, train the denoising neural network for the benchmark dynamic characteristics of the anchoring structure of bridge expansion devices; S3. Collect the dynamic sensing data of the anchoring structure of the bridge expansion device to be detected, and obtain the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected through preprocessing and neural network analysis; S4. Calculate the difference coefficient between the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected and the dynamic sensing data of the standard components under different working conditions; and determine the dynamic characteristic state of the anchoring structure of the bridge expansion device to be detected according to the difference coefficient.

2. The inspection method for the anchorage structure of the bridge expansion device according to claim 1, characterized in that S1 includes: S101. Fabricate the standard components of the anchoring structure for bridge expansion devices, and the standard components include: bridge expansion devices, concrete, and bridge slabs; S102. Divide the working condition areas of the standard components, and the working condition areas include: the working condition area of the bridge expansion device, the working condition area of the concrete, the interface working condition area between the bridge expansion device and the concrete, and the interface working condition area between the concrete and the end of the bridge slab; sensors are embedded in the working condition area of the concrete; S103. Set multiple sub - working condition states for each working condition area; S104. Collect the dynamic sensing data of the standard components in each sub - working condition state through the sensors.

3. The inspection method for the anchoring structure of the bridge expansion device according to claim 2, characterized in that, In S103, setting multiple sub - working condition states for each working condition area includes: S1031. Set the initial intact state, intact to yield state, yield state, yield to fracture state, and fracture state for the working condition area of the bridge expansion device; S1032. Set the concrete dense state, the first cavity state, the second cavity state, the third cavity state, and the fourth cavity state for the working condition area of the concrete; the cavity ratios corresponding to the first cavity state, the second cavity state, the third cavity state, and the fourth cavity state increase in sequence; S1033. Set the completely bonded state, the first detachment state, the second detachment state, the third detachment state, and the completely detached state for the interface working condition area between the bridge expansion device and the concrete; the degree of detachment of the first detachment state, the second detachment state, and the third detachment state increases in sequence; S1034. Set the completely bonded state, the fourth detachment state, the fifth detachment state, the sixth detachment state, and the seventh detachment state for the interface working condition area between the concrete and the end of the bridge slab; the degree of detachment of the fourth detachment state, the fifth detachment state, the sixth detachment state, and the seventh detachment state increases in sequence.

4. The inspection method for the anchoring structure of the bridge expansion device according to claim 3, characterized in that, S2 includes: S201. Preprocess the dynamic sensing data of the standard components under different working conditions, and the preprocessing includes: adding Gaussian noise; S202. Input the dynamic sensing data with added Gaussian noise into the Transformer neural network; S203. Calculate the loss function according to the output of the Transformer neural network; S204. Iteratively update the neural network parameters by minimizing the loss function, and repeat S201 - S203 until the loss error converges to a preset value to obtain the target neural network.

5. The detection method for the anchoring structure of the bridge expansion device according to claim 4, characterized in that, In S4, the working condition of the standard part corresponding to the minimum value of the difference coefficient is used as the dynamic characteristic state of the anchoring structure of the bridge expansion device to be detected.

6. A detection system for the anchoring structure of a bridge expansion device, characterized in that For implementing the method according to any one of claims 1 to 5, comprising: An acquisition module, configured to prefabricate a standard part of the anchoring structure of the bridge expansion device and acquire the dynamic sensing data of the standard part under different working conditions; A training module, configured to train a denoising neural network for the benchmark dynamic characteristics of the anchoring structure of the bridge expansion device based on the dynamic sensing data of the standard part under different working conditions; A collection module, configured to collect the dynamic sensing data of the anchoring structure of the bridge expansion device to be detected, and obtain the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected through preprocessing and neural network analysis; A calculation module, configured to calculate the difference coefficient between the dynamic characteristic data of the anchoring structure of the bridge expansion device to be detected and the dynamic sensing data of the standard part under different working conditions; and determine the dynamic characteristic state of the anchoring structure of the bridge expansion device to be detected according to the difference coefficient.

7. A computer storage medium, characterized in that, The computer storage medium stores a computer program; when the computer program runs on a computer, the computer is enabled to execute the method according to any one of claims 1 to 5.

8. An electronic device, characterized in that, Comprising: A memory, configured to store the computer program; A processor, configured to execute the computer program to implement the method according to any one of claims 1 to 5.

Images