Bridge roughness identification method based on limited vehicle response

Abstract

The application discloses a bridge pavement roughness identification method based on limited vehicle response, which comprises the following steps: step one, arranging a vehicle-mounted sensor on the central position or directly above the axle of a measuring vehicle; step two, using a tractor to guide the measuring vehicle to uniformly drive through a bridge to be measured, and collecting the vertical response of the measuring vehicle; step three, constructing a balance equation of the vehicle and the bridge, obtaining a state space equation of the vehicle-bridge coupling system according to a defined state vector, and discretely representing the state space equation of the vehicle-bridge coupling system; step four, taking the acceleration response as an observation value, and discretely representing the vehicle acceleration; step five, taking a state vector at k time as a state vector at k+2 time, and obtaining the vehicle displacement at k+2 time; step six, considering the observation noise existing in the actual situation, and obtaining an improved system observation equation; and step seven, obtaining the pavement roughness according to an extended Kalman filtering algorithm.

Description

Technical Field

[0001] This invention belongs to the field of bridge health monitoring technology, specifically a method for identifying bridge pavement unevenness based on finite vehicle response. Background Technology

[0002] Currently, bridge structural health monitoring technologies are mainly divided into direct measurement and indirect measurement technologies. Direct measurement technologies typically involve installing and deploying a series of sensors on the surface or key parts of the bridge structure to monitor and record the bridge's vibration response during operation. Indirect measurement mainly includes imaging technology, drone technology, and GPS, among others. Among these, vehicle scanning, as a novel indirect bridge condition assessment technology, has recently received widespread attention in the field of health monitoring for small and medium-sized bridges due to its advantages such as high efficiency, high mobility, and better economic efficiency.

[0003] The core idea of ​​vehicle scanning is to indirectly sense bridge vibrations through moving vehicles equipped with sensors, thereby assessing the bridge's dynamic characteristics and providing fundamental data for bridge health monitoring. However, road surface roughness is a significant factor that cannot be ignored in the application of vehicle scanning, as it significantly reduces the visibility of bridge frequencies in the vehicle response spectrum. Therefore, identifying road surface roughness from vehicle responses is a crucial step in the widespread application of vehicle scanning. In recent years, bridge roughness identification technology based on vehicle responses has emerged. This technology mainly involves relatively complex vehicle models and requires a certain number of sensors in practical applications—multiple sensors need to be deployed at different locations on the vehicle—combined with advanced signal processing techniques to identify road surface roughness. However, because the bridge responses recorded by different sensors are susceptible to external interference, it is difficult to maintain a consistent and stable state, making it difficult to apply the research results in practice. Furthermore, the large number of sensors required increases the cost of bridge health monitoring, greatly diminishing the advantages of vehicle scanning technology in terms of economy and ease of operation. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method for identifying bridge road surface unevenness based on finite vehicle response, which can identify bridge road surface unevenness using only a single vehicle-mounted sensor.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for identifying bridge pavement roughness based on finite vehicle response includes the following steps:

[0007] Step 1: Place the vehicle-mounted sensor at the center of the measuring vehicle's axle or on the vehicle body directly above the center of the measuring vehicle's axle;

[0008] Step 2: Use a tractor to guide the measuring vehicle to drive at a constant speed across the bridge to be measured. During the process of the measuring vehicle traveling on the bridge, the signal acquisition system records the vertical response of the measuring vehicle.

[0009] Step 3: Construct the equilibrium equations for the vehicle and the bridge. Based on the defined state vector Z, obtain the state-space equations of the vehicle-bridge coupled system, and then discretize the state-space equations of the vehicle-bridge coupled system.

[0010] Step 4: Using the acceleration response as the observation value, calculate the vehicle acceleration. Discretize the representation;

[0011] Step 5: Use the state vector Z at time k k Z represents the state vector at time k+2. k+2 The vehicle displacement y at time k+2 is obtained. v,k+2 ;

[0012] Step Six: Consider the actual observation noise v k+1 This yields an improved system observation equation;

[0013] Step 7: Solve for the road surface roughness using the extended Kalman filter algorithm.

[0014] Furthermore, the vehicle-mounted sensor is a displacement sensor, and in step two, the signal acquisition system records the vertical displacement response of the measuring vehicle.

[0015] Furthermore, in step three, the equilibrium equations for the vehicle and the bridge are:

[0016]

[0017]

[0018] Where: M b C b and K b These represent the bridge's mass matrix, damping matrix, and stiffness matrix, respectively; m v For the mass of the vehicle; k v For vehicle stiffness; y v This represents the vertical displacement of the vehicle. q represents the vertical acceleration of the vehicle. b Let n be the displacement vector of the bridge node, with a dimension of 1×n, where n is the number of degrees of freedom of the bridge. The velocity vector of the bridge node; r is the acceleration vector of the bridge node. c Let H be the road surface roughness to be identified; H be a row vector containing the Hamiltonian interpolation function; F be the interaction force between the vehicle and the bridge; and:

[0019] H 1×n =[0,…0,N,0,…0]

[0020] F = H T [m v g+k v (y v -r c )-k v Hq b ]

[0021] Where: N is the Hamiltonian interpolation function; g is the gravitational acceleration.

[0022] Furthermore, in step three, the state vector Z is defined as:

[0023]

[0024] in: Indicates the vehicle's vertical velocity;

[0025] The state-space equation of the vehicle-axle coupled system is:

[0026]

[0027] in: This represents the first derivative of the state vector Z with respect to time t; t represents the discrete time; A and B are the state matrix and input matrix of the state equation, respectively; F is the known load matrix; and:

[0028]

[0029]

[0030] The discretized state-space equations of the vehicle-axle coupled system are as follows:

[0031] Z k+1 =A k Z k +B k r c,k +F k +w k

[0032] A k =exp(Adt)≈(I+AΔt)

[0033] B k =[exp(Adt)-I]A -1 B≈dtB

[0034] F k =[exp(Adt)-I]A -1F

[0035] Where: I is the identity matrix; w represents system noise.

[0036] Furthermore, in step four, the vehicle acceleration... for:

[0037]

[0038]

[0039]

[0040] Acceleration of the vehicle Discretization is represented as:

[0041]

[0042] Where: L represents the output matrix; E represents the transfer matrix.

[0043] Furthermore, in step five, the state vector Z at time k+2... k+2 for:

[0044] Z k+2 =A k+1 A k Z k +A k+1 B k r c,k +A k+1 F k +B k+1 r c,k+1 +F k+1 +w k

[0045] A k+1 A k =[U 1,k+1 U 2,k+1 U 3,k+1 U 4,k+1 ] T

[0046] A k+1 B k =[V 1,k+1 V 2,k+1 V 3,k+1 V 4,k+1 ] T

[0047] The vehicle displacement y at time k+2 v,k+2 for:

[0048] y v,k+2 =U 3,k+1 Zk +V 3,k+1 r c,k

[0049] Among them: U 1,k+1 A represents k+1 A k The submatrix corresponding to the first row of the matrix; U 2,k+1 A represents k+1 A k The submatrices corresponding to rows 2 to n+1 of the matrix; U 3,k+1 A represents k+1 A k The submatrix corresponding to the (n+2)th row of the matrix; U 4,k+1 A represents k+1 A k The submatrix corresponding to rows n+3 to 2n+2 of the matrix; V 1,k+1 A represents k+1 B k The submatrix corresponding to the first row of the matrix; V 2,k+1 A represents k+1 B k The submatrices corresponding to rows 2 to n+1 of the matrix; V 3,k+1 A represents k+1 B k The submatrix corresponding to the (n+2)th row of the matrix; V 4,k+1 A represents k+1 B k The submatrix corresponding to rows n+3 to 2n+2 of the matrix.

[0050] Furthermore, in step six, the improved system observation equation is:

[0051]

[0052] Y k+1 =C k Z k+1 +D k r c,k+1 +v k+1 Where: Y k+1 Represents the improved observation vector; v k+1 Indicates observation noise; C k For the improved output matrix; D k For the improved transfer matrix.

[0053] Furthermore, in step seven, the road surface unevenness is calculated. The steps are as follows:

[0054] 71) Estimating initial state settings:

[0055]

[0056]

[0057]

[0058] Where Z0 and r0 are the initial values ​​of the state vector and the unknown input, respectively; and These are the initial estimates of the state vector and the unknown input, respectively; P z,0|0 The initial value of the state vector covariance; E[·] represents the mathematical expectation;

[0059] 72) Time update phase:

[0060]

[0061]

[0062] in: and P represents the state vectors calculated in the previous iteration step and the current iteration step, respectively; z,k+1|k Represents the state prediction value The error covariance matrix; P z,k|k Represents the state prediction value The error covariance matrix; Q k+1 The noise vector w at time k+1 k+1 The covariance matrix;

[0063] 73) Calculate the Kalman gain K z,k+1 :

[0064]

[0065] Where: C k+1|k R represents the output matrix obtained from the previous iteration; k+1 Indicates observation noise v k+1 The covariance matrix;

[0066] 74) Estimating unknown inputs:

[0067]

[0068]

[0069] Wherein: S k+1 Represents an unknown stimulus r k+1 The error covariance matrix; D k+1|k Y represents the transfer matrix obtained from the previous iteration; k+1 represents the improved observation vector; I represents the identity matrix;

[0070] 75) Measurement Update Phase:

[0071]

[0072] in: and These are the estimated values ​​of the state vector Z and the unknown input r, respectively.

[0073] The beneficial effects of this invention are as follows:

[0074] This invention presents a bridge pavement unevenness identification method based on finite vehicle response. First, the pavement unevenness is treated as an unknown excitation in a vehicle-bridge coupled system, and a displacement sensor is used to collect the vertical response of the vehicle crossing the bridge. Then, the state vector and observation vector of the vehicle-bridge coupled system are defined, and corresponding state equations and discrete equations are established to obtain the system matrix of the vehicle-bridge coupled state space equation. Finally, initial state values ​​are defined, and the pavement unevenness is reconstructed using Kalman filtering and the obtained system matrix. In summary, this invention's bridge pavement unevenness identification method based on finite vehicle response requires only a single vehicle-mounted sensor to identify bridge pavement unevenness. Since the vehicle-mounted sensor is a single sensor positioned at the center of the measuring vehicle's axle or on the vehicle body directly above the center of the axle, it not only offers advantages in terms of economy and ease of operation but also avoids external interference to the measured vertical response, thus maintaining measurement consistency and stability. Attached Figure Description

[0075] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following figures are provided for illustration:

[0076] Figure 1 This is a flowchart of an embodiment of the bridge pavement roughness identification method based on finite vehicle response of the present invention;

[0077] Figure 2 A schematic diagram of an embodiment of a road surface roughness identification system;

[0078] Figure 3 This is a schematic diagram of the measuring vehicle system.

[0079] Figure 4 The mathematical model of the bridge under test;

[0080] Figure 5 Bridge surface unevenness identified at different vehicle speeds;

[0081] Figure 6 Bridge surface unevenness identified under different environmental noise conditions;

[0082] Figure 7 To determine the road surface unevenness by placing the test vehicle on the road surface. Detailed Implementation

[0083] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0084] like Figure 1 As shown in the figure, the bridge pavement unevenness identification method based on finite vehicle response in this embodiment includes the following steps.

[0085] Step 1: Arrange the vehicle-mounted sensor 15 at the center of the axle 14 of the measuring vehicle 10 or on the carriage 16 located directly above the center of the axle 14 of the measuring vehicle 10. In this embodiment, the vehicle-mounted sensor 15 is a displacement sensor.

[0086] Step 2: The tractor 11 guides the measuring vehicle 10 to travel at a constant speed across the bridge to be measured. During the movement of the measuring vehicle 10 across the bridge, the signal acquisition system records the vertical response of the measuring vehicle. In this embodiment, the signal acquisition system records the vertical displacement response of the measuring vehicle.

[0087] Step 3: Construct the equilibrium equations for the vehicle and the bridge. Based on the defined state vector Z, obtain the state-space equations of the vehicle-bridge coupled system, and then discretize the state-space equations of the vehicle-bridge coupled system.

[0088] The equilibrium equations for the vehicle and the bridge are as follows:

[0089]

[0090]

[0091] Where: M b C b and K b These represent the bridge's mass matrix, damping matrix, and stiffness matrix, respectively; m v For the mass of the vehicle; k v For vehicle stiffness; y v This represents the vertical displacement of the vehicle. q represents the vertical acceleration of the vehicle. b Let n be the displacement vector of the bridge node, with a dimension of 1×n, where n is the number of degrees of freedom of the bridge. The velocity vector of the bridge node; r is the acceleration vector of the bridge node. c Let H be the road surface roughness to be identified; H be a row vector containing the Hamiltonian interpolation function, with all terms being zero except for the beam elements corresponding to the vehicle interaction; F be the interaction force between the vehicle and the bridge; and:

[0092] H1×n =[0,…0,N,0,…0]

[0093] F = H T [m v g+k v (y v -r c )-k v Hq b ]

[0094] Where: N is the Hamiltonian interpolation function; g is the gravitational acceleration;

[0095] The state vector Z is defined as:

[0096]

[0097] in: Indicates the vehicle's vertical velocity;

[0098] The state-space equations of the vehicle-bridge coupled system are obtained as follows:

[0099]

[0100] in: The first derivative of the state vector Z with respect to time Z is denoted by t; t represents the discrete time; A and B are the state matrix and input matrix of the state equation, respectively; F is the known load matrix; and:

[0101]

[0102]

[0103] Considering that the observed sequence is discrete and that system noise w exists, the state-space equations are discretized. In this embodiment, the discretized state-space equations of the vehicle-bridge coupling system are as follows:

[0104] Z k+1 =A k Z k +B k r c,k +F k +w k

[0105] A k =exp(Adt)≈(I+AΔt)

[0106] B k =[exp(Adt)-I]A -1 B≈dtB

[0107] F k =[exp(Ad)-I]A-1 F

[0108] Where: I is the identity matrix; w represents system noise.

[0109] Step 4: Using the acceleration response as the observation value, calculate the vehicle acceleration. Discretize the representation.

[0110] Vehicle acceleration can be obtained by differentiating the measured vehicle displacement signal, and for the vehicle scanning method, only the vehicle response can be measured. Specifically, vehicle acceleration... for:

[0111]

[0112]

[0113]

[0114] Acceleration of the vehicle Discretization is represented as:

[0115]

[0116] Where: L represents the output matrix; E represents the transfer matrix.

[0117] Step 5: Use the state vector Z at time k k Z represents the state vector at time k+2. k+2 The vehicle displacement y at time k+2 is obtained. v,k+2 .

[0118] The state vector Z at time k+2 k+2 for:

[0119] Z k+2 =A k+1 A k Z k +A k+1 B k r c,k +A k+1 F k +B k+1 r c,k+1 +F k+1 +w k

[0120] A k+1 A k =[U 1,k+1 U 2,k+1 U 3,k+1 U 4,k+1 ] T

[0121] Ak+1 B k =[V 1,k+1 V 2,k+1 V 3,k+1 V 4,k+1 ] T

[0122] The vehicle displacement y at time k+2 v,k+2 for:

[0123] y v,k+2 =U 3,k+1 Z k +V 3,k+1rc,k

[0124] Among them: U 1,k+1 A represents k+1 A k The submatrix corresponding to the first row of the matrix; U 2,k+1 A represents k+1 A k The submatrices corresponding to rows 2 to n+1 of the matrix; U 3,k+1 A represents k+1 A k The submatrix corresponding to the (n+2)th row of the matrix; U 4,k+1 A represents k+1 A k The submatrix corresponding to rows n+3 to 2n+2 of the matrix; V 1,k+1 A represents k+1 B k The submatrix corresponding to the first row of the matrix; V 2,k+1 A represents k+1 B k The submatrices corresponding to rows 2 to n+1 of the matrix; V 3,k+1 A represents k+1 B k The submatrix corresponding to the (n+2)th row of the matrix; V 4,k+1 A represents k+1 B k The submatrix corresponding to rows n+3 to 2n+2 of the matrix.

[0125] Step Six: Consider the actual observation noise v k+1 An improved system observation equation was obtained.

[0126] Let k = k + 1, the improved system observation equation is:

[0127]

[0128] Y k+1 =C k Z k+1 +D k r c,k+1 +v k+1

[0129] Where: Y k+1 Represents the improved observation vector; v k+1 Indicates observation noise; C k For the improved output matrix; D k For the improved transfer matrix.

[0130] Vehicle displacement response y v,k+3 The vehicle acceleration response can be directly recorded by displacement sensors installed on the vehicle. The displacement response can be obtained by differentiating the obtained displacement response twice with respect to time dt. Similarly, if the vehicle is equipped with an acceleration sensor, the vehicle displacement response can be obtained by integrating the vehicle acceleration response twice with respect to time dt.

[0131] Step 7: Solve for the road surface roughness using the extended Kalman filter algorithm.

[0132] The road surface roughness is obtained by solving. The steps are as follows:

[0133] 71) Estimating initial state settings:

[0134]

[0135]

[0136]

[0137] Where Z0 and r0 are the initial values ​​of the state vector and the unknown input, respectively; and These are the initial estimates of the state vector and the unknown input, respectively; P z,0|0 The initial value of the state vector covariance; E[·] represents the mathematical expectation;

[0138] 72) Time update phase:

[0139]

[0140]

[0141] Where: the subscript k|k indicates that the value was obtained from the synchronous iteration, and k+1|k indicates that the value was obtained from the previous iteration. Specifically, and P represents the state vectors calculated in the previous iteration step and the current iteration step, respectively; z,k+1|k Represents the state prediction value The error covariance matrix; P z,k|k Represents the state prediction value The error covariance matrix; Q k+1 w represents the system noise vector at time k+1. k+1 The covariance matrix;

[0142] 73) Calculate the Kalman gain K. z,k+1 :

[0143]

[0144] Where: C k+1|k R represents the output matrix obtained from the previous iteration; k+1 Indicates observation noise v k+1 The covariance matrix;

[0145] 74) Estimating unknown inputs:

[0146]

[0147]

[0148] Wherein: S k+1 Represents an unknown stimulus r k+1 The error covariance matrix; D k+1|k Y represents the transfer matrix obtained from the previous iteration; k+1 represents the improved observation vector; I represents the identity matrix;

[0149] 75) Measurement Update Phase:

[0150]

[0151] in: and These are the estimated values ​​of the state vector Z and the unknown input r, respectively.

[0152] like Figure 2 The diagram shown is a schematic of the road surface irregularity identification system applicable to the bridge road surface irregularity identification method based on finite vehicle response of the present invention. The road surface irregularity identification system of this embodiment includes a field measurement system, a data analysis and processing platform, and a data output and display terminal.

[0153] In this embodiment, the field measurement system includes a measurement vehicle system, a data acquisition module, a data conversion module, a data communication module, and a data storage module. The measurement vehicle system in this embodiment is used to measure vertical response data in real time. The data acquisition module collects vertical response data measured by sensors, the data conversion module converts the collected vertical response data, and the data communication module transmits the converted vertical response data to the data storage module for storage. Figure 3As shown, the measuring vehicle system of this embodiment includes a measuring vehicle 10 and a tractor 11. The tractor 11 guides the measuring vehicle 10 to move, and the measuring vehicle 10 can rotate relative to the tractor 11 about a pivot 12 and move relative to the tractor 11 along a sliding shaft 13. The pivot 12 and the sliding shaft 13 are perpendicular to each other. In this embodiment, the pivot 12 is rotatably engaged with the tractor 11, and a sliding sleeve is provided on the pivot 12. The sliding shaft 13 is slidably engaged with the sliding sleeve and is fixedly connected to the measuring vehicle 10. The measuring vehicle 10 includes a carriage 16, and an axle 14 is installed on the carriage 16. The axle 14 is perpendicular to both the pivot 12 and the sliding shaft 13, and the sliding shaft 13 is fixedly installed on the carriage 16. In this embodiment, an onboard sensor 15 for collecting the vertical response of the measuring vehicle 10 during its movement on the measured road surface is arranged at the center of the axle 14. Of course, in other embodiments, the onboard sensor 15 can also be arranged on the carriage 16 located directly above the center of the axle 14.

[0154] In this embodiment, the data analysis and processing platform is used to identify road surface unevenness based on the collected vertical response data. In this embodiment, the vehicle-mounted sensor 15 is a displacement sensor, and the data acquisition module collects the vertical displacement response measured by the displacement sensor.

[0155] In this embodiment, the data output and display terminal is used to output and display the calculation results of the data analysis and processing platform in real time.

[0156] The following describes the bridge pavement unevenness identification method based on finite vehicle response of the present invention with specific examples.

[0157] In numerical verification, the following methods are adopted: Figure 4 The mathematical model of the bridge under test is shown. Figure 4 The parameters for the bridge are set as follows: bridge length L = 25m, cross-sectional dimension A = 3.2m. 2 The bridge density is ρ = 4800 kg / m³ 3 The elastic modulus E = 2.75 × 10⁻⁶ 10 N / m 2 The measurement vehicle parameters are set as follows: vehicle stiffness k v =200kN / m, vehicle body mass m v =14,000 kg, moving speed v = 2 m / s. The simulation was performed using the Power Spectral Density (PSD) method defined by ISO 8608. ISO 8608 classifies road surface roughness into five different levels: A, E, and C, where A represents the best road surface condition and E represents the roughest. In this embodiment, the road surface roughness level is considered to be C.

[0158] To verify the robustness of the bridge pavement roughness identification method based on finite vehicle response in this embodiment, numerical studies were conducted to verify the identification effect of bridge pavement roughness under different vehicle speeds and different environmental noise interference. Furthermore, to verify the versatility of the bridge pavement roughness identification method based on finite vehicle response in this embodiment, the numerical study specifically simulated the scenario where the test vehicle was placed on the road surface. The roughness examples under the above three working conditions are as follows:

[0159] Condition 1: Bridge surface unevenness identified using the bridge pavement unevenness identification method based on finite vehicle response in this embodiment at different vehicle speeds, such as... Figure 5 As shown;

[0160] Working Condition 2: Bridge surface irregularities identified using the bridge pavement irregularity identification method based on finite vehicle response in this embodiment under different environmental noise conditions, such as... Figure 6 As shown;

[0161] Working Condition 3: Road surface unevenness identified using the bridge road surface unevenness identification method based on finite vehicle response in this embodiment in different application scenarios, such as... Figure 7 As shown.

[0162] Numerical verification results from the three operating conditions show that the calculated roughness agrees well with the theoretical value in both the time and frequency domains. Furthermore, the bridge pavement roughness identification method based on finite vehicle response in this embodiment has broad applicability, applicable not only to bridge pavement roughness identification but also to the roughness identification of other pavements. The roughness identification method of the pavement roughness identification system in this embodiment is efficient, simple in process, and accurate in results. It can provide new technical support for the identification of bridge pavement roughness in a large number of bridge structures, serving bridge health monitoring, operation management, and maintenance.

[0163] Furthermore, this embodiment can be well extended to load identification problems in other fields, such as vehicle driving comfort assessment.

[0164] The above-described embodiments are merely preferred embodiments provided to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are all within the scope of protection of the present invention. The scope of protection of the present invention is defined by the claims.

Claims

1. A method for identifying bridge pavement irregularities based on finite vehicle response, characterized in that: The steps include the following: Step 1: Place the vehicle-mounted sensor at the center of the measuring vehicle's axle or on the vehicle body directly above the center of the measuring vehicle's axle; Step 2: Use a tractor to guide the measuring vehicle to drive at a constant speed across the bridge to be measured. During the process of the measuring vehicle traveling on the bridge, the signal acquisition system records the vertical response of the measuring vehicle. Step 3: Construct the equilibrium equations for the vehicle and the bridge, based on the defined state vectors. The state-space equations of the vehicle-bridge coupled system are obtained, and then the state-space equations of the vehicle-bridge coupled system are discretized. Step 4: Using the acceleration response as the observation value, calculate the vehicle acceleration. Discretize the representation; Step 5: Use Time-state vector express State vector at time step ,get Vehicle displacement at any given moment ; Step Six: Consider the observation noise that exists in reality. An improved system observation equation is obtained; Step 7: Solve for the road surface roughness using the extended Kalman filter algorithm. ; In step three, the equilibrium equations for the vehicle and the bridge are: in: , and These are the bridge's mass matrix, damping matrix, and stiffness matrix, respectively. For the quality of the vehicle; k v For vehicle stiffness; This represents the vertical displacement of the vehicle. The vertical acceleration of the vehicle; Let be the displacement vector of the bridge node, with dimension . , The number of degrees of freedom of the bridge; The velocity vector of the bridge node; This represents the acceleration vector of the bridge node; The road surface unevenness to be identified; A row vector containing the Hamilton interpolation function; The interaction force between the vehicle and the bridge; and: F in: For Hamilton interpolation function; It is the acceleration due to gravity; In step three, the state vector Defined as: in: Indicates the vehicle's vertical velocity; The state-space equation of the vehicle-axle coupled system is: in: State vector Regarding time The first derivative; Represents discrete time; and These are the state matrix and input matrix of the state equation, respectively; Given the load matrix; and: The discretized state-space equations of the vehicle-axle coupled system are as follows: in: It is the identity matrix; This indicates system noise.

2. The bridge pavement roughness identification method based on finite vehicle response according to claim 1, characterized in that: The vehicle-mounted sensor is a displacement sensor. In step two, the signal acquisition system records the vertical displacement response of the measuring vehicle.

3. The bridge pavement roughness identification method based on finite vehicle response according to claim 1, characterized in that: In step four, the vehicle acceleration for: Acceleration of the vehicle Discretization is represented as: in: Indicates the output matrix; This represents the transfer matrix.

4. The bridge pavement roughness identification method based on finite vehicle response according to claim 3, characterized in that: In step five, State vector at time step for: Vehicle displacement at any given moment for: in: express The submatrix corresponding to the first row of the matrix; express Matrix second to The submatrix corresponding to the row; express Matrix number The submatrix corresponding to the row; express Matrix number to The submatrix corresponding to the row; express The submatrix corresponding to the first row of the matrix; express Matrix second to The submatrix corresponding to the row; express Matrix number The submatrix corresponding to the row; express Matrix number to The submatrix corresponding to each row.

5. The bridge pavement roughness identification method based on finite vehicle response according to claim 4, characterized in that: In step six, the improved system observation equation is: = in: Represents the improved observation vector; Indicates observation noise; For the improved output matrix; For the improved transfer matrix.

6. The bridge pavement roughness identification method based on finite vehicle response according to claim 5, characterized in that: In step seven, the road surface unevenness is obtained. The steps are as follows: 71) Estimating initial state settings: in: and These are the initial values ​​for the state vector and the unknown input, respectively. and These are the initial estimates of the state vector and the unknown input, respectively; The initial value for the state vector covariance; Represents the mathematical expectation; 72) Time update phase: in: These represent the state vectors calculated in the current iteration step and the previous iteration step, respectively. Represents the state prediction value The error covariance matrix; Represents the state prediction value The error covariance matrix; express Time-based system noise vector The covariance matrix; 73) Calculate the Kalman gain : in: This represents the output matrix obtained from the previous iteration. Indicates observation noise The covariance matrix; 74) Estimating unknown inputs: in: Indicates unknown incentives The error covariance matrix; This represents the transfer matrix obtained from the previous iteration. Represents the improved observation vector; Represents the identity matrix; 75) Measurement Update Phase: in: and They are state vectors and unknown input The estimated value.

Images