Road network axle load identification method and device based on crowd sensing and medium

Abstract

The invention relates to a road network axle load identification method and device based on crowd sensing and a medium, and the method comprises the steps: collecting dynamic response data through a sensor based on crowd sensing, and transmitting the data to a controller for road network axle load identification; the controller repeatedly executes the following steps until all road sections with unknown pavement structure parameters and vehicles with unknown axle loads are calibrated: acquiring first dynamic response data of the vehicles with known axle loads on the road sections with unknown pavement structure parameters; calibrating pavement structure parameters of the road section with unknown pavement structure parameters based on the first dynamic response data, and obtaining a mapping relation between the vehicle axle load and the dynamic response; second dynamic response data caused by the vehicle with the unknown axle load on the road section with the calibrated known pavement structure parameters are obtained, the axle load of the vehicle with the unknown axle load is calibrated based on the second dynamic response data and the mapping relation, and the equipment and the medium are used for achieving the method. Compared with the prior art, the method improves the calculation efficiency and the stability and reliability of axle load identification.

Description

Technical Field

[0001] The present invention relates to the field of road engineering, and in particular to a road network axle load identification method, equipment and medium based on crowd intelligence perception. Background Art

[0002] Accurately identifying and monitoring vehicle axle loads is crucial for the maintenance and management of road projects. Traditional axle load monitoring methods rely on manual measurement or fixed weighing equipment, which has problems such as high cost, low efficiency, and limited data coverage. In existing technologies, road axle load monitoring usually uses static weighing or pressure sensor-based methods. These methods have limitations in practical applications. Static weighing equipment cannot provide continuous monitoring data, while pressure sensor-based methods may cause inaccurate data due to sensor accuracy and durability issues. In addition, these methods often require the installation of equipment at specific locations, which cannot achieve comprehensive monitoring of the entire road network.

[0003] In the field of axle load identification, there are already a large number of freight vehicles with known axle loads in the existing road network. Therefore, by quickly deploying sensing nodes for the dynamic response of the pavement structure in the road network and realizing parameter estimation and calibration through vehicle-road interaction, axle load identification at the road network level can be achieved. However, similar methods currently still need to be improved in terms of the accuracy and reliability of axle load identification. Especially in a complex and changeable road network environment, how to accurately identify vehicle axle loads and update pavement structure parameters in real time is a problem that needs to be solved urgently in the current technical field. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a road network axle load identification method, equipment and medium based on crowd intelligence perception.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] With the development of Internet of Things technology, crowd sensing, as an emerging data collection method, provides new possibilities for road axle load monitoring. The core of crowd sensing is to utilize a large number of distributed sensor networks and realize the perception and analysis of core elements through spontaneous organization and interaction between sensors. In the field of axle load identification, there are already a large number of freight vehicles with known axle loads in the existing road network. Therefore, by quickly deploying perception nodes of the dynamic response of the pavement structure in the road network and realizing parameter estimation and calibration through vehicle-road interaction, axle load identification at the road network level can be achieved.

[0007] Based on this, the present invention provides a road network axle load identification method based on crowd intelligence perception. The method is based on crowd intelligence perception, uses sensors to collect dynamic response data, and sends it to a controller for road network axle load identification. The controller repeatedly performs the following steps until all sections with unknown road structure parameters and vehicles with unknown axle loads are calibrated:

[0008] Obtaining the first dynamic response data caused by a vehicle with a known axle load to a road section with unknown road structure parameters;

[0009] Calibrate the pavement structure parameters of the road section with unknown pavement structure parameters based on the first dynamic response data, and obtain the mapping relationship between the vehicle axle load and the dynamic response;

[0010] The second dynamic response data caused by the vehicle with unknown axle load on the road section with calibrated known road surface structural parameters is obtained, and the axle load of the vehicle with unknown axle load is calibrated based on the second dynamic response data and the mapping relationship.

[0011] As a preferred technical solution, the method for obtaining the first kinetic response data is:

[0012] Acquire the position information and dynamic response data of the sensor, and acquire the trajectory information of the known axle-loaded vehicle within the preset sensing range of the sensor based on the position information;

[0013] Acquire a time period when the vehicle passes the sensor based on the trajectory information, and extract first intermediate data that matches the time period from the dynamic response data;

[0014] Continuously calculating and adjusting the first intermediate data until the number of peaks of the first intermediate data is equal to the number of axles of the vehicle with a known axle load, thereby obtaining second intermediate data;

[0015] Data within a preset range in the second intermediate data is intercepted, and the data is the first dynamic response data.

[0016] As a preferred technical solution, the calculation adjustment method is: calculating the first intermediate data using a short-time energy calculation method, and continuously adjusting the calculation time window of the short-time energy.

[0017] As a preferred technical solution, the preset range refers to the starting point of the first peak to the end point of the last peak.

[0018] As a preferred technical solution, the method for calibrating the pavement structure parameters of a road section with unknown pavement structure parameters is:

[0019] Constructing a response data set based on the first dynamic response data, and constructing an axle load feature set corresponding to the response data set based on the axle load data of the vehicle with known axle load;

[0020] Extracting the spectrum distribution based on the response data set, and generating a real spectrum distribution feature data set based on the spectrum distribution;

[0021] Obtain design data of a road section with unknown pavement structure parameters, construct an initial model of a multi-layer pavement structure based on the design data, and determine the initial value of a vector of unknown structural parameters;

[0022] Based on the axle load feature set, the real spectrum distribution feature data set and the undetermined structural parameter vector, the following steps are repeatedly performed:

[0023] Obtaining upper and lower bounds of pavement structure parameters based on the undetermined structure parameter vectors, and randomly generating a batch of undetermined structure parameter vectors based on the upper and lower bounds;

[0024] Based on the batch of undetermined structural parameter vectors, finite element models of the multi-layer pavement structure are respectively constructed, the axle load feature set is used as the input of the finite element model, and the dynamic response parameters are calculated;

[0025] Extracting the undetermined frequency spectrum distribution characteristics based on the dynamic response parameters;

[0026] Calculate the error between the pending frequency spectrum distribution feature and the real frequency spectrum distribution feature, and if the errors corresponding to the batch of pending structural parameter vectors all meet the preset error threshold, select the pending structural parameter vector corresponding to the minimum error as the pavement structure parameter;

[0027] If any of the errors corresponding to the batch of undetermined structural parameter vectors does not meet the preset error threshold, the undetermined structural parameter vectors corresponding to the second smallest error value and the third smallest error value are selected as the upper and lower bounds of the next batch of undetermined structural parameter vectors, respectively.

[0028] As a preferred technical solution, the method for obtaining the mapping relationship between the vehicle axle load and the dynamic response is:

[0029] Calculating third dynamic response data using a corresponding finite element model based on the calibrated pavement structure parameters;

[0030] The first vibration energy of the specified frequency band of the third dynamic response data is extracted, and the relationship between the vibration energy and the vehicle axle load is solved by a regression fitting method based on the first vibration energy.

[0031] As a preferred technical solution, the method for calibrating the axle load of a vehicle with unknown axle load is:

[0032] Obtaining the mapping relationship between vehicle axle load and vibration energy in multiple road sections with calibrated pavement structure parameters;

[0033] Acquire the second dynamic response data of the unknown axle-loaded vehicle on each road section with calibrated road surface structural parameters, and extract the second vibration energy at the frequency band corresponding to the first vibration energy frequency band in each second dynamic response data;

[0034] Based on the second vibration energy, the axle load of the vehicle with unknown axle load is estimated by using the corresponding mapping relationship to obtain an estimated value of the axle load of the vehicle with unknown axle load in a plurality of sections with calibrated road surface structure parameters;

[0035] Fitting multiple axle load estimation values ​​to obtain an expected value, wherein the expected value is the final axle load data.

[0036] As a preferred technical solution, the controller further performs the following steps:

[0037] Select multiple calibrated pavement structure parameters and vehicle axle load parameters, fit them using Gaussian distribution, and calculate the variance based on the fitting results. If the variance meets the preset threshold, the calibration is considered reliable.

[0038] If the variance does not meet the preset threshold, the calibration is considered unreliable and the pavement structure parameters and vehicle axle loads need to be recalibrated.

[0039] According to a second aspect of the present invention, there is provided an electronic device for road network axle load identification, comprising a memory and a processor, wherein a computer program is stored in the memory, and the above method is implemented when the processor executes the program.

[0040] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein the program implements the above method when executed by a processor.

[0041] Compared with the prior art, the present invention has the following advantages:

[0042] 1) The present invention adopts a crowd-sensing method to realize the rapid calibration of road structure parameters, making full use of the interaction between vehicles and roads in the road network, realizing rapid estimation of axle loads of large-scale road networks, and completing comprehensive detection of the entire road network without installing equipment in specific places. This not only improves the estimation efficiency of axle loads of road networks, but also solves the problems of high cost and time-consuming and labor-intensive traditional methods;

[0043] 2) The present invention proposes a semi-supervised learning vehicle-road cycle calibration method, which reduces the errors in the single estimation of pavement structure parameter calibration and vehicle axle load through continuous iterative calculation, improves the estimation accuracy, and makes full use of the interaction between the vehicle and the road to realize the autonomous calibration of vehicle axle load and pavement structure parameters, thereby improving the stability and reliability of road network axle load identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of group intelligence perception of the present invention;

[0045] Figure 2 This is a schematic diagram of the vehicle-road autonomous calibration of the present invention;

[0046] Figure 3 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0048] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0049] Example 1

[0050] This embodiment provides a road network axle load identification method based on crowd intelligence perception. The method is based on crowd intelligence perception, uses multiple embedded sensors buried in the target road network to collect dynamic response data, and sends it to the controller in a semi-supervised learning manner of vehicle-road interaction to achieve the calibration of vehicle axle loads and road network structure parameters of the entire road network. Specifically, this embodiment is as follows Figure 1The structure shown implements crowd intelligence perception. The detailed steps of the method include:

[0051] In the expressway network, 100 sections were selected and corresponding sensors were buried. The sensors were vibration sensors with a collection frequency of 100 Hz embedded in the road structure layer to obtain the dynamic response of the vehicle under the excitation of the axle load in real time.

[0052] Define the vehicle set V with known axle load according to the actual road network conditions load , road set R with known pavement structure parameters modulus , the vehicle set U with unknown axle load parameters v-load and the road set U with unknown pavement structure parameters r-modulus , where the vehicle set U with unknown axle load parameters is v-load and the road set U with unknown pavement structure parameters r-modulus It can be dynamically expanded based on actual road network data.

[0053] Define a set of roads R where known vehicles pass through unknown roads sample ={R1,R2,......,R n} and unknown vehicles passing through the known road vehicle set V sample = {V1, V2, ..., V i}, where R n Indicates that the nth unknown road has been passed by a known vehicle, V i represents the i-th unknown axle load vehicle passing through a road with known pavement structure parameters. When the set R sample The number of samples exceeds the preset value T1 = 15 and the set V sample When the number of samples exceeds the preset value T2 = 10, Figure 2 The loop calibration architecture shown is performed as follows Figure 3 The method of the flowchart shown in FIG. 1 is performed until all U r-modulus The samples in are calibrated and transferred to R modulus , the detailed steps are:

[0054] S1. Obtaining the first dynamic response data caused by a vehicle with a known axle load to a road section with unknown road structure parameters.

[0055] S11. According to the Beidou / GPS information of the embedded sensor buried location, the maximum collection range is set to 10m, the vehicle trajectory information passing through the sensor is filtered out, and the dynamic response data is extracted.

[0056] S12, based on the trajectory information, intercepting the time period when the vehicle passes the sensor, extracting the data in the dynamic response data that matches the time period, and further filtering out the dynamic response data of the vehicle acting on the corresponding sensor as the first intermediate data according to the distribution of the signal peaks.

[0057] S13, performing short-time energy calculation on the first intermediate data and adjusting the short-time energy calculation time window until the number of peaks of the first intermediate data is equal to the number of axles of the vehicle with known axle load, thereby obtaining second intermediate data.

[0058] S14, intercepting the data from the starting point of the first signal peak to the ending point of the last signal peak in the second intermediate data, and the data is the first dynamic response data.

[0059] S2. Calibrate the pavement structure parameters of the road section with unknown pavement structure parameters based on the first dynamic response data, and obtain the mapping relationship between the vehicle axle load and the dynamic response.

[0060] S21. Calibrate the pavement structure parameters of the road section with unknown pavement structure parameters.

[0061] When the number of vehicles passing through a road with unknown pavement structure parameters reaches 15, the pavement structure parameters are calibrated and the road with calibrated pavement structure parameters is put into the set R sample In and build with R sample Corresponding pavement structure parameter sample set R cal , the detailed steps are as follows:

[0062] S211. Construct a response data set R based on the first dynamic response data, and construct an axle load feature set L corresponding to the response data set based on the axle load data of the vehicle with known axle load.

[0063] S212: extract the spectrum distribution of each response data in the set based on the response data set R, and generate a real spectrum distribution feature data set R based on the spectrum distribution. f .

[0064] S213, obtaining design data of the road section with unknown pavement structure parameters, constructing an initial model of the multi-layer pavement structure based on the design data, and determining the initial value P0 of the undetermined structural parameter vector = [E 01 ,E 02 ,……,E 0m ], where E 01 ,E 02 ,……,E 0m It represents the initial values ​​of the parameters of the 1st to mth structural layers, and all of them are elastic moduli.

[0065] S214, repeatedly performing the following steps based on the axle load feature set, the real spectrum distribution feature data set, and the undetermined structural parameter vector:

[0066] A1. Obtain the upper and lower bounds of the pavement structure parameters based on the undetermined structure parameter vectors, and randomly generate a batch of undetermined structure parameter vectors [P1, P2, P3, ..., P x ].

[0067] A2. Based on the batch of undetermined structural parameter vectors, finite element models of the multi-layer pavement structure are respectively constructed, and the axle load feature set L is used as the input of the finite element model to calculate the dynamic response parameters.

[0068] A3. Extract the unknown spectrum distribution characteristics based on the dynamic response parameters.

[0069] A4. Calculate the error between the undetermined spectral distribution characteristics and the true spectral distribution characteristics. If any of the errors corresponding to the batch of undetermined structural parameter vectors does not meet the preset error threshold, select the undetermined structural parameter vectors corresponding to the second smallest error and the third smallest error as the upper and lower bounds of the next batch of undetermined structural parameter vectors, respectively, to generate the next batch of undetermined structural parameter vectors for feedback optimization. The feedback optimization process can use one or more of the particle swarm algorithm and the genetic algorithm. If the errors corresponding to the batch of undetermined structural parameter vectors all meet the preset error threshold, select the undetermined structural parameter vector corresponding to the minimum error as the pavement structure parameter, and put the calibrated pavement structure parameter into R cal middle.

[0070] S22. Obtain a mapping relationship between the vehicle axle load and the dynamic response.

[0071] S221. Calculate third dynamic response data using a corresponding finite element model based on the calibrated pavement structure parameters.

[0072] S222. Extract the vibration energy in the third dynamic response data frequency band between 0.5 and 20 Hz, and solve the relationship between the vibration energy and the vehicle axle load based on the vibration energy using methods such as linear fitting and polynomial fitting.

[0073] S3. Obtain second dynamic response data of the vehicle with unknown axle load on the road section with calibrated known road surface structure parameters, and calibrate the axle load of the vehicle with unknown axle load based on the second dynamic response data and the mapping relationship.

[0074] When a vehicle with unknown axle load passes through more than 10 sections with calibrated pavement structure parameters, the vehicle axle load is calibrated and the vehicle is placed in the set V after the calibration is completed. sample In and build with V sample The corresponding vehicle axle load set V cal , the detailed steps are as follows:

[0075] S31. Obtain a mapping relationship between vehicle axle load and vibration energy in a plurality of road sections with calibrated pavement structure parameters.

[0076] S32, obtaining the second dynamic response data of the unknown axle-loaded vehicle on each road section with calibrated road surface structural parameters, and extracting the vibration energy in the frequency band between 0.5 and 20 Hz in each second dynamic response data.

[0077] S33. Estimate the axle load of the vehicle with unknown axle load based on the vibration energy using a corresponding mapping relationship, and obtain an estimated value of the axle load of the vehicle with unknown axle load in a plurality of road sections with calibrated road surface structure parameters.

[0078] S34, perform normal distribution fitting on multiple axle load estimation values, use the expected value obtained by fitting as the final axle load data, and put it into the set V cal .

[0079] S4. From the set R cal and V cal Select multiple calibrated pavement structure parameters and vehicle axle load parameters from the unknown set U, use Gaussian distribution to fit them respectively, and calculate the variance based on the fitting results. If the variance meets the preset threshold, the calibration is considered reliable. The roads and vehicles corresponding to the calibrated pavement structure parameters and vehicle axle load parameters can be removed from the unknown set U. v-load and U r-modulus Transfer to V load and R modulus If the variance does not meet the preset threshold, the calibration is considered unreliable and the pavement structure and vehicle axle load of the road and vehicle corresponding to the calibrated pavement structure parameters and vehicle axle load parameters need to be recalibrated.

[0080] Example 2

[0081] The above is an introduction to the method embodiment. The following is a further description of the scheme of the present invention through the device embodiment. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process described can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0082] The present embodiment provides an electronic device for road network axle load identification, which includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or computer program instructions loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The CPU, ROM and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0083] Multiple components in the device are connected to the I / O interface, including: input units, such as keyboards, mice, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0084] The processing unit performs the various methods and processes described above, such as methods S1 to S4 and A1 to A4. For example, in some embodiments, methods S1 to S4 and A1 to A4 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of methods S1 to S4 and A1 to A4 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S4 and A1 to A4 in any other appropriate manner (e.g., by means of firmware).

[0085] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0086] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0087] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0088] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A road network axle load identification method based on crowd intelligence perception, characterized in that: The method is based on crowd sensing, using sensors to collect dynamic response data and send it to the controller for road network axle load identification. The controller repeatedly performs the following steps until all sections with unknown road structure parameters and vehicles with unknown axle loads are calibrated: Obtaining the first dynamic response data caused by a vehicle with a known axle load to a road section with unknown road structure parameters; Calibrate the pavement structure parameters of the road section with unknown pavement structure parameters based on the first dynamic response data, and obtain the mapping relationship between the vehicle axle load and the dynamic response; The second dynamic response data caused by the vehicle with unknown axle load on the road section with calibrated known road surface structural parameters is obtained, and the axle load of the vehicle with unknown axle load is calibrated based on the second dynamic response data and the mapping relationship.

2. According to the method for identifying axle loads of a road network based on crowd intelligence perception according to claim 1, it is characterized in that: The method for obtaining the first kinetic response data is: Acquire the position information and dynamic response data of the sensor, and acquire the trajectory information of the known axle-loaded vehicle within the preset sensing range of the sensor based on the position information; Acquire a time period when the vehicle passes the sensor based on the trajectory information, and extract first intermediate data that matches the time period from the dynamic response data; Continuously calculating and adjusting the first intermediate data until the number of peaks of the first intermediate data is equal to the number of axles of the vehicle with a known axle load, thereby obtaining second intermediate data; Data within a preset range in the second intermediate data is intercepted, and the data is the first dynamic response data.

3. The method for identifying axle loads of a road network based on crowd intelligence perception according to claim 2 is characterized in that: The calculation adjustment method is: calculating the first intermediate data using a short-time energy calculation method, and continuously adjusting the calculation time window of the short-time energy.

4. The method for identifying axle loads of a road network based on crowd intelligence perception according to claim 2 is characterized in that: The preset range refers to the starting point of the first peak to the end point of the last peak.

5. The method for identifying axle loads of a road network based on crowd intelligence perception according to claim 1 is characterized in that: The method for calibrating the pavement structure parameters of a road section with unknown pavement structure parameters is as follows: Constructing a response data set based on the first dynamic response data, and constructing an axle load feature set corresponding to the response data set based on the axle load data of the vehicle with known axle load; Extracting the spectrum distribution based on the response data set, and generating a real spectrum distribution feature data set based on the spectrum distribution; Obtain design data of a road section with unknown pavement structure parameters, construct an initial model of a multi-layer pavement structure based on the design data, and determine the initial value of a vector of unknown structural parameters; Based on the axle load feature set, the real spectrum distribution feature data set and the undetermined structural parameter vector, the following steps are repeatedly performed: Obtaining upper and lower bounds of pavement structure parameters based on the undetermined structure parameter vectors, and randomly generating a batch of undetermined structure parameter vectors based on the upper and lower bounds; Based on the batch of undetermined structural parameter vectors, finite element models of the multi-layer pavement structure are respectively constructed, the axle load feature set is used as the input of the finite element model, and the dynamic response parameters are calculated; Extracting the undetermined frequency spectrum distribution characteristics based on the dynamic response parameters; Calculate the error between the pending frequency spectrum distribution feature and the real frequency spectrum distribution feature, and if the errors corresponding to the batch of pending structural parameter vectors all meet the preset error threshold, select the pending structural parameter vector corresponding to the minimum error as the pavement structure parameter; If any of the errors corresponding to the batch of undetermined structural parameter vectors does not meet the preset error threshold, the undetermined structural parameter vectors corresponding to the second smallest error value and the third smallest error value are selected as the upper and lower bounds of the next batch of undetermined structural parameter vectors, respectively.

6. The method for identifying axle loads of a road network based on crowd intelligence perception according to claim 5 is characterized in that: The method for obtaining the mapping relationship between the vehicle axle load and the dynamic response is: Calculating third dynamic response data using a corresponding finite element model based on the calibrated pavement structure parameters; The first vibration energy of the specified frequency band of the third dynamic response data is extracted, and the relationship between the vibration energy and the vehicle axle load is solved by a regression fitting method based on the first vibration energy.

7. The method for identifying axle loads of a road network based on crowd intelligence perception according to claim 1 is characterized in that: The method for calibrating the axle load of a vehicle with unknown axle load is: Obtaining the mapping relationship between vehicle axle load and vibration energy in multiple road sections with calibrated pavement structure parameters; Acquire the second dynamic response data of the unknown axle-loaded vehicle on each road section with calibrated road surface structural parameters, and extract the second vibration energy at the frequency band corresponding to the first vibration energy frequency band in each second dynamic response data; Based on the second vibration energy, the axle load of the vehicle with unknown axle load is estimated by using the corresponding mapping relationship to obtain an estimated value of the axle load of the vehicle with unknown axle load in a plurality of sections with calibrated road surface structure parameters; Fitting multiple axle load estimation values ​​to obtain an expected value, wherein the expected value is the final axle load data.

8. The method for identifying axle loads of a road network based on crowd intelligence perception according to claim 1 is characterized in that: The controller further performs the following steps: Select multiple calibrated pavement structure parameters and vehicle axle load parameters, fit them using Gaussian distribution, and calculate the variance based on the fitting results. If the variance meets the preset threshold, the calibration is considered reliable. If the variance does not meet the preset threshold, the calibration is considered unreliable and the pavement structure parameters and vehicle axle loads need to be recalibrated.

9. An electronic device for road network axle load identification, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images