A method for identifying the axle load of heavy vehicles based on the distribution of bridge deflection influence lines

Abstract

The present application discloses a method for identifying the axle weight of heavy vehicles based on the distribution of the influence line of the bridge deflection, comprising the following steps: setting up a monitoring device for vehicle parameter identification on the bridge deck, arranging a structural displacement detection device in the middle of the span of the bridge to identify the mid-span deflection of the bridge; conducting a field sports car test to obtain the influence lines of small heavy vehicles of different models and weights on the mid-span deflection of the bridge; performing simulation amplification based on the influence line data of the small heavy vehicles; constructing a heavy vehicle axle weight identification model and training the model through the influence line data corresponding to the obtained heavy vehicles; inputting the data of the target heavy vehicle when crossing the bridge into the heavy vehicle axle weight identification model to obtain the axle weight data of the target heavy vehicle. The beneficial effects of the present application: by estimating the vehicle axle weight using visual recognition on the bridge and deflection calculation under the bridge, the impact of the device installation on the bridge deck traffic can be avoided; at the same time, the recyclable structural displacement detection device can effectively reduce the investment cost of the equipment.

Description

Technical Field

[0001] The present application relates to the technical field of bridges, and particularly to a method for identifying the axle weight of heavy vehicles based on the distribution of bridge deflection influence lines. Background Art

[0002] The dynamic weighing system of vehicle loads can be divided into the pavement dynamic weighing system (PWIM) and the bridge dynamic weighing system (BWIM). Among them, for the bridge dynamic weighing system, sensors are usually installed at the bottom of the bridge to collect structural responses (such as strain gauges, accelerometers, fiber optic sensors, etc.), and relevant identification algorithms (such as the Moses algorithm) are used to calculate the axle weight of the driving vehicle.

[0003] To ensure the measurement accuracy and coverage, it is usually necessary to install multiple sensors at multiple positions on the bridge, which will increase the complexity and cost of the system. Especially on large bridges, a large number of sensors and cable connections are required, which may lead to problems in wiring and power management. Over time, traditional contact sensors will experience performance degradation due to long-term exposure to the external environment, resulting in measurement errors or complete failure; and the construction of the system will affect the normal traffic operation.

[0004] At the same time, the vehicle load information that can be obtained only from the bridge deflection response at the algorithm level is very limited, which results in less overweight vehicle load and corresponding bridge deflection data in actual situations, and cannot meet the requirements of deep learning model inference applications. And the inference of the deep learning model is easily affected by noise and abnormal data in the input data, causing a large error in the accuracy of the result. Summary of the Invention

[0005] One object of the present application is to provide a method for identifying the axle weight of heavy vehicles based on the distribution of bridge deflection influence lines, which can solve at least one defect in the above background art.

[0006] To achieve at least one of the above objects, the technical solution adopted in the present application is: a method for identifying the axle weight of heavy vehicles based on the distribution of bridge deflection influence lines, including the following steps:

[0007] S100: Install a monitoring device for vehicle parameter identification on the bridge deck, and at the same time arrange a structural displacement detection device at the mid-span of the bridge to identify the mid-span deflection of the bridge;

[0008] S200: Conduct on-site vehicle running tests on the bridge using small heavy vehicles of different vehicle types and weights to obtain the influence lines of small heavy vehicles of different vehicle types and weights on the mid-span deflection of the bridge;

[0009] S300: Perform simulation amplification based on the influence line data of small heavy vehicles to obtain the influence line simulation data of the mid-span deflection of the bridge corresponding to large heavy vehicles;

[0010] S400: Construct a heavy vehicle axle weight identification model and train the model with the influence line data corresponding to all heavy vehicles obtained;

[0011] S500: Substitute the vehicle type of the target heavy vehicle when crossing the bridge and the influence line data of the mid-span deflection of the bridge generated into the trained heavy vehicle axle weight identification model, and then inversely obtain the axle weight data of the target heavy vehicle.

[0012] Preferably, the specific process of vehicle parameter identification in step S100 by the monitoring device is as follows:

[0013] S110: Use the monitoring device to perform image tracking and shooting on the vehicles driving on the bridge deck;

[0014] S120: Obtain the type of the target vehicle according to the captured image, and calculate the vehicle speed of the target vehicle according to the initial position and calibration position demarcated on the bridge deck;

[0015] S130: According to the calculated vehicle speed of the target vehicle, intercept the images of the target vehicle at at least one position in the original image data to identify the number of axles and axle distance of the target vehicle;

[0016] S140: Construct a parameter set of the target vehicle regarding license plate, vehicle type, number of axles and axle distance.

[0017] Preferably, the monitoring device includes a traffic flow camera and a lidar; during the day, the monitoring device performs camera image tracking and shooting on the vehicles through the traffic flow camera; at night, the monitoring device performs collaborative tracking and shooting of camera images and point cloud images on the vehicles through the traffic flow camera and the lidar.

[0018] Preferably, the structural displacement detection device includes a target and a corner reflector installed at the mid-span position of the bridge main girder, and an industrial camera and a millimeter wave radar installed at the pier position; the industrial camera monitors the displacement of the target to obtain the mid-span deflection of the bridge, and the millimeter wave radar monitors the phase difference of the electromagnetic wave signals of the corner reflector to obtain the mid-span deflection of the bridge.

[0019] Preferably, when performing step S200, a plurality of structural displacement detection devices for monitoring the mid-span deflection of the bridge are provided and arranged in sequence according to the main girders corresponding to the bridge; the transverse deflection ratio coefficient η under different lanes and vehicle weights is calculated based on the mid-span deflections of the main girders of each lane obtained from the on-site vehicle running test; after completing step S200, some of the structural displacement detection devices are removed, and based on the mid-span deflections of the corresponding main girders obtained by the remaining structural displacement detection devices, the mid-span deflections of the remaining main girders are calculated through the obtained transverse deflection ratio coefficient η.

[0020] Preferably, the transverse deflection ratio coefficient η corresponding to the i-th main girder i is calculated by the formula: ;

[0021] where ω i represents the mid-span deflection of the i-th main girder, represents the sum of the mid-span deflections of the n main girders corresponding to the bridge.

[0022] Preferably, when disassembling the structural displacement detection device after completing step S200, at least two structural displacement detection devices are left. The process of judging the transverse connection stiffness between the main girders through the transverse ratio coefficients corresponding to the remaining structural displacement detection devices is as follows: Set the transverse connection stiffness judgment formula: (ω x / ω y ) = (η x / η y ); If the transverse connection stiffness judgment formula holds, the transverse connection stiffness between the main girders corresponding to the bridge meets the requirements; otherwise, it is determined that the transverse connection stiffness between the main girders corresponding to the bridge does not meet the requirements, and the transverse deflection ratio coefficients of each main girder are recalibrated through a single vehicle type; where ω x and ω y respectively represent the mid-span deflections of the main girders corresponding to two of the remaining structural displacement detection devices, and η x and η y respectively represent the transverse deflection ratio coefficients of the main girders corresponding to two of the remaining structural displacement detection devices.

[0023] Preferably, step S300 includes the following process:

[0024] S310: Construct a finite element model of the bridge;

[0025] S320: Use the measured deflection influence line data and corresponding vehicle information of the small heavy vehicle running test to correct the stiffness of the finite element model of the bridge;

[0026] S330: Apply the load corresponding to the heavy vehicle to the corrected bridge finite element model, and then obtain the simulation data of the influence line of the mid-span deflection of the bridge corresponding to the heavy vehicle loading.

[0027] Preferably, step S400 includes the following specific processes:

[0028] S410: Construct the deep learning model architecture of the heavy vehicle axle load identification model, and select the root mean square error as the loss function L driven by the model data. data ;

[0029] S420: Embed the bridge finite element model with completed stiffness correction into the deep learning model training process;

[0030] S430: Input the given bridge deflection influence line data into the deep learning model to generate the corresponding axle load vector and axle distance information;

[0031] S440: Convert the predicted axle load vector and axle distance information into the load of the bridge finite element model and perform structural simulation to obtain the corresponding bridge deflection influence line simulation data;

[0032] S450: Take the error between the given bridge deflection influence line data and the bridge deflection influence line simulation data as the physical information loss function L of the deep learning model. phy ;

[0033] S460: Take the loss functions L data and L phy as the optimization objects of the deep learning model, and perform parameter backpropagation to iterate the model parameters.

[0034] Preferably, step S500 includes the following specific processes:

[0035] S510: When the monitoring device detects that the target heavy vehicle is on the bridge, trigger the bridge structure displacement detection device to calculate the mid-span deflection of the bridge;

[0036] S520: Use signal processing technology to remove the structural dynamic response and environmental noise components in the measured deflection influence line data to obtain the bridge deflection influence line data at the measured points;

[0037] S530: Estimate the bridge deflection influence line at other mid-span points of the bridge according to the calculated deflection transverse proportional coefficient and the bridge deflection influence line data at the measured points;

[0038] S540: After the target heavy vehicle crosses the bridge, input the bridge deflection influence line data at all points into the vehicle axle load identification model to estimate each axle load.

[0039] Compared with the prior art, the beneficial effects of this application are as follows:

[0040] By means of a heavy vehicle axle load identification method based on the distribution of bridge deflection influence lines, the axle load of the vehicle is estimated by using on-bridge visual recognition and off-bridge deflection calculation. Compared with the traditional method, it can avoid the impact of device installation on bridge traffic, and at the same time, the recoverable structural displacement detection device can effectively reduce the equipment input cost. Brief Description of the Drawings

[0041] Figure 1 It is a schematic diagram of the overall work flow of this application.

[0042] Figure 2 It is a schematic diagram of the construction process of the vehicle axle load identification model of this application. Detailed Embodiments

[0043] Next, in combination with the detailed embodiments, the present application will be further described. It should be noted that in the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification.

[0044] In the description of the present application, it should be noted that for the orientation terms, such as the terms "center", "transverse", "longitudinal", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., the indicated orientation and position relationships are based on the orientation or position relationships shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and should not be understood as limiting the specific protection scope of the present application.

[0045] It should be noted that the terms "first", "second", etc. in the description and claims of the present application are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence.

[0046] In this application, unless otherwise clearly specified or defined, the terms "installed", "connected", "joined", "fixed", etc. shall be understood in a broad sense. For example, it can be a connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0047] In this application, unless otherwise clearly specified or defined, the first feature being "on" or "under" the second feature may include the direct contact between the first and second features, or may include the situation where the first and second features are not in direct contact but in contact through other features between them. Moreover, the first feature being "above", "over" and "on top of" the second feature includes that the first feature is directly above and obliquely above the second feature, or merely indicates that the horizontal height of the first feature is higher than that of the second feature. The first feature being "under", "beneath" and "underneath" the second feature includes that the first feature is directly below and obliquely below the second feature, or merely indicates that the horizontal height of the first feature is lower than that of the second feature.

[0048] The terms "comprising" and "having" and any variations thereof in the description and claims of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0049] One preferred embodiment of this application, as Figure 1 shown, a heavy vehicle axle load identification method based on the distribution of bridge deflection influence lines, comprising the following steps:

[0050] S100: Install a monitoring device for vehicle parameter identification on the bridge deck, and at the same time arrange a structural displacement detection device at the mid-span of the bridge to identify the mid-span deflection of the bridge.

[0051] S200: Conduct on-site running tests on the bridge using small heavy vehicles of different vehicle types and weights to obtain the influence lines of small heavy vehicles of different vehicle types and weights on the mid-span deflection of the bridge.

[0052] S300: Perform simulation amplification based on the influence line data of small heavy vehicles to obtain the simulation data of the influence lines of the mid-span deflection of the bridge corresponding to large heavy vehicles.

[0053] S400: Construct a heavy vehicle axle load identification model and train the model with all the obtained influence line data corresponding to heavy vehicles.

[0054] S500: Substitute the vehicle type and the influence line data of the bridge mid-span deflection generated when the target heavy vehicle crosses the bridge into the trained heavy vehicle axle weight identification model, and then inversely obtain the axle weight data of the target heavy vehicle.

[0055] It can be understood that, first of all, in the technical solution of the present application, the monitoring device is generally installed on the side of the bridge, and the installation position of the structural displacement detection device is under the bridge; therefore, the technical solution of the present application will basically not affect the normal traffic of the bridge during the installation and construction of the equipment.

[0056] Secondly, in the traditional method, the finite element simulation method is generally used for the bridge mid-span deflection of different vehicles. Although this method is efficient, the obtained data has poor authenticity, especially the simulation data of large heavy vehicles. In the technical solution of the present application, the measured data of the bridge mid-span deflection is obtained through the vehicle running test on the bridge site. Compared with the traditional method, the data accuracy can be effectively improved. Of course, considering that the number of large heavy vehicles passing through the bridge per unit time is small, the obtained data is difficult to meet the training of the subsequent large heavy vehicle axle weight identification model; and large heavy vehicles may be special vehicles, which require professional drivers and corresponding escort personnel, etc., and do not meet the actual vehicle running test requirements. Therefore, the technical solution of the present application obtains the bridge mid-span deflection data corresponding to small heavy vehicles, and then performs simulation amplification on the obtained measured data to obtain the influence line simulation data of the bridge mid-span deflection corresponding to large heavy vehicles. Finally, the obtained measured deflection influence line data of small heavy vehicles and the deflection influence line simulation data of large heavy vehicles are used for the training of the heavy vehicle axle weight identification model. Compared with the traditional method, while ensuring the training accuracy of the model, the data acquisition speed can be effectively improved.

[0057] It should be known that generally, vehicles with a vehicle weight of more than 15 tons can be defined as heavy vehicles. The division of large heavy vehicles and small heavy vehicles can generally be limited according to the load capacity of the bridge; for example, heavy vehicles with a vehicle weight of less than 40 tons can be defined as small heavy vehicles, and vehicles with a vehicle weight of more than 40 tons can be defined as large heavy vehicles.

[0058] In this embodiment, as Figure 2 shown, the vehicle parameters identified by the monitoring device in step S100 include vehicle type, license plate, wheels and vehicle speed. According to the wheel position and number of the vehicle, the corresponding axle number and axle distance of the vehicle can be calculated. There are various specific structures of the monitoring device for identifying vehicle parameters. For the convenience of understanding, one of the structures will be described in detail below.

[0059] Specifically, as Figure 2As shown in the figure, the monitoring device includes a traffic flow camera. The specific structure and working principle of the traffic flow camera are well-known technologies to those skilled in the art, so they will not be elaborated in detail here. For example, a 25-frame-rate 8-megapixel high-definition zoom network camera can be used. The traffic flow camera can be installed on the side of the bridge by borrowing a pole or erecting a pole at a certain distance from the vehicle's on-ramp position. The traffic flow camera can take pictures of vehicles, compress the captured image data, and transmit it to the edge computing terminal through a wired network for analysis and processing to obtain data such as the vehicle type, license plate, number and position of wheels corresponding to the vehicle. Considering that the light is dim at night, which will reduce the detection accuracy of the traffic flow camera, the monitoring device of this embodiment further includes a lidar. The specific structure and working principle of the lidar are well-known technologies to those skilled in the art, so they will not be elaborated in detail here. The monitoring device performs camera image tracking and shooting of vehicles through the traffic flow camera during the day; the monitoring device performs collaborative tracking and shooting of camera images and point cloud images of vehicles through the traffic flow camera and the lidar at night.

[0060] In this embodiment, the specific process of vehicle parameter identification based on the above monitoring device is as follows:

[0061] S110: Perform image tracking and shooting of vehicles driving on the bridge deck through the monitoring device.

[0062] S120: Obtain the type of the target vehicle according to the captured image, and calculate the vehicle speed of the target vehicle based on the initial position and calibration position defined on the bridge deck.

[0063] S130: According to the calculated vehicle speed of the target vehicle, intercept the images of the target vehicle at at least one position in the original image data to identify the number of axles and wheelbase of the target vehicle.

[0064] S140: Construct a parameter set of the target vehicle regarding the license plate, vehicle type, number of axles and wheelbase.

[0065] Specifically, in the daytime, for the picture data captured by the traffic flow camera, first detect the type of the target vehicle through the vehicle type recognition algorithm, and track the trajectory of the target vehicle to the calibration position for vehicle speed estimation; that is, the distance between the monitoring initial position and the calibration position defined on the bridge deck in the picture data is known, and the vehicle speed of the target vehicle is estimated according to the time when the target vehicle moves from the initial position to the calibration position. Then, use the inferred position bounding box to intercept the target vehicle image in the original captured image, and then perform wheel detection on the intercepted target vehicle image to identify the number of axles and wheelbase of the target vehicle, and at the same time identify the license plate position and characters of the target vehicle. In the night, the traffic flow camera and the lidar collect data at the same time, fuse and superimpose the data collected by the lidar and the data collected by the traffic flow camera, and then perform the above recognition process through the fused data.

[0066] It should be noted that during the process of the traffic flow camera tracking and photographing a target vehicle, the target vehicle may be blocked and interfered by other vehicles, which may interfere with the calculation of the wheelbase and the number of axles of the target vehicle. Therefore, when identifying the wheels of the target vehicle, it is often necessary to intercept the image information at different positions during the vehicle's driving process. If directly performing positioning and identification on the entire driving process of the target vehicle, it may cause a relatively long data identification process. Therefore, in this embodiment, after estimating the speed of the target vehicle, the position of the target vehicle can be estimated according to the selected moment, and then the image can be directly intercepted at the corresponding estimated position in the original image data, which can achieve rapid positioning and identification of the target vehicle to improve the calculation process of the wheelbase and the number of axles.

[0067] In this embodiment, as Figure 2 shown, the structural displacement detection device for bridge mid-span deflection monitoring includes a target installed at the mid-span position of the bridge and an industrial camera installed on the pier capping beam on one side of the bridge. The specific structure and working principle of the industrial camera are well-known technologies to those skilled in the art, so they will not be elaborated in detail here; the industrial camera is selected with a right-handed rotation in terms of accuracy, for example, the whole-pixel accuracy is 0.1 mm / pixel. When monitoring the mid-span deflection of the bridge, the industrial camera can monitor the target for displacement calculation, and then the mid-span deflection of the bridge can be obtained and the influence line can be drawn. The displacement calculation of the bridge mid-span deflection is also to transmit the image data collected by the industrial camera to the edge computing terminal through wired network transmission for analysis and processing.

[0068] Considering that there are situations such as uneven illumination or overexposure in the industrial camera imaging, and the image clarity of the target will decrease in foggy weather. Therefore, on the basis of the above structure, the structural displacement detection device in this embodiment further includes a corner reflector installed at the mid-span of the bridge and a millimeter-wave radar installed on the pier capping beam on the other side of the bridge. The specific structure and working principle of the millimeter-wave radar are well-known technologies to those skilled in the art, so they will not be elaborated in detail here. The millimeter-wave radar can emit electromagnetic waves to the corner reflector and measure the mid-span deflection of the bridge according to the phase difference of the received electromagnetic wave signals. That is, in bad weather, the collected data of the millimeter-wave radar can be used to correct the collected data of the industrial camera to maintain the measurement accuracy of the bridge mid-span deflection.

[0069] It should be noted that the deflection change of the bridge is mainly reflected in the deflection deformation of the main girder, and generally there are multiple main girders corresponding to a single bridge. Therefore, the detection of the bridge mid-span deflection is the detection of the mid-span deflection of the main girder; then, in order to ensure the measurement accuracy, a target can be installed at the mid-span of each main girder, and each target corresponds to an industrial camera. Since the cost of the millimeter-wave radar is relatively high and it is difficult to achieve full coverage, in this embodiment, it is preferably to install a corner reflector and the corresponding millimeter-wave radar on the middle main girder.

[0070] It is understandable that the bridge deck is generally divided into multiple lanes. When vehicles drive in different lanes, the deflection responses of each main girder are different. Therefore, when performing step S200, running tests need to be carried out on all lanes. The running tests can use the natural driving data of social vehicles or send designated vehicles to conduct running tests during periods with less traffic flow. Considering the efficiency of the test and the stability of data collection, this embodiment preferably adopts the method of using designated vehicles for running tests. The selection of designated vehicles can be set according to actual needs. For example, the vehicle types of designated vehicles can be selected as two-axle vehicles, three-axle vehicles, four-axle vehicles, five-axle vehicles, and six-axle vehicles, the vehicle weight range can be selected as 10 tons, 20 tons, 30 tons, and 40 tons, the driving speed of the vehicle can be selected as 30 km / h, 40 km / h, and 50 km / h, and the actual driving lane positions when the vehicle gets on the bridge are evenly distributed.

[0071] After completing the selection of the designated vehicle, obtain the vehicle information under the above different working conditions and the measured bridge mid-span deflection data through on-site running tests; match the deflection and vehicle axle weight information, use signal processing technologies (such as variational mode decomposition and wavelet transform, etc.) to extract the bridge deflection influence line, and then calculate the deflection transverse proportionality coefficient η under the action of different lanes and different vehicle weights according to the deflection curve data of different main girders; that is, the change of the bridge mid-span deflection is characterized by the deflection transverse proportionality coefficient η.

[0072] Specifically, the deflection transverse proportionality coefficient η corresponding to the i-th main girder i The calculation formula is: ; where ω i represents the mid-span deflection of the i-th main girder, represents the sum of the mid-span deflections of the n main girders corresponding to the bridge. Then, the deflection transverse proportionality coefficient η generated by the vehicle on each main girder in any lane x can be expressed in the form of a matrix, that is, η x =[η x1 , η x2 , ……, η xn . After obtaining the corresponding deflection transverse proportionality coefficient matrix, perform signal filtering processing on it and output smooth curve data for training the heavy vehicle recognition model.

[0073] For the convenience of understanding, the following will be described in detail through specific examples. Assume that the target bridge has 5 main girders and corresponds to 3 lanes. Then, the deflection transverse proportionality coefficient matrices of each main girder corresponding to the 3 lanes are respectively η 1 =[η 11 , η 12 , η 13 , η 14 , η 15 , η 2 =[η21 , η 22 , η 23 , η 24 , η 25 , η 3 = [η 31 , η 32 , η 33 , η 34 , η 35 . Taking η 11 as an example, η 11 represents a column vector, characterizing the deflection lateral proportionality coefficient when the vehicle acts at different driving positions on the first lane.

[0074] It can be understood that during the actual use of the bridge, within a relatively short service time of the bridge, such as within 3 - 5 years, and when the traffic flow conditions are normal, the decrease in the lateral connection stiffness between the main girders in a multi - girder system is very small and can be ignored. However, after the bridge has been in service for a long time, such as 5 - 10 years, the decrease in the lateral connection stiffness between its main girders is relatively large. At this time, in order to ensure the accuracy of the prediction of the heavy - vehicle recognition model, it is necessary to recalibrate the deflection lateral proportionality coefficient of the main girders. The specific calibration method can be to recalibrate the deflection lateral proportionality coefficient of each main girder through a single vehicle type.

[0075] It should be noted that in order to ensure the accuracy of the on - site vehicle - running test, corresponding industrial cameras and targets, etc. are set for each main girder in this embodiment; after the on - site vehicle - running test is completed, for the subsequent monitoring of the bridge, only the deflection of some main girders needs to be identified to judge the decrease in the lateral connection stiffness of the main girders. Then, from the perspective of cost savings, after step S200 is completed, some of the structural displacement detection devices are removed, and based on the mid - span deflection of the corresponding main girders obtained by the remaining structural displacement detection devices, the mid - span deflection of the other main girders is calculated through the obtained deflection lateral proportionality coefficient η. For the convenience of understanding, the following will be described in detail through specific examples.

[0076] Specifically, for example, if the bridge corresponds to 5 main girders, after the on - site vehicle - running test is completed, 3 of the industrial cameras and the corresponding targets can be recycled, and the remaining 2 industrial cameras are used for subsequent long - term bridge deflection measurement, and the deflections of the other 3 main girders are calculated according to the deflection lateral proportionality coefficient of each main girder. The 5 main girders are numbered from #1 to #5 in sequence. Assuming that the vehicle is driving in the first lane and the remaining two industrial cameras are located at the positions of main girder #2 and main girder #4 respectively, then the deflection corresponding to main girder #2 is ω 2 , and the deflection corresponding to main girder #4 is ω 4 , then according to the lateral deflection proportionality coefficient, the deflections of the other main girders are:

[0077] ω 1 = (η11 ×ω 2 ) / η 12 。

[0078] ω 3 =[(η 13 ×ω 2 ) / η 12 +(η 13 ×ω 4 ) / η 14 / 2。

[0079] ω 5 =(η 15 ×ω 4 ) / η 14 。

[0080] It should be noted that in the above calculation formula, the values of η 11 , η 12 , η 13 , η 14 and η 15 are all obtained from on-site vehicle running tests; that is, after the on-site vehicle running test is completed, the deflection transverse ratio coefficients of each main girder corresponding to each lane can be regarded as fixed values.

[0081] In this embodiment, when disassembling the structural displacement detection device after completing step S200, at least two structural displacement detection devices are left. The process of judging the transverse connection stiffness between the main girders through the transverse ratio coefficients corresponding to the remaining structural displacement detection devices is as follows: Set the transverse connection stiffness judgment formula: (ω x / ω y ) = (η x / η y ); if the transverse connection stiffness judgment formula holds, the transverse connection stiffness between the corresponding main girders of the bridge meets the requirements, otherwise it will be determined that the transverse connection stiffness between the corresponding main girders of the bridge does not meet the requirements, where ω x and ω y respectively represent the mid-span deflections of the main girders corresponding to two of the remaining structural displacement detection devices, and η x and η y respectively represent the deflection transverse ratio coefficients of the main girders corresponding to two of the remaining structural displacement detection devices.

[0082] For the convenience of understanding, take the industrial cameras reserved on the main girders #2 and #4 in the first lane as an example above; when the lateral connection stiffness of the main girders does not decrease significantly, the ratio of the deflections between any two main girders is equal to the ratio of their lateral deflection proportionality coefficients. From the foregoing content, it can be seen that after the running vehicle test is completed, the lateral deflection proportionality coefficients of the main girders corresponding to each lane can be regarded as fixed values. Then, the ratio of the lateral deflection proportionality coefficients corresponding to the main girders #2 and #4 can be regarded as a fixed value k. During the subsequent use of the bridge, it is only necessary to continuously monitor the deflections of the main girders #2 and #4 and calculate ω 2 and ω 4 and determine whether the ratio is equal to k or the difference is within the set error range, so as to realize the monitoring of the decrease in the lateral connection stiffness of the main girders.

[0083] In this embodiment, step S300 includes the following processes:

[0084] S310: Construct a finite element model of the bridge based on the bridge construction plan.

[0085] S320: Use the measured deflection influence line data and the corresponding vehicle information of the small heavy vehicle running vehicle test to correct the stiffness of the bridge finite element model.

[0086] S330: Apply the load corresponding to the large heavy vehicle to the corrected bridge finite element model, and then obtain the simulation data of the influence line of the bridge mid-span deflection corresponding to the large heavy vehicle loading.

[0087] For the convenience of understanding, the following can be described in detail through specific examples. Assume that the vehicle types of the heavy vehicles are selected as 4-axle, 5-axle, 6-axle, etc., and the vehicle weights are 50 tons, 60 tons, 70 tons, 80 tons and above, etc. Step S300 mainly forms a training data set of the vehicle axle weight identification model through the bridge mid-span deflection obtained from the on-site running vehicle test and the corresponding driving vehicle information (vehicle axle number and axle weight). And use the measured data to correct and optimize the main girder stiffness and the lateral connection stiffness between the main girders of the bridge finite element model, and amplify the deflection data of the bridge running vehicle condition with a vehicle weight of more than 40 tons.

[0088] Specifically, first, the measured deflection data from the sports car test with a vehicle weight of less than 40 tons and the vehicle information are used to correct the stiffness of the finite element model. For example, the known vehicle load information is applied to the established finite element model to obtain the simulated bridge deflection data, and the error analysis is carried out with the measured deflection data corresponding to the vehicle weight. At the same time, the stiffness of the main beam and the lateral connection stiffness between the main beams are continuously adjusted until the error between the two reaches the minimum. Then, vehicle loads of 50 tons, 60 tons, 70 tons, and over 80 tons are applied to the above-mentioned bridge finite element model with corrected stiffness, and the bridge deflection influence line data under the vehicle load of over 40 tons are calculated to expand the bridge deflection data under the working conditions where the sports car test cannot be actually carried out. Finally, the bridge deflection influence line data of heavy vehicles over 40 tons obtained by simulation are used as the input of the vehicle axle weight identification model, and the corresponding axle weight is used as the output of the vehicle axle weight identification model; the measured and simulated data samples are mixed to construct the training data set of the model.

[0089] In this embodiment, step S400 includes the following specific processes:

[0090] S410: Construct the deep learning model architecture of the heavy vehicle axle weight identification model, and select the root mean square error as the loss function L driven by the model data data .

[0091] S420: Embed the bridge finite element model with corrected stiffness into the deep learning model training process.

[0092] S430: Input the given bridge deflection influence line data into the deep learning model to generate the corresponding axle weight vector and axle distance information.

[0093] S440: Convert the predicted axle weight vector and axle distance information into the load of the bridge finite element model and perform structural simulation to obtain the corresponding bridge deflection influence line simulation data.

[0094] S450: Take the error between the given bridge deflection influence line data and the bridge deflection influence line simulation data as the physical information loss function L of the deep learning model phy .

[0095] S460: Take the loss functions L data and L phy as the optimization objects of the deep learning model, and perform parameter backpropagation to iterate the model parameters.

[0096] It can be understood that the loss function L of the heavy vehicle axle weight identification model is L = L data + λL phy ; where λ is the penalty term coefficient, and the specific value needs to be adjusted according to the actual training situation, and the value range is [0, 1].

[0097] In this embodiment, step S500 includes the following specific processes:

[0098] S510: When the monitoring device detects that the target heavy vehicle gets on the bridge, trigger the under-bridge structure displacement detection device to calculate the mid-span deflection of the bridge.

[0099] S520: Use signal processing technology to eliminate the structural dynamic response and environmental noise components in the measured deflection influence line data, and obtain the bridge deflection influence line data of the measured points.

[0100] S530: Estimate the bridge deflection influence lines of other points at the mid-span of the bridge based on the calculated deflection transverse proportionality coefficient and the bridge deflection influence line data of the measured points.

[0101] S540: After the target heavy vehicle crosses the bridge, input the bridge deflection influence line data of all points into the vehicle axle load identification model to estimate each axle load.

[0102] The above describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present application. Without departing from the spirit and scope of the present application, the present application will have various changes and improvements, and these changes and improvements all fall within the scope of the present application claimed. The scope of protection required by the present application is defined by the appended claims and their equivalents.

Claims

1. A method for identifying axle weight of heavy vehicles based on the distribution of bridge deflection influence lines, characterized in that: The steps include: S100: a monitoring device for identifying vehicle parameters is set up on the bridge deck, and a structural displacement detection device is arranged at the mid-span of the bridge to identify the mid-span deflection of the bridge; S200: Use small and heavy vehicles of different models and weights to conduct on-site running test on the bridge to obtain the influence lines of small and heavy vehicles of different models and weights on the mid-span deflection of the bridge; S300: performing simulation amplification according to the influence line data of the small heavy vehicle to obtain influence line simulation data of the mid-span deflection of the bridge corresponding to the large heavy vehicle; S400: constructing a heavy vehicle axle weight recognition model and performing model training through the influence line data corresponding to all heavy vehicles obtained; S500: Substituting the model of the target heavy vehicle when crossing the bridge and the influence line data of the generated bridge mid-span deflection into the trained heavy vehicle axle weight identification model, and then inverting to obtain the axle weight data of the target heavy vehicle; Step S300 includes the following process: S310: Construct a finite element model of the bridge; S320: Using the measured deflection influence line data of the small heavy vehicle running test and the corresponding vehicle information, the finite element model of the bridge is corrected for stiffness; S330: applying a load corresponding to a large heavy vehicle to the modified bridge finite element model, thereby obtaining influence line simulation data of the mid-span deflection of the bridge corresponding to the large heavy vehicle loading; Step S400 includes the following specific processes: S410: Build a deep learning model architecture for the heavy vehicle axle weight recognition model and select the root mean square error as the model data-driven loss function L data ; S420: embedding the bridge finite element model with completed stiffness correction into the deep learning model training process; S430: Input given bridge deflection influence line data into the deep learning model to generate corresponding axle weight vector and wheelbase information; S440: converting the predicted axle weight vector and wheelbase information into loads of a bridge finite element model and performing structural simulation to obtain corresponding bridge deflection influence line simulation data; S450: The error between the given bridge deflection influence line data and the bridge deflection influence line simulation data is used as the physical information loss function L of the deep learning model phy ; S460: Loss function L data and L phy As the optimization object of the deep learning model, parameter backpropagation is performed to iterate the model parameters.

2. The heavy vehicle axle weight identification method based on the bridge deflection influence line distribution according to claim 1 is characterized in that: The specific process of performing vehicle parameter identification by the monitoring device in step S100 is as follows: S110: Tracking and photographing images of vehicles traveling on the bridge deck by using a monitoring device; S120: Acquire the type of the target vehicle according to the captured image, and calculate the speed of the target vehicle according to the initial position and the calibrated position defined on the bridge deck; S130: According to the estimated speed of the target vehicle, intercepting an image of the target vehicle at at least one position in the original image data to identify the number of axles and wheelbase of the target vehicle; S140: Construct a parameter set of the target vehicle regarding the license plate, vehicle model, number of axles, and wheelbase.

3. The heavy vehicle axle weight identification method based on the bridge deflection influence line distribution according to claim 2 is characterized in that: The monitoring device includes a traffic camera and a laser radar; the monitoring device uses the traffic camera to track and shoot the vehicle with camera images during the day; and the monitoring device uses the traffic camera and the laser radar to coordinately track and shoot the vehicle with camera images and point cloud images at night.

4. The heavy vehicle axle weight identification method based on the bridge deflection influence line distribution according to claim 2 is characterized in that: The structural displacement detection device includes a target and a corner reflector installed at the mid-span position of the bridge main beam, and an industrial camera and a millimeter-wave radar installed at the bridge pier; the industrial camera monitors the displacement of the target to obtain the mid-span deflection of the bridge, and the millimeter-wave radar monitors the phase difference of the electromagnetic wave signal of the corner reflector to obtain the mid-span deflection of the bridge.

5. The heavy vehicle axle weight identification method based on bridge deflection influence line distribution according to claim 1, characterized in that: When performing step S200, a plurality of structural displacement detection devices for monitoring the mid-span deflection of the bridge are provided and are arranged in sequence according to the main beams corresponding to the bridge; According to the mid-span deflection of the main beam of each lane obtained from the on-site sports car test, the lateral proportional coefficient η of the deflection under the action of different lanes and vehicle weights is calculated; After completing step S200, some of the structural displacement detection devices are removed, and based on the mid-span deflections of the corresponding main beams obtained by the remaining structural displacement detection devices, the mid-span deflections of the remaining main beams are calculated using the obtained deflection lateral proportional coefficient η.

6. The heavy vehicle axle weight identification method based on the bridge deflection influence line distribution according to claim 5 is characterized in that: The transverse proportional coefficient η of the deflection corresponding to the i-th main beam i The calculation formula is: ; Among them, ω i represents the mid-span deflection of the ith main beam, Represents the sum of the mid-span deflections of the n main beams corresponding to the bridge.

7. The heavy vehicle axle weight identification method based on the bridge deflection influence line distribution according to claim 5 is characterized in that: When the structural displacement detection device is disassembled after step S200, at least two structural displacement detection devices remain. The process of determining the transverse connection stiffness between the main beams by using the transverse proportional coefficients corresponding to the remaining structural displacement detection devices is as follows: Set the lateral connection stiffness judgment formula: (ω x / ω y ) = (η x / η y ); if the transverse connection stiffness judgment formula is established, the transverse connection stiffness between the main beams corresponding to the bridge meets the requirements, otherwise it will be determined that the transverse connection stiffness between the main beams corresponding to the bridge does not meet the requirements, and the transverse proportional coefficient of the deflection of each main beam is recalibrated through a single vehicle model; Among them, ω x and ω y Respectively represent the mid-span deflection of the main beam corresponding to the remaining two structural displacement detection devices, η x and η y They respectively represent the lateral proportional coefficients of the deflection of the main beams corresponding to the remaining two structural displacement detection devices.

8. The heavy vehicle axle weight identification method based on the bridge deflection influence line distribution according to claim 1 is characterized in that: Step S500 includes the following specific processes: S510: When the monitoring device detects that a target heavy vehicle is on the bridge, the displacement detection device of the underbridge structure is triggered to calculate the mid-span deflection of the bridge; S520: using signal processing technology to eliminate structural dynamic response and environmental noise components in the measured deflection influence line data, and obtaining the bridge deflection influence line data of the measured point; S530: estimating the bridge deflection influence lines of other points in the mid-span of the bridge according to the calculated deflection lateral proportionality coefficient and the bridge deflection influence line data of the measured points; S540: After the target heavy vehicle passes the bridge, the bridge deflection influence line data of all points are input into the vehicle axle weight identification model to estimate the axle weight of each axle.

Images