A method, device, equipment and storage medium for detecting bridge vehicle travel time

Abstract

The embodiments of the present application provide a method, device, equipment and storage medium for detecting the vehicle travel time of a bridge, which relates to the field of deep learning. Random traffic samples, noise data and a bridge coupled vibration analysis model are obtained to establish a random traffic bridge response theory sample library. A response theory curve image is generated according to the response theory sample library and annotated to obtain a response training sample. Model training is performed according to the response training sample to obtain a target curve detection model. Dynamic response data is collected in real time. A dynamic response curve image is generated according to the dynamic response data and the corresponding relationship between the pixel coordinates of the image and the vehicle travel time is determined. The dynamic response curve image is input into the target curve detection model, and the pixel coordinates of the dynamic response curve image are output. The vehicle travel time is calculated according to the pixel coordinates, which reduces the complexity of data collection, reduces costs, improves work efficiency, and improves the detection accuracy of bridge health monitoring.

Description

Technical Field

[0001] The present invention relates to the field of deep learning technology, and in particular to a method, device, equipment and storage medium for detecting bridge vehicle travel time. Background Art

[0002] In the bridge health monitoring system, it is very important to monitor the travel time of vehicles passing through the bridge. The accurate travel time is directly related to the safety assessment and maintenance of the bridge.

[0003] At present, the existing technology usually adopts setting thresholds to detect the time when a vehicle passes through a bridge or using sensors for auxiliary positioning, such as ground induction coils, radars, cameras, and acoustic sensors. However, detection through thresholds is easily affected by interference information, which reduces the detection accuracy and cannot guarantee the accuracy of the collected data. Assisted positioning through sensors requires expensive installation and maintenance costs, and when the sensor fails or is damaged, the data cannot be guaranteed to be accurate. In addition, the linkage and data fusion of multiple sensors increase the complexity of the system and reduce the safety assessment and maintenance efficiency of the bridge. Therefore, there is an urgent need for a method that can simply, efficiently and accurately monitor the passage time of vehicles passing through bridges. Summary of the invention

[0004] In view of this, the purpose of the present invention is to overcome the deficiencies in the prior art and to provide a method, device, equipment and storage medium for detecting the passage time of vehicles on a bridge.

[0005] The present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present disclosure provides a method for detecting a bridge vehicle travel time, the method comprising:

[0007] Obtain random traffic samples and noise data of the bridge to be detected, establish a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establish a random traffic bridge response theoretical sample library according to the random traffic samples, noise data and bridge coupling vibration analysis model;

[0008] Generate a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and annotate the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample;

[0009] Performing neural network model training according to the random traffic flow bridge response training samples to obtain a target curve detection deep learning model;

[0010] According to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected, dynamic response data is collected in real time, a bridge dynamic response curve image is generated according to the dynamic response data, and a corresponding relationship between pixel coordinates of the bridge dynamic response curve image and vehicle travel time is determined;

[0011] Inputting the bridge dynamic response curve image into the target curve detection deep learning model, and outputting the pixel coordinates of the bridge dynamic response curve image;

[0012] The vehicle travel time of the bridge to be detected is calculated according to the pixel coordinates of the bridge dynamic response curve image.

[0013] According to a specific implementation method disclosed in the present application, the bridge coupling vibration analysis model includes a vibration model of the bridge to be detected, a vehicle vibration equation and an interaction force model between the bridge and the vehicle. The steps of obtaining random traffic samples and noise data of the bridge to be detected, establishing the bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establishing a random traffic bridge response theory sample library according to the random traffic samples, noise data and the bridge coupling vibration analysis model include:

[0014] Collecting the noise data according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected;

[0015] Applying the random traffic flow samples to the vibration model, the vehicle vibration equation and the bridge-vehicle interaction force model to obtain a random traffic flow bridge response sample library;

[0016] The noise data is fused with the random traffic flow bridge response sample library based on a data fusion algorithm to obtain a random traffic flow bridge response theoretical sample library.

[0017] According to a specific implementation method disclosed in the present application, the step of generating a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and annotating the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample includes:

[0018] Obtaining the category number, center point coordinates, curve height and curve width of the random traffic flow bridge response theory curve image;

[0019] The category number, center point coordinates, curve height and curve width of the random traffic flow bridge response theoretical curve image are annotated according to a preset data format to obtain a random traffic flow bridge response training sample.

[0020] According to a specific implementation method disclosed in the present application, the step of training a neural network model according to the random traffic bridge response training sample to obtain a target curve detection deep learning model includes:

[0021] Performing data enhancement processing on the random traffic flow bridge response training samples to obtain enhanced random traffic flow bridge response training samples;

[0022] Performing data division on the enhanced random traffic flow bridge response training samples to obtain enhanced random traffic flow bridge response training data and enhanced random traffic flow bridge response test data;

[0023] The YOLO image model was trained based on the enhanced random traffic bridge response training data to obtain a preliminary target curve detection deep learning model;

[0024] The preliminary target curve detection deep learning model is iteratively optimized using enhanced random traffic bridge response test data to obtain the target curve detection deep learning model.

[0025] According to a specific implementation method disclosed in the present application, the steps of collecting dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be detected, generating a bridge dynamic response curve image according to the dynamic response data and determining the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time include:

[0026] Performing denoising, filtering and normalization processing on the dynamic response data to obtain pre-processed dynamic response data;

[0027] Taking time as the horizontal axis and the dynamic response data as the vertical axis, the pre-processed dynamic response data is used to generate a bridge dynamic response curve image;

[0028] The corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time is defined as , where T is the time range and W is the width of the bridge dynamic response curve image.

[0029] According to a specific implementation method disclosed in the present application, the step of inputting the bridge dynamic response curve image into the target curve detection deep learning model and outputting the pixel coordinates of the bridge dynamic response curve image includes:

[0030] Performing convolution processing on the bridge dynamic response image to obtain an image feature map;

[0031] Generate a bounding box on the image feature map and predict whether the bounding box includes a target response curve, wherein the prediction result includes center point coordinates, width, height, and confidence of the target category;

[0032] The bounding boxes with low confidence of the target category are filtered to obtain a bridge dynamic response curve image including the target response curve, and the pixel coordinates of the bridge dynamic response curve image are output, wherein the pixel coordinates include the center point coordinates of the bounding box, the width and height of the bounding box, and the pixel coordinates of the upper left corner and the lower right corner of the bounding box.

[0033] According to a specific implementation method disclosed in the present application, the step of calculating the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image includes:

[0034] Calculation formula based on vehicle travel time , , Calculate the vehicle travel time of the bridge to be detected, where: is the X coordinate of the upper left corner of the bounding box, is the X coordinate of the lower right corner of the bounding box, T is the time range, and W is the width of the bridge dynamic response curve image.

[0035] In a second aspect, an embodiment of the present disclosure provides a device for detecting a bridge vehicle travel time, the device comprising:

[0036] An acquisition module is used to acquire random traffic samples and noise data of the bridge to be detected, establish a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establish a random traffic bridge response theory sample library according to the random traffic samples, noise data and bridge coupling vibration analysis model;

[0037] A labeling module, used for generating a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and labeling the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample;

[0038] A training module, used for training a neural network model according to the random traffic bridge response training samples to obtain a target curve detection deep learning model;

[0039] An acquisition module, used for acquiring dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be detected, generating a bridge dynamic response curve image according to the dynamic response data and determining the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time;

[0040] An output module, used for inputting the bridge dynamic response curve image into the target curve detection deep learning model, and outputting the pixel coordinates of the bridge dynamic response curve image;

[0041] The calculation module is used to calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image.

[0042] In a third aspect, an electronic device is provided in an embodiment of the present disclosure, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the method for detecting the bridge vehicle passage time described in any one of the first aspects when executing the computer program.

[0043] In a fourth aspect, a computer-readable storage medium is provided in an embodiment of the present disclosure, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for detecting the bridge vehicle passage time described in any one of the first aspects are implemented.

[0044] The above-mentioned method for detecting the vehicle travel time of a bridge provided by the present application obtains random traffic samples and noise data of the bridge to be detected, establishes a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, establishes a random traffic flow bridge response theory sample library according to the random traffic flow samples, noise data and bridge coupling vibration analysis model, generates a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, annotates the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample, trains a neural network model according to the random traffic flow bridge response training sample to obtain a target curve detection deep learning model, and installs a neural network model on the bridge to be detected according to the random traffic flow bridge response training sample to obtain a target curve detection deep learning model. The installed dynamic displacement sensor, accelerometer and dynamic strain gauge collect dynamic response data in real time, generate a bridge dynamic response curve image according to the dynamic response data, and determine the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time, input the bridge dynamic response curve image into the target curve detection deep learning model, output the pixel coordinates of the bridge dynamic response curve image, and calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image, thereby reducing the complexity of data collection, reducing costs, and improving work efficiency. The vehicle travel time is calculated by detecting the vehicle travel curve through the deep learning model, thereby improving the detection accuracy of bridge health monitoring.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and understandable, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work. In each of the drawings, similar components are numbered similarly.

[0047] Figure 1 A schematic flow chart of a method for detecting a bridge vehicle travel time provided in an embodiment of the present application is shown;

[0048] Figure 2 A schematic diagram showing the structure of a device for detecting the bridge vehicle travel time provided in an embodiment of the present application is shown;

[0049] Figure 3 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0051] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. In contrast, when an element is referred to as being "directly on" another element, there is no intermediate element. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0052] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0053] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of the template are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0055] Example 1

[0056] like Figure 1 FIG. 1 is a flow chart of a method for detecting a bridge vehicle passing time provided in an embodiment of the present application. The method for detecting a bridge vehicle passing time provided in an embodiment of the present application comprises the following steps:

[0057] Step S101, obtaining random traffic samples and noise data of a bridge to be detected, establishing a bridge coupling vibration analysis model of the bridge to be detected based on a finite element method, and establishing a random traffic bridge response theoretical sample library according to the random traffic samples, noise data and the bridge coupling vibration analysis model.

[0058] Specifically, random traffic samples refer to a large amount of typical random traffic data with the characteristics of the bridge to be tested, which is output by using the Monte Carlo method and a random traffic simulation analysis program written in a programming language after obtaining the types and proportions of vehicles on the bridge to be tested. The noise data refers to the sensing data collected when there are no vehicles passing by the dynamic displacement sensors, accelerometers and dynamic strain gauges installed on the bridge to be tested. This type of data mainly comes from the bridge response caused by external factors such as environmental vibration, wind load and temperature change. The noise data can reflect the natural response characteristics of the bridge under the condition of no traffic load.

[0059] The bridge coupling vibration analysis model includes a vibration model of the bridge to be detected, a vehicle vibration equation and an interaction force model between the bridge and the vehicle. The steps of obtaining random traffic samples and noise data of the bridge to be detected, establishing the bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establishing a random traffic bridge response theory sample library according to the random traffic samples, noise data and the bridge coupling vibration analysis model include:

[0060] Collecting the noise data according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected;

[0061] Applying the random traffic flow samples to the vibration model, the vehicle vibration equation and the bridge-vehicle interaction force model to obtain a random traffic flow bridge response sample library;

[0062] The noise data is fused with the random traffic flow bridge response sample library based on a data fusion algorithm to obtain a random traffic flow bridge response theoretical sample library.

[0063] Preferably, the vibration model of the bridge to be monitored is expressed as:

[0064]

[0065] The vehicle vibration equation of the bridge to be monitored is expressed as:

[0066]

[0067] The interaction force model between the bridge to be monitored and the vehicle is expressed as:

[0068]

[0069] in, is the vehicle speed, For vehicle quality, is the vehicle interval time, N is the total number of vehicles in the random traffic flow, w(x,t) is the displacement of the bridge at position x and time t, E is the Young's modulus of the beam material, I is the section moment of inertia of the beam, ρ is the density of the beam, A is the cross-sectional area of ​​the beam, q(x,t) is the distributed load that varies along the beam length x and time t, (t) is the vertical displacement of the vehicle, is the damping coefficient of the vehicle, is the spring stiffness of the vehicle, g is the acceleration due to gravity, is the Dirac delta function.

[0070] Step S102: generating a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and annotating the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample.

[0071] Specifically, the theoretical response data in the random traffic flow bridge response theoretical sample library is used as the vertical axis, and the vehicle travel time is used as the horizontal axis to draw a curve image to form a random traffic flow bridge response theoretical curve image library. The curve model generated when the vehicle passes on the bridge to be detected is obtained by annotation, so as to obtain the random traffic flow bridge response training sample.

[0072] The step of generating a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and annotating the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample comprises:

[0073] Obtaining the category number, center point coordinates, curve height and curve width of the random traffic flow bridge response theory curve image;

[0074] The category number, center point coordinates, curve height and curve width of the random traffic flow bridge response theoretical curve image are annotated according to a preset data format to obtain a random traffic flow bridge response training sample.

[0075] Preferably, the annotation file of the random traffic bridge response theory curve image is a plain text file, each annotation file corresponds to a random traffic bridge response theory curve image, each line of the file represents the annotation information of a random traffic bridge response theory curve image, and the annotation information contains five fields, namely, the category number, the X coordinate of the center point of the curve, the Y coordinate of the center point of the curve, the curve width and the curve height. The category number refers to the category of the random traffic bridge response theory curve, which is an integer starting from 0. The center point X coordinate and the center point Y coordinate are the ratio of the image width and height, and the value range is [0,1]. If the random traffic bridge response theory curve image file is named response_curve1.jpg, the corresponding annotation file is response_curve1.txt. When there are two vehicle traffic response curves in the image, the content of the annotation file is 0, 0.45, 0.55, 0.1, 0.2 and 1, 0.30, 0.40, 0.15, 0.25.

[0076] Step S103: Perform neural network model training based on the random traffic bridge response training samples to obtain a target curve detection deep learning model.

[0077] It can be understood that by training the neural network model with the labeled random traffic bridge response training samples, the model can detect the curve range with the training sample characteristics from the image, that is, the bridge response curve range when there are vehicles passing.

[0078] The step of training a neural network model according to the random traffic bridge response training sample to obtain a target curve detection deep learning model includes:

[0079] Performing data enhancement processing on the random traffic flow bridge response training samples to obtain enhanced random traffic flow bridge response training samples;

[0080] Performing data division on the enhanced random traffic flow bridge response training samples to obtain enhanced random traffic flow bridge response training data and enhanced random traffic flow bridge response test data;

[0081] The YOLO image model was trained based on the enhanced random traffic bridge response training data to obtain a preliminary target curve detection deep learning model;

[0082] The preliminary target curve detection deep learning model is iteratively optimized using enhanced random traffic bridge response test data to obtain the target curve detection deep learning model.

[0083] Specifically, the data enhancement processing includes scaling processing, translation processing and shearing processing, and the diversity of random traffic bridge response training samples can be improved through enhancement processing. Preferably, the PyTorch framework is used for training, and the network architecture is designed based on the YOLO network to obtain a preliminary target curve detection deep learning model, which can perform bounding box prediction and classification on the image. In order to prevent the preliminary target curve detection deep learning model from overfitting, cross-validation processing, early stopping processing and regularization processing are performed to obtain the target curve detection deep learning model and evaluate its performance. The precision, recall rate and F1 score are used to verify the performance of the target curve detection deep learning model on the enhanced random traffic bridge response test data.

[0084] Step S104, collecting dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected, generating a bridge dynamic response curve image according to the dynamic response data and determining the correspondence between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time.

[0085] Specifically, the dynamic displacement sensor is used to measure the displacement change of the bridge to be detected when a vehicle passes through, the accelerometer is used to measure the acceleration response of the bridge to be detected, and the dynamic strain gauge is used to measure the strain change of the bridge to be detected. The vibration, stress, displacement and other response data of the bridge are collected by setting the acquisition frequency of the sensor to ensure data accuracy.

[0086] The step of collecting dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected, generating a bridge dynamic response curve image according to the dynamic response data and determining the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time includes:

[0087] Performing denoising, filtering and normalization processing on the dynamic response data to obtain pre-processed dynamic response data;

[0088] Taking time as the horizontal axis and the dynamic response data as the vertical axis, the pre-processed dynamic response data is used to generate a bridge dynamic response curve image;

[0089] The corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time is defined as , where T is the time range and W is the width of the bridge dynamic response curve image.

[0090] Step S105: input the bridge dynamic response curve image into the target curve detection deep learning model, and output the pixel coordinates of the bridge dynamic response curve image.

[0091] Specifically, the target curve detection deep learning model is used to detect the response curve of the vehicle passing through the bridge in the bridge dynamic response curve image, and the pixel coordinates of the response curve are returned.

[0092] The step of inputting the bridge dynamic response curve image into the target curve detection deep learning model and outputting the pixel coordinates of the bridge dynamic response curve image comprises:

[0093] Performing convolution processing on the bridge dynamic response image to obtain an image feature map;

[0094] Generate a bounding box on the image feature map and predict whether the bounding box includes a target response curve, wherein the prediction result includes center point coordinates, width, height, and confidence of the target category;

[0095] The bounding boxes with low confidence of the target category are filtered to obtain a bridge dynamic response curve image including the target response curve, and the pixel coordinates of the bridge dynamic response curve image are output, wherein the pixel coordinates include the center point coordinates of the bounding box, the width and height of the bounding box, and the pixel coordinates of the upper left corner and the lower right corner of the bounding box.

[0096] Specifically, the dynamic response image of the bridge is scaled and filled and then input into the target curve detection deep learning model. The feature information in the dynamic response image of the bridge is extracted through the convolutional neural network structure, and non-maximum suppression is used to ensure that the bounding box of the final output is non-overlapping and covers the possible target area of ​​the image.

[0097] Step S106, calculating the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image.

[0098] The step of calculating the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image comprises:

[0099] Calculation formula based on vehicle travel time , , Calculate the vehicle travel time of the bridge to be detected, where: is the X coordinate of the upper left corner of the bounding box, is the X coordinate of the lower right corner of the bounding box, T is the time range, and W is the width of the bridge dynamic response curve image.

[0100] In specific implementation, the noise and abnormal data of sensor data will affect the detection of vehicle travel time. By collecting bridge dynamic response data in real time, adversarial training is performed on the target curve detection deep learning model to improve the model's detection capability and robustness, as well as its stability and accuracy in complex environments.

[0101] The above-mentioned method for detecting the vehicle travel time of a bridge provided by the present application obtains random traffic samples and noise data of the bridge to be detected, establishes a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, establishes a random traffic flow bridge response theory sample library according to the random traffic flow samples, noise data and bridge coupling vibration analysis model, generates a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, annotates the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample, trains a neural network model according to the random traffic flow bridge response training sample to obtain a target curve detection deep learning model, and uses a dynamic displacement sensor installed on the bridge to be detected to obtain a target curve detection deep learning model. , accelerometers and dynamic strain gauges collect dynamic response data in real time, generate a bridge dynamic response curve image according to the dynamic response data and determine the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time, input the bridge dynamic response curve image into the target curve detection deep learning model, output the pixel coordinates of the bridge dynamic response curve image, and calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image, thereby reducing the use of sensors, reducing the complexity and use cost of data collection, and improving the work efficiency of bridge health and safety monitoring. The vehicle travel time is calculated by detecting the vehicle travel curve through the deep learning model, thereby improving the detection accuracy of bridge health monitoring.

[0102] Example 2

[0103] like Figure 2 FIG. 2 is a schematic diagram of a bridge vehicle travel time detection device 200 provided in an embodiment of the present application, wherein the device comprises:

[0104] The acquisition module 201 is used to acquire random traffic samples and noise data of the bridge to be detected, establish a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establish a random traffic bridge response theory sample library according to the random traffic samples, noise data and the bridge coupling vibration analysis model;

[0105] A labeling module 202 is used to generate a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and label the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample;

[0106] A training module 203 is used to perform neural network model training according to the random traffic bridge response training samples to obtain a target curve detection deep learning model;

[0107] The acquisition module 204 is used to acquire dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be detected, generate a bridge dynamic response curve image according to the dynamic response data, and determine the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time;

[0108] An output module 205 is used to input the bridge dynamic response curve image into the target curve detection deep learning model, and output the pixel coordinates of the bridge dynamic response curve image;

[0109] The calculation module 206 is used to calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image.

[0110] The bridge vehicle passing time detection device 200 provided in this embodiment can implement the bridge vehicle passing time detection method shown in Example 1, and will not be described again to avoid repetition.

[0111] The detection device for bridge vehicle travel time provided by the embodiment of the present application obtains random traffic samples and noise data of the bridge to be detected, establishes a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, establishes a random traffic flow bridge response theory sample library according to the random traffic flow samples, noise data and bridge coupling vibration analysis model, generates a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, annotates the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample, trains a neural network model according to the random traffic flow bridge response training sample to obtain a target curve detection deep learning model, and installs a neural network model on the bridge to be detected according to the random traffic flow bridge response training sample to obtain a target curve detection deep learning model. The installed dynamic displacement sensor, accelerometer and dynamic strain gauge collect dynamic response data in real time, generate a bridge dynamic response curve image according to the dynamic response data, and determine the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time, input the bridge dynamic response curve image into the target curve detection deep learning model, output the pixel coordinates of the bridge dynamic response curve image, and calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image, thereby reducing the complexity of data collection, reducing costs, and improving work efficiency. The vehicle travel time is calculated by detecting the vehicle travel curve through the deep learning model, thereby improving the detection accuracy of bridge health monitoring.

[0112] Example 3

[0113] In addition, an embodiment of the present disclosure provides an electronic device 300, including a memory and a processor, wherein the memory stores a computer program.

[0114] For details, see Figure 3 The electronic device 300 includes: a receiver 301, a bus interface and a processor 302.

[0115] In the embodiment of the present invention, the electronic device 300 further includes a memory 303. Figure 3 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 302 and memory represented by memory 303. The bus architecture may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The receiver 301 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. The processor 302 is responsible for managing the bus architecture and general processing, and the memory 303 may store data used by the processor 302 when performing operations.

[0116] The electronic device 300 provided in the embodiment of the present invention can implement the method for detecting the bridge vehicle passing time shown in Embodiment 1, which will not be described again to avoid repetition.

[0117] The electronic device provided in the embodiment of the present application obtains random traffic samples and noise data of the bridge to be detected, establishes a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, establishes a random traffic flow bridge response theory sample library according to the random traffic flow samples, noise data and bridge coupling vibration analysis model, generates a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, annotates the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample, performs neural network model training according to the random traffic flow bridge response training sample, obtains a target curve detection deep learning model, and performs deep learning on the dynamic bridge response image installed on the bridge to be detected. The displacement sensor, accelerometer and dynamic strain gauge collect dynamic response data in real time, generate a bridge dynamic response curve image according to the dynamic response data, determine the correspondence between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time, input the bridge dynamic response curve image into the target curve detection deep learning model, output the pixel coordinates of the bridge dynamic response curve image, and calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image, thereby reducing the complexity of data collection, reducing costs, and improving work efficiency. The vehicle travel time is calculated by detecting the vehicle travel curve through the deep learning model, thereby improving the detection accuracy of bridge health monitoring.

[0118] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and structure diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the structure diagram and / or the flow chart, and the combination of boxes in the structure diagram and / or the flow chart, can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0119] In addition, the functional modules or units in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0120] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0121] The above description is only a specific implementation mode 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 changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for detecting the vehicle travel time on a bridge, characterized in that: The method comprises: Obtain random traffic samples and noise data of the bridge to be detected, establish a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establish a random traffic bridge response theoretical sample library according to the random traffic samples, noise data and bridge coupling vibration analysis model; Generate a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and annotate the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample; Performing neural network model training according to the random traffic flow bridge response training samples to obtain a target curve detection deep learning model; According to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected, dynamic response data is collected in real time, a bridge dynamic response curve image is generated according to the dynamic response data, and a corresponding relationship between pixel coordinates of the bridge dynamic response curve image and vehicle travel time is determined; Inputting the bridge dynamic response curve image into the target curve detection deep learning model, and outputting the pixel coordinates of the bridge dynamic response curve image; The vehicle travel time of the bridge to be detected is calculated according to the pixel coordinates of the bridge dynamic response curve image.

2. The method for detecting the bridge vehicle travel time according to claim 1, characterized in that: The bridge coupling vibration analysis model includes a vibration model of the bridge to be detected, a vehicle vibration equation and an interaction force model between the bridge and the vehicle. The steps of obtaining random traffic samples and noise data of the bridge to be detected, establishing the bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establishing a random traffic bridge response theory sample library according to the random traffic samples, noise data and the bridge coupling vibration analysis model include: Collecting the noise data according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected; Applying the random traffic flow samples to the vibration model, the vehicle vibration equation and the bridge-vehicle interaction force model to obtain a random traffic flow bridge response sample library; The noise data is fused with the random traffic flow bridge response sample library based on a data fusion algorithm to obtain a random traffic flow bridge response theoretical sample library.

3. The method for detecting the bridge vehicle travel time according to claim 1, characterized in that: The step of generating a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and annotating the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample comprises: Obtaining the category number, center point coordinates, curve height and curve width of the random traffic flow bridge response theory curve image; The category number, center point coordinates, curve height and curve width of the random traffic flow bridge response theoretical curve image are annotated according to a preset data format to obtain a random traffic flow bridge response training sample.

4. The method for detecting the bridge vehicle travel time according to claim 1, characterized in that: The step of training a neural network model according to the random traffic bridge response training sample to obtain a target curve detection deep learning model includes: Performing data enhancement processing on the random traffic flow bridge response training samples to obtain enhanced random traffic flow bridge response training samples; Performing data division on the enhanced random traffic flow bridge response training samples to obtain enhanced random traffic flow bridge response training data and enhanced random traffic flow bridge response test data; The YOLO image model was trained based on the enhanced random traffic bridge response training data to obtain a preliminary target curve detection deep learning model; The preliminary target curve detection deep learning model is iteratively optimized using enhanced random traffic bridge response test data to obtain the target curve detection deep learning model.

5. The method for detecting the bridge vehicle travel time according to claim 1, characterized in that: The step of collecting dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be inspected, generating a bridge dynamic response curve image according to the dynamic response data and determining the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time includes: Performing denoising, filtering and normalization processing on the dynamic response data to obtain pre-processed dynamic response data; Taking time as the horizontal axis and the dynamic response data as the vertical axis, the pre-processed dynamic response data is used to generate a bridge dynamic response curve image; The corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time is defined as , where T is the time range and W is the width of the bridge dynamic response curve image.

6. The method for detecting the bridge vehicle travel time according to claim 1, characterized in that: The step of inputting the bridge dynamic response curve image into the target curve detection deep learning model and outputting the pixel coordinates of the bridge dynamic response curve image comprises: Performing convolution processing on the bridge dynamic response image to obtain an image feature map; Generate a bounding box on the image feature map and predict whether the bounding box includes a target response curve, wherein the prediction result includes center point coordinates, width, height, and confidence of the target category; The bounding boxes with low confidence of the target category are filtered to obtain a bridge dynamic response curve image including the target response curve, and the pixel coordinates of the bridge dynamic response curve image are output, wherein the pixel coordinates include the center point coordinates of the bounding box, the width and height of the bounding box, and the pixel coordinates of the upper left corner and the lower right corner of the bounding box.

7. The method for detecting the bridge vehicle travel time according to claim 1, characterized in that: The step of calculating the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image comprises: Calculation formula based on vehicle travel time , , Calculate the vehicle travel time of the bridge to be detected, where: is the X coordinate of the upper left corner of the bounding box, is the X coordinate of the lower right corner of the bounding box, T is the time range, and W is the width of the bridge dynamic response curve image.

8. A device for detecting the vehicle travel time on a bridge, characterized in that: The device comprises: An acquisition module is used to acquire random traffic samples and noise data of the bridge to be detected, establish a bridge coupling vibration analysis model of the bridge to be detected based on the finite element method, and establish a random traffic bridge response theory sample library according to the random traffic samples, noise data and bridge coupling vibration analysis model; A labeling module, used for generating a random traffic flow bridge response theory curve image according to the random traffic flow bridge response theory sample library, and labeling the random traffic flow bridge response theory curve image to obtain a random traffic flow bridge response training sample; A training module, used for training a neural network model according to the random traffic bridge response training samples to obtain a target curve detection deep learning model; An acquisition module, used for acquiring dynamic response data in real time according to the dynamic displacement sensor, accelerometer and dynamic strain gauge installed on the bridge to be detected, generating a bridge dynamic response curve image according to the dynamic response data and determining the corresponding relationship between the pixel coordinates of the bridge dynamic response curve image and the vehicle travel time; An output module, used for inputting the bridge dynamic response curve image into the target curve detection deep learning model, and outputting the pixel coordinates of the bridge dynamic response curve image; The calculation module is used to calculate the vehicle travel time of the bridge to be detected according to the pixel coordinates of the bridge dynamic response curve image.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method for detecting the bridge vehicle passage time according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for detecting the bridge vehicle passage time according to any one of claims 1 to 7 are implemented.

Images