Bridge influence area identification method and device based on mobile vehicle driving strategy

Abstract

The application discloses a bridge influence surface identification method and device based on a new driving strategy of a mobile vehicle, a computer and a storage medium, and relates to the field of bridge structure safety detection.The application solves the problems of low calculation efficiency and low accuracy of existing identification technology without considering the efficiency and accuracy of loading on various bridges.The method comprises the following steps: setting a driving path of a vehicle, and obtaining a plurality of influence lines with a lateral distance of a vehicle width; collecting a bridge dynamic response; constructing an influence surface identification model according to the bridge dynamic response and the plurality of influence lines with the lateral distance of the vehicle width; obtaining a plurality of longitudinal influence lines of the bridge according to the influence surface identification model with singular value decomposition; and obtaining a fitted bridge influence surface by using a modified Hermite interpolation to process the plurality of longitudinal influence lines of the bridge.The application is applied to the field of bridge health monitoring.

Description

Technical Field

[0001] This invention relates to the field of bridge structural safety inspection, and in particular to a method for identifying the impact surface of a bridge based on the driving strategy of moving vehicles. Background Technology

[0002] Bridges, as key structures in transportation infrastructure, play a vital role in ensuring people's daily lives and promoting socio-economic development. Due to factors such as vehicle loads, temperature loads, wind loads, and environmental effects, the technical condition of bridges may change, even leading to catastrophic safety accidents like collapses. Various bridge accidents cause significant losses to public safety and socio-economic property. Therefore, it is necessary to obtain information on bridge influence lines and influence surfaces through dynamic load tests based on practical bridge operation and maintenance, to effectively assess the structural condition of bridges, and to prevent accidents from occurring.

[0003] Influence lines and influence surfaces contain structural mechanical property information directly related to the bridge's structural state, enabling effective assessment of bridge conditions. Current methods for identifying influence lines on actual bridges are typically based on the quasi-static response caused by slow-moving vehicles crossing the bridge, often resulting in low efficiency in practical testing. Compared to influence lines, influence surfaces more comprehensively reflect the state information of any location on the bridge in space. However, current research on influence surface identification under moving vehicle loads is relatively scarce, and existing studies have not considered the actual conditions of bridge load tests. The essence of identifying bridge influence surfaces using moving vehicle loads is to identify the influence lines at the wheel trajectories and then use interpolation methods to fit these lines to obtain the influence surface. The denser the wheel impact points, the denser the obtained influence lines, and the more accurate the fitted influence surface. Therefore, in actual measurements, it is necessary to maximize the range of the load acting on the bridge surface. Thus, selecting an appropriate moving vehicle strategy to cover as much of the bridge surface as possible while considering the effects of vehicle travel back and forth, and accurately and efficiently identifying the bridge influence surface, is essential for effectively improving the safety inspection capabilities of bridge structures.

[0004] Existing technology CN110781607A discloses a bridge influence surface identification method considering the spatial distribution of vehicle wheel loads. The steps are as follows: Step 1: Load the bridge with a loading vehicle along a selected loading path and collect response data; Step 2: Number the two-dimensional positions of the bridge and establish a mathematical model for influence surface identification; Step 3: Solve the influence surface identification equation using the L2 regularization method. However, this method uses a single loading method and does not consider the efficiency and accuracy issues of loading multiple bridges. Although it is more accurate in solving the mathematical model, the computational efficiency is reduced. Summary of the Invention

[0005] This invention solves the problems of existing identification technologies not considering the efficiency and accuracy of loading on various bridges, and having low computational efficiency.

[0006] This invention provides a method for identifying the impact surface of a bridge based on the driving strategy of moving vehicles, the method comprising:

[0007] Set the vehicle's driving path and obtain multiple influence lines with lateral spacing between the vehicle width;

[0008] Collect bridge dynamic response;

[0009] An influence surface identification model is constructed based on multiple influence lines relating the bridge's dynamic response and the lateral spacing between vehicle widths.

[0010] Based on the influence surface identification model that incorporates singular value decomposition, multiple longitudinal influence lines of the bridge are obtained;

[0011] The bridge's multiple longitudinal influence lines are processed using modified Hermite interpolation to obtain the fitted bridge influence surface.

[0012] Furthermore, a preferred embodiment is also provided, wherein setting the vehicle's driving path and obtaining multiple influence lines with lateral spacing between the vehicle width includes:

[0013] Driving Path 1: The inspection vehicle travels back and forth on the bridge surface. Each time the inspection vehicle crosses the bridge, it moves laterally by one vehicle width. The wheel trajectory on one side of the vehicle during each trip coincides with the wheel trajectory on the other side of the vehicle during the previous trip. Multiple influences with a lateral spacing of one vehicle width are calculated and obtained.

[0014] Driving Path 2: The inspection vehicle travels back and forth on the bridge surface. After each U-turn, the inspection vehicle moves laterally by half a vehicle width. The center of gravity of the vehicle in each trip coincides with the wheel trajectory on the side of the previous trip. Multiple influence lines with a lateral spacing of half a vehicle width are calculated and obtained.

[0015] Furthermore, a preferred embodiment is also provided, wherein the construction of the influence surface identification model based on multiple influence lines of the bridge dynamic response and the lateral spacing vehicle width includes:

[0016] ,

[0017] Among them, R s S represents the superimposed bridge response; R is the bridge response vector under moving vehicles; Φ is the influence line vector after converting from two dimensions to one dimension; S is the vehicle load information matrix calculated from the influence surface; S1 is the load information block matrix considering the two sides of the wheels separately; S2 represents the load information block matrix of the resultant force of the wheels.

[0018] Furthermore, a preferred embodiment is also provided, wherein obtaining multiple longitudinal influence lines of the bridge based on the influence surface identification model incorporating singular value decomposition includes:

[0019] ,

[0020] in, It is a singular value. To adjust the regularization parameters of the penalty function weights, For the elements in the regularization matrix, It is a left singular vector. To measure the bridge response, It is a right singular vector.

[0021] Furthermore, a preferred embodiment is also provided, wherein the step of processing multiple longitudinal influence lines of the bridge using modified Hermite interpolation to obtain a fitted bridge influence surface includes:

[0022] ,

[0023] ,

[0024] ,

[0025] in, The weights to the right of data point 3, The weights of data point 3 are the co-occurrence weights. The slope between data points 2 and 3. The slope between data points 3 and 4. The slope between data points 4 and 5. Let be the derivative at data point 3.

[0026] Based on the same inventive concept, the present invention also provides a bridge impact surface identification device based on a moving vehicle driving strategy, the device comprising:

[0027] The vehicle width multi-influence line acquisition unit is used to set the vehicle's driving path and acquire multiple influence lines of the lateral spacing of the vehicle width;

[0028] Bridge dynamic response acquisition unit, used to acquire bridge dynamic response;

[0029] The influence surface identification model construction unit is used to construct an influence surface identification model based on multiple influence lines of the bridge dynamic response and the lateral spacing vehicle width.

[0030] The bridge multiple longitudinal influence line acquisition unit is used to acquire multiple longitudinal influence lines of the bridge based on the influence surface mathematical model with singular value decomposition.

[0031] The bridge influence surface acquisition unit is used to process multiple longitudinal influence lines of the bridge using modified Hermite interpolation to obtain a fitted bridge influence surface.

[0032] Furthermore, a preferred embodiment is also provided, wherein the multiple influence lines acquisition unit for the vehicle width includes:

[0033] Driving Path Module 1: Used to detect the vehicle traveling back and forth on the bridge. Each time the vehicle crosses the bridge, it moves laterally by one vehicle width. The wheel trajectory on one side of the vehicle during each crossing coincides with the wheel trajectory on the other side of the vehicle during the previous crossing. Multiple influences with a lateral spacing of one vehicle width are calculated and obtained.

[0034] The second module of the driving path is used to detect the vehicle traveling back and forth on the bridge. After each turn, the vehicle moves laterally by half a vehicle width. The center of gravity of the vehicle in each trip coincides with the wheel trajectory on the side of the previous trip. Multiple influence lines with a lateral spacing of half a vehicle width are calculated and obtained.

[0035] Furthermore, a preferred embodiment is also provided, wherein the influence surface identification model construction unit includes:

[0036] ,

[0037] Among them, R s S represents the superimposed bridge response; R is the bridge response vector under moving vehicles; Φ is the influence line vector after converting from two dimensions to one dimension; S is the vehicle load information matrix calculated from the influence surface; S1 is the load information block matrix considering the two sides of the wheels separately; S2 represents the load information block matrix of the resultant force of the wheels.

[0038] Based on the same inventive concept, the present invention also provides a computer-readable storage medium for storing a computer program that executes the bridge impact surface identification method based on the moving vehicle driving strategy described in any of the preceding claims.

[0039] Based on the same inventive concept, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the bridge impact surface identification method based on the moving vehicle driving strategy described in any one of the preceding claims.

[0040] The advantages of this invention are:

[0041] This invention solves the problems of existing identification technologies not considering the efficiency and accuracy of loading on various bridges, and having low computational efficiency.

[0042] The bridge impact surface identification method based on mobile vehicle driving strategy described in this invention differs from the single loading method in the prior art. This invention proposes two different vehicle driving paths, which can effectively consider the efficiency and accuracy issues of loading various bridges.

[0043] Existing technologies do not consider the reversing process during vehicle loading, thus wasting one detection data point. This invention considers the vehicle reversing process, doubling the detection time. By taking into account the vehicle's round-trip effect, it can effectively reduce the number of trips and lower detection time.

[0044] While existing technologies are more accurate in solving mathematical models, their computational efficiency is reduced. This invention introduces singular value decomposition to improve efficiency. After performing dynamic-static separation processing on the bridge displacement response obtained from bridge dynamic load tests under new driving paths, a two-dimensional mathematical model for identifying bridge influence surfaces under vehicle loads is established. Furthermore, a solution method based on Tikhonov regularization that incorporates singular value decomposition is proposed, which improves the computational efficiency and identification accuracy of influence lines.

[0045] This invention employs a modified Akima piecewise cubic Hermite interpolation method to perform lateral fitting on the longitudinal influence line, obtaining the bridge influence surface. This provides a more comprehensive reflection of the bridge's structural condition and effectively assists in bridge structural safety inspection. Furthermore, this invention offers advantages such as wide coverage, high efficiency and accuracy, high identification precision, low cost, and the elimination of the need for complete road closures.

[0046] This invention is applied to the field of bridge health monitoring. Attached Figure Description

[0047] Figure 1 This is a flowchart of the bridge impact surface identification method based on the driving strategy of mobile vehicles as described in Implementation Method 1.

[0048] Figure 2 This is a schematic diagram of the driving path described in Implementation Method 2;

[0049] Figure 3 This is a schematic diagram of the second driving path described in Implementation Method 2;

[0050] Figure 4 This is a mid-span cross-sectional view of the 30m simply supported T-beam described in Implementation Method Eleven;

[0051] Figure 5 The finite element model diagram of the 30m simply supported T-beam described in Implementation Method Eleven;

[0052] Figure 6 This is a cross-sectional view of the 16m simply supported hollow slab described in Embodiment Eleven;

[0053] Figure 7 The finite element model diagram of the 16m simply supported hollow slab described in Implementation Method Eleven;

[0054] Figure 8 A schematic diagram of the load test vehicle described in Implementation Method Eleven;

[0055] Figure 9 Figure 1 is an influence line diagram of the influence surface of the hollow slab beam #4 and the wheel track as described in Implementation Method 11; wherein, Figure (a) is an influence line diagram of the influence surface of the hollow slab beam #4, and Figure (b) is an influence line diagram of the wheel track.

[0056] Figure 10 The eleventh implementation method describes the influence surface diagram calculated using the finite element model. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0058] Implementation Method 1, see Figure 1 This embodiment describes a bridge impact surface identification method based on moving vehicle driving strategies. The method includes:

[0059] Set the vehicle's driving path and obtain multiple influence lines with lateral spacing between the vehicle width;

[0060] Collect bridge dynamic response;

[0061] An influence surface identification model is constructed based on multiple influence lines relating the bridge's dynamic response and the lateral spacing between vehicle widths.

[0062] Based on the influence surface identification model that incorporates singular value decomposition, multiple longitudinal influence lines of the bridge are obtained;

[0063] The bridge's multiple longitudinal influence lines are processed using modified Hermite interpolation to obtain the fitted bridge influence surface.

[0064] This implementation method establishes a two-dimensional mathematical model for identifying the bridge influence surface under vehicle load by separating the dynamic and static responses obtained from bridge dynamic load tests through setting vehicle travel paths. It also proposes a Tikhonov regularization-based solution method incorporating singular value decomposition, improving the computational efficiency and identification accuracy of the influence lines. A modified Akima piecewise cubic Hermite interpolation method is used to laterally fit the longitudinal influence lines, obtaining the bridge influence surface, which can more comprehensively reflect the bridge structural state information and provide effective assistance for bridge structural safety inspection. Furthermore, this method has the advantages of wide coverage, high efficiency and accuracy, high identification precision, low cost, and does not require complete traffic closure.

[0065] Implementation Method 2, see below Figure 2This embodiment further defines the bridge influence surface identification method based on moving vehicle driving strategy described in Embodiment 1. The step of setting the vehicle's driving path and obtaining multiple influence lines with lateral spacing equal to the vehicle width includes:

[0066] Driving Path 1: The inspection vehicle travels back and forth on the bridge surface. Each time the inspection vehicle crosses the bridge, it moves laterally by one vehicle width. The wheel trajectory on one side of the vehicle during each trip coincides with the wheel trajectory on the other side of the vehicle during the previous trip. Multiple influences with a lateral spacing of one vehicle width are calculated and obtained.

[0067] Driving Path 2: The inspection vehicle travels back and forth on the bridge surface. After each U-turn, the inspection vehicle moves laterally by half a vehicle width. The center of gravity of the vehicle in each trip coincides with the wheel trajectory on the side of the previous trip. Multiple influence lines with a lateral spacing of half a vehicle width are calculated and obtained.

[0068] Unlike the single loading method of existing technologies, this embodiment proposes two different vehicle travel paths, which can effectively consider the efficiency and accuracy of loading various bridges. The first travel path proposed in this embodiment is more suitable for T-beam bridges, while the second travel path is more suitable for hollow slab bridges, so that the load can cover the bridge deck as completely as possible.

[0069] This embodiment solves the problem that the prior art does not consider the reverse process during vehicle loading, thus wasting one detection data. This embodiment takes into account the vehicle reverse process, thus doubling the detection time.

[0070] Implementation Method 3: This implementation method further defines the bridge influence surface identification method based on the driving strategy of moving vehicles described in Implementation Method 1. The step of constructing the influence surface identification model based on multiple influence lines derived from the bridge's dynamic response and the lateral spacing between vehicle widths includes:

[0071] ,

[0072] Among them, R s S represents the superimposed bridge response; R is the bridge response vector under moving vehicles; Φ is the influence line vector after converting from two dimensions to one dimension; S is the vehicle load information matrix calculated from the influence surface; S1 is the load information block matrix considering the two sides of the wheels separately; S2 represents the load information block matrix of the resultant force of the wheels.

[0073] The influence surface identification model constructed from multiple influence lines relating the bridge's dynamic response and the lateral spacing between vehicle widths is specifically as follows:

[0074] The relationship between the bridge response vector and the influence line vector can be expressed as:

[0075]

[0076] Where R represents the response vector of the moving vehicle under the bridge; L represents the influence line vector of the bridge; L represents the vehicle load information matrix.

[0077] Considering the spatial effects of vehicles, the row vectors in the load information matrix L, representing the vehicles themselves, need to be transformed into two-dimensional matrices. Simultaneously, the influence surface is also a two-dimensional matrix. Therefore, converting the load information matrix into a three-dimensional matrix is ​​both difficult and computationally expensive. To reduce computational complexity, the following dimensionality reduction method is used:

[0078] The influence lines of the bridge sections of concern along each trajectory of the car are regarded as longitudinal connections rather than transverse connections in reality. Thus, the bridge influence surface (multiple influence lines) is represented in one dimension, which can be expressed by equation (2).

[0079]

[0080] in, Indicates the impact surface after dimensionality reduction; i represents the influence line at different wheel trajectory lines.

[0081] Because the number of vehicle trips is less than the number of wheel trajectory lines, the number of known calculation conditions is far less than the number of unknown solutions. Therefore, it is necessary to increase the number of rows in the transfer matrix of the final equation to improve the robustness of the calculation program. To fully utilize the known information, it is proposed to input the condition of the resultant force of the two wheels of the car acting on a certain influence line to obtain the response R into the equation system. A resultant force load term S2 of the two wheels of the car is added to the load information matrix, and the influence surface vector... The influence line of the position of the resultant force is added to the middle.

[0082] Because the influence surface of the bridge is largely nonlinearly distributed laterally, the resultant force of the two rows of wheels does not act at the centerline of the two trajectory lines. Based on the lateral continuity of the influence surface, and according to the intermediate value theorem (Bolzano-Cauchy second theorem), the resultant force can be determined to act at an uncertain location between the two rows of wheels. Therefore, the following mathematical model can be constructed:

[0083]

[0084] Among them, R s Φ represents the superimposed bridge response; Φ represents the response that can be represented by equation (4), where Φ is...; S represents the vehicle load information matrix calculated from the influence surface; S1 represents the load information block matrix considering the two sides of the wheel separately; S2 represents the load information block matrix of the resultant force of the wheel.

[0085]

[0086] in, Indicates the interval between the first and second halves of the ... third halves of the second hai Article and ( i +1) The influence line at a certain position in the trajectory line needs to be removed from the result because its position is uncertain.

[0087] In actual driving paths, cars travel back and forth across the bridge surface, with different directions each time, a situation that needs to be considered in the solution. Representing the multiple influence lines of the bridge as one-dimensional vectors is equivalent to fixing the vehicle's driving direction; therefore, the consistency of the vehicle's driving direction needs to be ensured in the influence surface identification model. For travel along the bridge direction, the vehicle load information matrix L... f =L:

[0088]

[0089] in, m This indicates the number of discrete data points in the bridge response time history that were extracted. n This represents the number of discrete data points in the influence line vector of the calculation result; Indicates the axle load of the first axle of the car

[0090] When a vehicle is traveling in the opposite direction on its return trip, the calculation treats this process as the vehicle reversing along the bridge direction. In this case, the vehicle load information matrix L... b It can be represented as:

[0091]

[0092] Taking into account the vehicle's position as it crosses each trajectory line and the corresponding bridge response, the two wheels of the vehicle are considered separately, and the load information block matrix for influence surface calculation is constructed as follows:

[0093]

[0094] in, l Indicates the number of times the test vehicle has been driven. This represents a matrix representing vehicle load information when a car is traveling along the bridge. This represents the vehicle load information matrix when a car is traveling in the reverse direction of the bridge.

[0095] Add the resultant force load term S2 of the two wheels on both sides of the car to the load information matrix, as follows:

[0096]

[0097] Implementation Method Four: This implementation method further defines the bridge influence surface identification method based on moving vehicle driving strategy described in Implementation Method One. The step of obtaining multiple longitudinal influence lines of the bridge based on the influence surface identification model incorporating singular value decomposition includes:

[0098] ,

[0099] in, It is a singular value. To adjust the regularization parameters of the penalty function weights, For the elements in the regularization matrix, It is a left singular vector. To measure the bridge response, It is a right singular vector.

[0100] Specifically, due to the load information matrix L For non-square matrices, Tikhonov regularization is introduced to resolve the ill-posed equation:

[0101]

[0102] in, This represents the sum of squares of the error terms; This represents the penalty function in the regularization algorithm; This represents the regularization parameter that adjusts the weights of the penalty function; T represents the regularization matrix.

[0103]

[0104] To improve computational efficiency, singular value decomposition (SVD) is introduced to transform the matrix... L Represented as:

[0105]

[0106] in, U for m × m The matrix, Σ is m × n The matrix is ​​equal to diag{ e i},in e i It is a singular value; V for n × n A matrix, in which vectors v i It is a right singular vector.

[0107] Equation (1) can then be written in SVD form:

[0108]

[0109] Considering the noise and environmental load effects, the measured bridge response... R m It can be represented as:

[0110]

[0111] in, r This refers to the components remaining in the measured bridge response after removing the quasi-static response caused by moving vehicles.

[0112] This allows us to obtain multiple longitudinal influence lines for the bridge:

[0113]

[0114] The regularization matrix T used in this embodiment will make the result second-order continuous during the solution process. The expression of the calculation result of equation (3) is that the influence lines at different horizontal positions are connected vertically, as shown in equation (4). However, in reality, the endpoints of different influence lines do not have continuity. Therefore, the regularization matrix T needs to be transformed as follows:

[0115]

[0116] Implementation Method 5: This implementation method further defines the bridge influence surface identification method based on moving vehicle driving strategy described in Implementation Method 1. The step of using modified Hermite interpolation to process multiple longitudinal influence lines of the bridge to obtain a fitted bridge influence surface includes:

[0117] ,

[0118] ,

[0119] ,

[0120] in, The weights to the right of data point 3, The weights of data point 3 are the co-occurrence weights. The slope between data points 2 and 3. The slope between data points 3 and 4. The slope between data points 4 and 5. Let be the derivative at data point 3.

[0121] Specifically, this implementation method uses the modified Akima piecewise cubic Hermite interpolation method, and the specific process is as follows:

[0122] Given five data points 1, 2, 3, 4, and 5 from left to right in a planar coordinate system, the derivative at the 3rd data point can be expressed as:

[0123]

[0124] In the formula, d 1-2 , d 2-3， d 3-4，d 4-5 These are the slopes of the line segments between two adjacent points.

[0125] Therefore, two points can be determined using four conditional expressions. x 1, y 1) and ( x 2, y 2) Polynomials between:

[0126]

[0127] Based on the above conditions, a cubic polynomial can be determined:

[0128]

[0129] In the formula, ; ; ;

[0130] ;

[0131] This implementation uses a modified Akima algorithm, which assigns more weight to the side with the lower slope by changing the weights, thus expressing the derivative of the calculation point (Equation 16) as:

[0132]

[0133] The bridge type in this embodiment is a beam bridge.

[0134] Implementation Method Six: The bridge impact surface identification device based on mobile vehicle driving strategy described in this implementation method includes:

[0135] The vehicle width multi-influence line acquisition unit is used to set the vehicle's driving trajectory and acquire multiple influence lines of the lateral spacing of the vehicle width;

[0136] Bridge dynamic response acquisition unit, used to acquire bridge dynamic response;

[0137] The influence surface identification model construction unit is used to construct an influence surface identification model based on multiple influence lines of the bridge dynamic response and the lateral spacing vehicle width.

[0138] The bridge multiple longitudinal influence line acquisition unit is used to acquire multiple longitudinal influence lines of the bridge based on the influence surface mathematical model with singular value decomposition.

[0139] The bridge influence surface acquisition unit is used to process multiple longitudinal influence lines of the bridge using modified Hermite interpolation to obtain a fitted bridge influence surface.

[0140] Implementation Method Seven: This implementation method further defines the bridge influence surface identification device based on the driving strategy of moving vehicles described in Implementation Method One. The multiple influence lines acquisition unit for the vehicle width includes:

[0141] Driving trajectory module 1: used to detect the vehicle traveling back and forth on the bridge surface. The detection vehicle moves laterally by one vehicle width each time it crosses the bridge. The wheel trajectory on one side of the vehicle in each trip coincides with the wheel trajectory on the other side of the vehicle in the previous trip. Multiple influences with a lateral spacing of one vehicle width are calculated and obtained.

[0142] The second module of the driving trajectory is used to detect the vehicle traveling back and forth on the bridge surface. After each turn, the vehicle moves laterally by half a vehicle width. The center of gravity of the vehicle in each trip coincides with the wheel trajectory of the previous trip. Multiple influence lines with a lateral spacing of half a vehicle width are calculated and obtained.

[0143] Implementation Method Eight: This implementation method further defines the bridge impact surface identification device based on moving vehicle driving strategy described in Implementation Method One. The impact surface identification model construction unit includes:

[0144] ,

[0145] Among them, R s S represents the superimposed bridge response; R is the bridge response vector under moving vehicles; Φ is the influence line vector after converting from two dimensions to one dimension; S is the vehicle load information matrix calculated from the influence surface; S1 is the load information block matrix considering the two sides of the wheels separately; S2 represents the load information block matrix of the resultant force of the wheels.

[0146] Implementation Method Nine: A computer-readable storage medium according to this implementation method, the computer-readable storage medium being used to store a computer program, the computer program executing the bridge impact surface identification method based on the driving strategy of moving vehicles as described in any one of Implementation Methods One to Five.

[0147] Implementation Method 10: A computer device according to this implementation method includes a memory and a processor. The memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the bridge impact surface identification method based on the driving strategy of moving vehicles as described in any one of Implementation Methods 1 to 5.

[0148] Implementation Method 11, see below Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 , Figure 9 and Figure 10This embodiment describes a specific implementation of the bridge impact surface identification method based on moving vehicle driving strategy described in Embodiment 1. It also explains Embodiments 2 through 5. Specifically:

[0149] A 30m simply supported T-beam bridge and a 16m simply supported hollow slab bridge were analyzed. Finite element models were established using the beam-grid method in Midas / civil. Lateral connections were simulated as virtual beams with only stiffness considered and weight ignored. The hinge joints were designed to transmit only shear force, achieved by releasing the beam end moment constraints at the hinge joints. The cross-sections and finite element models of the two bridges are shown below. Figure 4 , Figure 5 , Figure 6 and Figure 7 As shown.

[0150] Car parameters such as Figure 8 As shown, P1=7t, P2=P3=9t, D1=3.2m, D2=1.8m, D3=1.4m. Two driving paths are used, each traveling at a constant speed of 30km / h. The bridge surface unevenness level is "good". Driving path one: After each U-turn, the vehicle moves laterally by 1.8m, for a total of 5 trips. Driving path two: After each U-turn, the vehicle moves laterally by 0.9m, for a total of 9 trips. Both strategies cover a bridge surface width of 9m.

[0151] In driving path one, the inspection vehicle travels back and forth across the bridge. Each time the inspection vehicle crosses the bridge, it moves laterally by one vehicle width. The wheel trajectory on one side of the vehicle during each crossing coincides with the wheel trajectory on the other side of the vehicle during the previous crossing. Multiple influence lines with a lateral spacing of one vehicle width are calculated. In driving path two, the inspection vehicle travels back and forth across the bridge. Each time the inspection vehicle turns around, it moves laterally by half a vehicle width. The center of gravity of the vehicle during each crossing coincides with the wheel trajectory on the other side of the vehicle during the previous crossing. Multiple influence lines with a lateral spacing of half a vehicle width are calculated.

[0152] Output the displacement response at mid-span of hollow slab bridge #4 and T-beam bridge #3. Extract the time interval from the front axle entering the bridge to the rear axle exiting the bridge during each crossing. Calculate and extract the influence lines at the wheel trajectory lines according to the method mentioned in Implementation Method 3. Fit the results according to the method mentioned in Implementation Method 5 to obtain the complete influence surface. Taking the hollow slab bridge as an example, the results are as follows: Figure 9 As shown. The above results are compared with the influence surface calculated in the finite element model. The influence surface diagram calculated in the finite element model is shown below. Figure 10 As shown. To evaluate the error in influence surface identification, three error metrics were constructed as follows:

[0153] Overall relative error :

[0154]

[0155] in, The Frobenius norm of the matrix; The baseline influence surface matrix; To identify the influence surface matrix.

[0156] Lateral relative error :

[0157]

[0158] in, This is the transverse influence line at the mid-span position extracted from the influence surface.

[0159] Maximum absolute error D M :

[0160]

[0161] Error analysis was performed on the two bridges mentioned above:

[0162] Table 1 Error Analysis of Influence Surface Identification

[0163]

[0164] As can be seen from the table above, the overall error, lateral error, and maximum error of the two bridges are all relatively small, indicating that the bridge influence surface results obtained based on the new driving strategy for moving vehicles are very accurate and reliable.

[0165] The technical solutions provided by the present invention have been described in further detail above with reference to the accompanying drawings in order to highlight their advantages and benefits, and are not intended to limit the present invention. Any modifications, combinations, improvements and equivalent substitutions of the present invention based on the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A bridge impact surface identification method based on moving vehicle driving strategy, characterized in that, The method includes: Set the vehicle's driving trajectory and obtain multiple influence lines with lateral spacing between the vehicle width; Collect bridge dynamic response; An influence surface identification model is constructed based on multiple influence lines relating the bridge's dynamic response and the lateral spacing between vehicle widths. Based on the influence surface identification model that incorporates singular value decomposition, multiple longitudinal influence lines of the bridge are obtained; The modified Hermite interpolation is used to process multiple longitudinal influence lines of the bridge to obtain the fitted bridge influence surface; The process of setting the vehicle's driving trajectory and obtaining multiple influence lines with lateral spacing between the vehicle width includes: Driving trajectory 1: The inspection vehicle travels back and forth on the bridge surface. Each time the inspection vehicle crosses the bridge, it moves laterally by one vehicle width. The wheel trajectory on one side of the vehicle in each trip coincides with the wheel trajectory on the other side of the vehicle in the previous trip. Multiple influence lines with a lateral spacing of one vehicle width are calculated and obtained. Driving trajectory 2: The inspection vehicle travels back and forth on the bridge surface. After each U-turn, the inspection vehicle moves laterally by half a vehicle width. The center of gravity of the vehicle in each trip coincides with the wheel trajectory on the side of the previous trip. Multiple influence lines with a lateral spacing of half a vehicle width are calculated and obtained. The construction of the influence surface identification model based on multiple influence lines of the bridge dynamic response and the lateral spacing vehicle width includes: ， Among them, R s The superimposed bridge response; R is the bridge response vector under moving vehicles; Φ is the influence line vector after converting from two dimensions to one dimension; S is the vehicle load information matrix calculated from the influence surface; S1 is the load information block matrix considering the two sides of the wheel separately; S2 represents the load information block matrix of the resultant force of the wheel. The process of using modified Hermite interpolation to process multiple longitudinal influence lines of the bridge and obtain a fitted bridge influence surface includes: ， ， ， in, The weights to the right of data point 3, The weights of data point 3 are the co-lateral weights. The slope between data points 2 and 3. The slope between data points 3 and 4. The slope between data points 4 and 5. Let be the derivative at data point 3.

2. The bridge impact surface identification method based on mobile vehicle driving strategy according to claim 1, characterized in that, The method for obtaining multiple longitudinal influence lines of the bridge based on the influence surface identification model incorporating singular value decomposition includes: ， in, It is a singular value. To adjust the regularization parameters of the penalty function weights, The elements in the regularization matrix, It is a left singular vector. To measure the bridge response, It is a right singular vector.

3. A bridge impact surface identification device based on mobile vehicle driving strategy, characterized in that, The device includes: The vehicle width multi-influence line acquisition unit is used to set the vehicle's driving trajectory and acquire multiple influence lines of the lateral spacing of the vehicle width; Bridge dynamic response acquisition unit, used to acquire bridge dynamic response; The influence surface identification model construction unit is used to construct an influence surface identification model based on multiple influence lines of the bridge dynamic response and the lateral spacing vehicle width. The bridge multiple longitudinal influence line acquisition unit is used to acquire multiple longitudinal influence lines of the bridge based on the influence surface mathematical model with singular value decomposition. The bridge influence surface acquisition unit is used to process multiple longitudinal influence lines of the bridge using modified Hermite interpolation to obtain a fitted bridge influence surface. The multiple influence line acquisition unit for vehicle width includes: The driving trajectory module is used to detect the vehicle's back-and-forth travel on the bridge. Each time the vehicle crosses the bridge, it moves laterally by one vehicle width. The wheel trajectory on one side of the vehicle in each trip coincides with the wheel trajectory on the other side of the vehicle in the previous trip. Multiple influences with a lateral spacing of one vehicle width are calculated and obtained. The second module of driving trajectory is used to detect the vehicle traveling back and forth on the bridge. After each turn, the vehicle moves laterally by half a vehicle width. The center of gravity of the vehicle in each movement coincides with the wheel trajectory of the previous movement. Multiple influence lines with a lateral spacing of half a vehicle width are calculated and obtained. The influence surface identification model construction unit includes: ， Among them, R s The superimposed bridge response; R is the bridge response vector under moving vehicles; Φ is the influence line vector after converting from two dimensions to one dimension; S is the vehicle load information matrix calculated from the influence surface; S1 is the load information block matrix considering the two sides of the wheel separately; S2 represents the load information block matrix of the resultant force of the wheel. The process of using modified Hermite interpolation to process multiple longitudinal influence lines of the bridge and obtain a fitted bridge influence surface includes: ， ， ， in, The weights to the right of data point 3, The weights of data point 3 are the co-lateral weights. The slope between data points 2 and 3. The slope between data points 3 and 4. The slope between data points 4 and 5. Let be the derivative at data point 3.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that executes the bridge impact surface identification method based on the driving strategy of moving vehicles as described in any one of claims 1-2.

5. A computer device, characterized in that: The system includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the bridge impact surface identification method based on a moving vehicle driving strategy as described in any one of claims 1-2.

Images