A system and method for monitoring and evaluating small and medium span bridge structures

Abstract

The application provides a small and medium span bridge structure monitoring and evaluation system and method, and belongs to the technical field of bridge engineering. The small and medium span bridge structure monitoring and evaluation system comprises a data acquisition and transmission module and a data processing module, and the monitoring and evaluation method comprises data acquisition and transmission and processing of corresponding data; the application realizes acquisition and transmission of bridge data through the data acquisition and transmission module, processes the transmitted data through the data processing module, realizes dynamic evaluation and real-time early warning of the bearing capacity of the bridge according to the processed data, thereby providing a bridge preventive maintenance management scheme, facilitating maintenance of the good state of the bridge, reducing the safety risk of the bridge, and ensuring the safety of road network operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bridge engineering, in particular to a small and medium span bridge structure monitoring and evaluation system and method. BACKGROUND

[0002] As a controlling engineering in traffic transportation, the bridge is the key to keep the road unobstructed. However, in the actual use process, due to the influence of its own structure and external environment, any form and any material of the bridge will inevitably cause structural degradation, leading to the failure of the bridge component, the reduction of the bearing capacity, and even the collapse of the bridge. Therefore, ensuring the good state of the bridge has become the research direction of the industry.

[0003] At present, the existing bridges include small and medium span bridges, large bridges and super large bridges. The safety monitoring technology of large and super large bridges has been mature in China. However, due to the particularity of the monitoring demand and the restriction of economic factors, the safety monitoring technology of large and super large bridges cannot be directly used for small and medium span bridges. There is a lack of technology for monitoring and evaluating the structural safety of small and medium span bridges, which is not conducive to the bearing capacity evaluation and safety warning of small and medium span bridges during operation, and there is a lack of scientific maintenance and management scheme for bridges, which has a great safety risk. SUMMARY

[0004] In view of the above problems existing in the prior art, the present application provides a small and medium span bridge structure monitoring and evaluation system and method to collect and process bridge data by including a data acquisition and transmission module and a data processing module, realize dynamic evaluation and real-time warning of the bearing capacity of the bridge, provide a preventive maintenance and management scheme for the bridge, and provide convenience for ensuring the good state of the bridge and reducing the safety risk of the bridge.

[0005] The specific technical solutions are as follows:

[0006] A small and medium span bridge structure monitoring and evaluation system has the following characteristics, comprising:

[0007] The data acquisition and transmission module comprises a sensor module, a data acquisition module and a transmission module. The sensor module has a plurality of sensors, and the plurality of sensors are arranged on the required part of the bridge to be monitored. The plurality of sensors are electrically connected to the data acquisition module, and the data acquisition module is electrically connected to the transmission module.

[0008] The data processing module is in communication connection with the transmission module, used for processing the monitoring signals of each sensor transmitted by the transmission module, and establishing a database for the processed data. At the same time, the evaluation and safety warning are made.

[0009] A small and medium span bridge structure monitoring and evaluation method comprises the following steps:

[0010] Step S1, data acquisition and transmission;

[0011] The sensor data collected by each sensor in the sensor module is transmitted to the data acquisition module through an independent cable, and the data acquisition module transmits the sensor data to the data processing module through the transmission module;

[0012] Step S2, data processing;

[0013] Step S21, load deduction in the road network;

[0014] Taking the structural state of part of the bridges in a regional road network as sample data, the state of all bridges in the entire road network is evaluated;

[0015] Step S22, network-level information fusion;

[0016] A bridge information database is established to store the structural state information of the bridge, and the representation objects of the structural state of the bridge include the entire bridge, a beam, and a plate. The data of the structural state information of the bridge is expressed in "attributes" and "attribute values". Each object instantiated from the "attribute" class defines a feature of a specific entity type, and each object instantiated from the "attribute value" class defines the corresponding data of the domain entity belonging to the entity type;

[0017] Step S23, hierarchical early warning;

[0018] Select an alarm threshold and set the secondary and primary warning limits;

[0019] Step S24, comprehensive evaluation;

[0020] S241, state comprehensive evaluation;

[0021] The structure of the bridge is decomposed layer by layer in order of main substructure, components, and members, each sub-item is scored, and then the scores of each sub-item are aggregated into the overall state evaluation of the bridge by setting a weight coefficient;

[0022] S242, bearing capacity evaluation;

[0023] Through analysis of the collected video image data, the vehicle position information on the road is obtained, and the combination of vehicle type recognition and dynamic weighing is realized to obtain multi-dimensional information of vehicle position, vehicle type, and vehicle weight. The load distribution state of the vehicle on the bridge deck is obtained, and the obtained vehicle load on the bridge deck is applied to the bridge. Combined with finite element analysis calculation, the theoretical load is obtained. At the same time, the theoretical load and the measured sensor data are compared to realize online evaluation of the safety of the bridge.

[0024] Step S25, maintenance decision suggestion;

[0025] According to the importance of the components of the bridge structure and the comprehensive evaluation, the management and maintenance decision of the target bridge in the same year is optimized.

[0026] The method for monitoring and evaluating the small and medium span bridge structure, wherein the load deduction in the road network takes the load as the input of the bridge, the load includes the environmental load and the vehicle load, and the environmental load is the temperature load.

[0027] The method for monitoring and evaluating the small and medium span bridge structure, wherein the vehicle load deduction includes a traffic distribution model, an OD matrix estimation and a flow measurement point layout, and

[0028] The traffic distribution model takes the bridge as the representation of the road section, takes the vehicle as the travel unit, takes the road section travel time as the impedance, takes the road section traffic volume as the flow, and analyzes the interaction relationship between the impedance and the flow by using the "equilibrium" analysis method.

[0029] The OD matrix estimation estimates the unknown OD matrix by the observed road section traffic volume and the prior information, and the prior information includes the historical OD matrix.

[0030] The flow measurement point layout needs to consider the information supplement, the number of measurement points and the coverage problem, the information supplement is to supplement the data of other road sections to the used road section flow information, the number of measurement points and the coverage are the minimum number of flow measurement points and the layout needs to meet the OD matrix coverage principle, to ensure that the travel between any OD points can be observed by at least one flow measurement point.

[0031] The method for monitoring and evaluating the small and medium span bridge structure, wherein in the hierarchical early warning process, the determination of the alarm threshold is based on the extreme statistical theory, first, the hour maximum value distribution model of the monitoring quantity is established, then the month maximum value and the year maximum value distribution models of the monitoring quantity are deduced, and the expected level of the month response extreme value and the year response extreme value are selected as the secondary and primary early warning limits respectively.

[0032] The method for monitoring and evaluating the small and medium span bridge structure, wherein the state comprehensive evaluation is a comprehensive evaluation of the superstructure of the target bridge, which is composed of two sub-levels of detection report and monitoring information, the evaluation based on the detection report is weighted according to the three indexes of the super load bearing component, the general load bearing component and the support, and the evaluation based on the monitoring information is according to the direct index and the core index.

[0033] The method for monitoring and evaluating the structure of a medium-small span bridge, wherein, in the process of bearing capacity evaluation, the video image data is obtained by a binocular stereo vision method, including the processes of image acquisition, camera calibration, feature extraction, feature matching and disparity generation, depth perception, and interpolation reconstruction, the image acquisition process takes the binocular images collected by the binocular camera as input, adopts a video acquisition method, and takes into account the contrast of the target background, the performance indicators of the camera under the condition of illumination, and the shooting scene; the camera calibration process calculates the intrinsic parameters of the camera and the rotation and translation matrix between the binocular cameras by an algorithm; the feature extraction process extracts the corner points in the collected binocular images and the extreme points under the action of different scale blurs; the feature matching and disparity generation process matches the same features between the corner points and the extreme points extracted in the feature extraction process to find the homonymous points in the binocular images; the depth perception process calculates the three-dimensional coordinates of the objects in space according to the principle of triangular approximation measurement; and the interpolation reconstruction process fits the complete surface information of the object by interpolation when the three-dimensional coordinates calculated in the depth perception process are incomplete.

[0034] The method for monitoring and evaluating the structure of a medium-small span bridge, wherein, in the process of bearing capacity evaluation, the vehicle type recognition adopts a Faster-RCNN deep learning algorithm based on a ZF framework to classify and recognize the vehicle types passing through the bridge, and the passing vehicle types are classified into large passenger cars, medium passenger cars, small passenger cars, large sedans, medium sedans, and small sedans.

[0035] The method for monitoring and evaluating the structure of a medium-small span bridge, wherein, in the process of bearing capacity evaluation, the theoretical load and the measured sensor data are compared to obtain a verification coefficient When the verification coefficient is lower than a preset limit value, it is determined that the bridge is within a safe range, and the calculation formula of the verification coefficient is as follows:

[0036]

[0037] In the above formula, is the measured sensor data; is the theoretical load.

[0038] The method for monitoring and evaluating the structure of a medium-small span bridge, wherein, in the process of data acquisition and processing, the sensor data further includes artificial inspection data.

[0039] The positive effects of the above technical solutions are:

[0040] The small and medium span bridge structure monitoring and evaluation system and method have the monitoring and evaluation system comprising the data acquisition and transmission module, the data processing module, realize the acquisition, transmission and processing of the bridge data, and dynamically evaluate and real-time early warn the bridge bearing capacity according to the processing result, provide the data reference for the bridge preventive maintenance management scheme, facilitate the bridge to maintain the good state, reduce the safety risk of the bridge, and benefit the normal operation of the road network. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 It is a structure diagram of an embodiment of the small and medium span bridge structure monitoring and evaluation system of the application.

[0042] Figure 2 It is a flow chart of the small and medium span bridge structure monitoring and evaluation method of the application.

[0043] In the drawings: 1, sensor module; 2, data acquisition module; 3, transmission module; 4, data processing module. DETAILED DESCRIPTION

[0044] In order to make the technical means, creative features, purposes and effects realized by the application easy to understand, the following embodiments are combined with the accompanying drawings to further describe the application. Figure 1 to the accompanying drawings Figure 2 The technical solutions provided by the application are described in detail, but the following content is not limited to the application.

[0045] Figure 1 It is a structure diagram of an embodiment of the small and medium span bridge structure monitoring and evaluation system of the application. As shown in the figure, Figure 1 The small and medium span bridge structure monitoring and evaluation system provided by the embodiment comprises a data acquisition and transmission module 3, the data acquisition and transmission module further comprises a sensor module 1, a data acquisition module 2 and a transmission module 3, and the sensor module 1 has a plurality of sensors, the plurality of sensors are arranged on the required parts of the bridge to be monitored, and the different parts of each component on the bridge to be monitored are monitored to obtain the measured sensing data. At this time, the plurality of sensors are electrically connected to the data acquisition module 2, that is, the measured sensing data collected by the sensor module 1 can be collected and transmitted to the data acquisition module 2. Preferably, the plurality of sensors and the data acquisition module 2 are connected by a limited star-shaped structure, the signals collected by each sensor are transmitted to the data acquisition module 2 through independent cables, at the same time, the data acquisition module 2 is electrically connected to the transmission module 3, the signals transmitted by the data acquisition module 2 are transmitted to the data processing module 4 through the transmission module 3,

[0046] Specifically, the data processing module 4 is in communication connection with the transmission module 3, preferably, the data processing module 4 can be a control computer, the transmission module 3 and the data processing module 4 are connected through a network, the monitoring signals collected by each sensor are analyzed and processed through the data processing module 4, and the processed data is established into a database, meanwhile, comprehensive evaluation and safety warning are made, which provides data reference for bridge maintenance decision suggestion, provides convenience for bridge maintenance in good state, effectively reduces the safety risk of the bridge, and is beneficial to the normal operation of the road network.

[0047] Figure 2 A flow chart of the method for monitoring and evaluating the small and medium span bridge structure according to the present application is shown in Fig. Figure 2 The present embodiment further provides a monitoring and evaluating method using the small and medium span bridge structure monitoring and evaluating system, which comprises the following steps:

[0048] Step S1, data acquisition and transmission;

[0049] The sensing data collected by each sensor in the sensor module 1 is transmitted to the data acquisition module 2 through an independent cable, and the sensing data is centrally processed through the data acquisition module 2 and then transmitted to the data processing module 4 through the transmission module 3, and the sensing data transmitted is analyzed and processed through the data processing module 4.

[0050] Step S2, data processing;

[0051] The data processing comprises load deduction in the road network, network-level information fusion, hierarchical early warning, comprehensive evaluation and maintenance decision suggestion process, wherein,

[0052] Step S21, load deduction in the road network;

[0053] The structural state of part of the bridges in a regional road network is taken as sample data, the state of all the bridges in the road network is evaluated, and the bridge structure network level is detected, evaluated and warned.

[0054] Specifically, the load deduction in the road network takes the load as the input of the bridge, at this time, the load includes the environmental load and the vehicle load, and the environmental load is the temperature load. Since the temperature of the bridge is directly affected by the environmental factors such as climate, sunshine and the like, and the distribution of the environmental factors in the road network is continuous, the distribution of the temperature of the bridge in the range of the road network is also continuous. It is worth pointing out that the temperature of the bridge fluctuates only in a limited range within a certain period of time, and the error existing in the theoretical model and the actual situation can be ignored. When obtaining the temperature data, part of the data points in the plane of the road network can be obtained, when the number of data points is sufficient and uniformly distributed in the region, the deduction of the temperature distribution is converted into a curve fitting problem, the least square algorithm is adopted, the sum of squares between the values of the fitted curve and the actual values at the sampling points is minimized, after the temperature distribution curve is determined, the structure temperature data of the unknown bridge can be analogically inferred, and the corresponding temperature load is obtained.

[0055] In addition, the vehicle load deduction in the load deduction in the road network further includes a traffic distribution model, OD matrix estimation and flow measurement point layout. Since the road network is composed of multiple road sections, and the bridge is a form of the road section, the traffic distribution model takes the bridge as a form of the road section, takes the vehicle as a travel unit, takes the road section travel time as impedance, takes the road section traffic volume as flow, and uses the "equilibrium" analysis method to analyze the interaction relationship between the impedance and the flow. It is worth pointing out that the "equilibrium" analysis method refers to finding the balance point of the characteristic / demand two functions in different environments in the transportation system, wherein the demand function is a relationship describing that the service quality decreases with the increase of the flow, and the characteristic function is a relationship describing that the flow increases with the improvement of the service quality, at this time, the independent variable is the congestion degree, and the increase of the congestion degree will lead to the increase of the impedance. At this time, the starting point O and the ending point D are set, there are multiple paths connecting the starting point O and the ending point D, and the number of travel units from the starting point O to the ending point D is known. If all the travel units select a certain path, the impedance on the path will increase until the path is no longer the optimal path, leading to the fact that the travel units will tend to select other paths, and the impedance of the selected path will also increase with the increase of the flow, and the cycle is repeated, and finally the equilibrium state of the traffic model is reached, that is, the application of the OD matrix in the road network to obtain the vehicle flow of the to-be-measured road section.

[0056] In addition, the OD matrix estimation is to estimate the unknown OD matrix through the observed road section traffic volume and prior information, and the prior information includes the historical OD matrix, fully utilizes the traffic volume data obtained by the detection or monitoring means, is more efficient and has a shorter period.

[0057] More specifically, since the bridge detection or monitoring data in the road network is an important source of vehicle flow data of the road to be tested, the flow measurement point layout for detection or monitoring needs to consider information supplement, measurement point number and coverage. At this time, information supplement is to supplement other road data to the road flow information used, to reduce the influence of the limited flow measurement points that can provide flow information in a road network due to manpower and material limitations, and to improve the accuracy of OD matrix estimation. The number and coverage of the measurement points are the minimum number of the measurement points and the layout needs to meet the OD matrix coverage principle to ensure that the travel between any OD points can be observed by at least one flow measurement point, so as to obtain reliable OD matrix estimation results.

[0058] Step S22, network-level information fusion;

[0059] A bridge information database is established to store the structural state information of the bridge. The characterization objects of the structural state of the bridge include the whole bridge, a beam, and a plate. Since there are similar attributes and mutual relationships between different characterization objects, the data of the structural state information of the bridge needs to be expressed as "attributes" and "attribute values" in the database. On the one hand, each object instantiated from the "attribute" class defines a feature of a specific entity type. On the other hand, each object instantiated from the "attribute value" class defines the corresponding data of a domain entity belonging to the entity type, which facilitates subsequent efficient information fusion. For example, the "attribute" can be defined as the "material" of the entity type "beam", and the "attribute value" can be located as the related data of a domain entity "No. 1 beam" belonging to the entity type "beam". It is worth pointing out that the domain entity can be understood as a container that stores all related information of an object in the real world.

[0060] Step S23, hierarchical early warning;

[0061] The alarm threshold is selected, and the secondary and primary warning limits are set. At this time, the determination of the alarm threshold is based on the extreme statistical theory. First, the hourly maximum value distribution model of the monitoring quantity is established, that is, the maximum value of the monitoring data in the time interval is calculated, a suitable time interval width is selected, and no matter how many extreme value events occur in the time interval, only the maximum value is entered into the subsequent statistical process. The preferred time interval width is one hour. Then, the monthly maximum value and annual maximum value distribution models of the monitoring quantity are inferred, and the expected levels of the monthly response extreme value and the annual response extreme value are selected as the secondary and primary warning limits, respectively, which provides a condition for timely issuing a warning when the bridge is at risk of damage.

[0062] Step S24, comprehensive evaluation;

[0063] S241, state comprehensive evaluation;

[0064] The structure of the bridge is decomposed layer by layer in order of main substructures, parts and components, each sub-item is scored, and the scores of each sub-item are aggregated into the overall state evaluation of the bridge by setting a weight coefficient. It is worth pointing out that the comprehensive evaluation of the bridge state in the embodiment is a comprehensive evaluation of the superstructure of the target bridge, and the comprehensive evaluation of the technical condition of the whole superstructure is composed of two sub-levels of detection reports and monitoring information. The evaluation based on the detection report is weighted according to three indexes of super load-bearing components, general load-bearing components and bearings, and the evaluation based on the monitoring information is according to direct indexes and core indexes. At this time, the direct indexes are collected physical quantities, including degree of disturbance, strain, displacement and frequency; the core indexes are advanced indexes obtained by extracting and analyzing the direct indexes, including correlation functions of the direct indexes and neutral axis positions. The hierarchical upward induction of each monitoring index is realized, and finally the comprehensive evaluation score of the superstructure of the bridge is obtained. It is worth pointing out that the state comprehensive evaluation provided in the embodiment is for the superstructure of the bridge, and the substructure and the bridge deck of the bridge are not within the scope of the monitoring and evaluation method provided in the embodiment, and there are relatively comprehensive detection means to judge them, and the original evaluation standard can be followed, and details are not repeated here.

[0065] S242, bearing capacity evaluation;

[0066] Through analysis of the collected video image data, the vehicle position information on the road is obtained, the vehicle position, vehicle type and vehicle weight multi-dimensional information are combined by combining vehicle type recognition and dynamic weighing, the vehicle load distribution state on the bridge deck is obtained, the obtained vehicle load on the bridge deck is applied to the bridge, the theoretical load is obtained by combining finite element analysis calculation, and the theoretical load and the measured sensor data of the sensor are compared to realize online evaluation of the safety of the bridge.

[0067] More specifically, the video image data is obtained by a binocular stereo vision method, including image acquisition, camera calibration, feature extraction, feature matching, disparity generation, depth perception, and interpolation reconstruction process. At this time, the image acquisition process takes the binocular images collected by the binocular camera as the input item, adopts a video acquisition method, and it is worth noting that in the acquisition process, the contrast of the target background, the performance indicators of the camera under the light condition, and the shooting scene need to be considered. In addition, the camera calibration process is a process of calculating the intrinsic parameters of the camera and the rotation and translation matrix between the binocular cameras. At this time, the intrinsic parameters of the camera include the internal parameters and the external parameters. The internal parameter calibration is obtained according to the pinhole imaging model of the camera. The external parameter calibration is the process of converting the coordinate information of the external world into a distance that can be understood by the computer with the camera, that is, the process of converting the world coordinate system into the camera coordinate system. The rotation and translation matrix between the binocular cameras is to extract the internal and external parameters of the left and right cameras respectively, find the internal corner point coordinates of the calibration graph corresponding to the left and right cameras, and combine the three-dimensional world coordinates, the internal parameter matrix and the distortion coefficient of the left and right cameras to obtain the rotation matrix, the translation matrix, the eigenmatrix and the basis matrix of the binocular camera. The feature extraction process is to extract the corner points in the collected binocular images and the extreme points under the action of different scale blurs. In addition, the feature matching and disparity generation process is to match the same features between the corner points and extreme points extracted in the feature extraction process to find the homonymic points in the binocular images. The depth perception process is a process of calculating the three-dimensional coordinates of objects in space according to the principle of triangular approximation measurement. The interpolation reconstruction process is a process of fitting the complete surface information of the object by interpolation when the three-dimensional coordinates calculated in the depth perception process are incomplete. At this time, the interpolation method refers to the whole process of using the characteristics of the adjacent points as sampling points to estimate the missing information, so as to restore the complete and continuous three-dimensional model of the object. After the above multiple processes, the extracted video image data is more accurate and has higher precision, which improves the accuracy of bridge monitoring and has higher reference value.

[0068] More specifically, the vehicle type recognition in the bearing capacity evaluation is to classify and recognize the vehicle types passing through the bridge by using the Faster-RCNN deep learning algorithm based on the ZF framework, to improve the accuracy of vehicle type recognition and further improve the accuracy of monitoring. Preferably, the passing vehicle types are divided into six types, namely large passenger cars, medium passenger cars, small passenger cars, large cars, medium cars and small cars, so that the vehicle types are covered more widely and the monitoring data obtained are more comprehensive.

[0069] More specifically, in the bearing capacity evaluation process, after comparing the theoretical load with the measured sensor data, the verification coefficient , can be obtained. When the verification coefficient If the value is lower than the preset limit, it is determined that the bridge is in a safe range, otherwise the bridge is in a damaged risk. At this time, the verification coefficient The calculation formula of the verification coefficient is as follows:

[0070]

[0071] In the above formula, is the measured sensing data; is the theoretical load.

[0072] Step S25, pipeline maintenance decision suggestion;

[0073] According to the importance of the components of the bridge structure and the comprehensive evaluation, the management and maintenance decision of the target bridge in the same year is optimized. For example, the load-bearing components and abutments of the superstructure of the plate bridge are main components, when the comprehensive evaluation score is between 88 and 90, the maintenance and reinforcement work of the abutments and load-bearing components of the superstructure should be focused on, and the pipeline maintenance budget is reasonably allocated and adjusted.

[0074] The small and medium span bridge structure monitoring and evaluation system and method provided by the embodiment, the monitoring and evaluation system comprises a data acquisition and transmission module 3 and a data processing module 4, and the monitoring and evaluation method comprises data acquisition and transmission and processing of corresponding data; the data acquisition and transmission module 3 realizes the acquisition and transmission of bridge data, and the data processing module 4 processes the transmitted data, realizes the dynamic evaluation and real-time early warning of the bridge bearing capacity according to the processed data, and provides a preventive maintenance management scheme for the bridge, facilitates the maintenance of the bridge in good condition, reduces the safety risk of the bridge, and ensures the safety of the road network operation.

[0075] The above is only the preferred embodiment of the present application, and does not limit the implementation and protection scope of the present application. For those skilled in the art, it should be realized that any equivalent replacement and obvious change made according to the content of the present application should be included in the protection scope of the present application.

Claims

1. A method for monitoring and evaluating small and medium span bridge structures, characterized by, The small and medium span bridge structure monitoring and evaluation system comprises a data acquisition and transmission module, a data processing module and a data storage module. The data processing module is in communication connection with the transmission module, used for processing the monitoring signals of each sensor transmitted by the transmission module, and establishing a database for the processed data, and meanwhile, making an evaluation and a safety early warning. The small and medium span bridge structure monitoring and evaluation method comprises the following steps: Step S1, data acquisition and transmission; The sensor module is connected with the data acquisition module through a cable. Step S2, data processing; Step S21, load deduction in a road network; The load deduction in a road network takes load as the input of the bridge, wherein the load comprises an environmental load and a vehicle load, and the environmental load is a temperature load. Step S22, network level information fusion; The bridge information database is established to store the structural state information of the bridge. Step S23, hierarchical early warning; The alarm threshold is selected to set the secondary and primary early warning limits. Step S24, comprehensive evaluation; S241, state comprehensive evaluation; The bridge structure is decomposed layer by layer in the order of main substructure, component and member, each subitem is scored, and the scores of the subitems are summarized into the overall state evaluation of the bridge through the setting of the weight coefficient. S242, bearing capacity evaluation; The vehicle position information is obtained through the analysis of the collected video image data, the vehicle position, vehicle type and vehicle weight multi-dimensional information are combined through the vehicle type recognition and dynamic weighing, the load distribution state of the vehicle on the bridge is obtained, the theoretical load is obtained through the finite element analysis combined with the bridge load of the vehicle, and the theoretical load and the measured sensor data are compared to realize the online evaluation of the bridge safety. Step S25, management and maintenance decision suggestion; According to the importance of the components of the bridge structure and the comprehensive evaluation, the management and maintenance decision of the target bridge in the same year is optimized.

2. The method for monitoring and evaluating small and medium span bridge structures according to claim 1, characterized in that, The deduction of the vehicle load comprises a traffic distribution model, an OD matrix estimation and a flow measurement point layout. The traffic assignment model takes the bridge as a road section, takes the vehicle as a travel unit, takes the road section travel time as impedance, takes the road section traffic volume as flow, and analyzes the interaction between impedance and flow by using the "equilibrium" analysis method; The OD matrix estimation is to estimate the unknown OD matrix by using the observed road section traffic volume and prior information, and the prior information includes historical OD matrix; The flow measurement point layout needs to consider information supplement, measurement point quantity and coverage, the information supplement is to supplement other road section data to the used road section flow information, and the measurement point quantity and coverage are the minimum number of flow measurement points and the layout needs to meet the OD matrix coverage principle to ensure that the travel between any OD points can be observed by at least one flow measurement point.

3. The method for monitoring and evaluating small and medium span bridge structures according to claim 1, characterized in that, In the hierarchical early warning process, the determination of the alarm threshold is based on the extreme statistical theory, first, the hourly maximum value distribution model of the monitoring quantity is established, then the monthly maximum value and annual maximum value distribution models of the monitoring quantity are deduced, and the expected levels of the monthly response extreme value and the annual response extreme value are selected as the secondary and primary warning limits respectively.

4. The method for monitoring and evaluating small and medium span bridge structures according to claim 1, characterized in that, The state comprehensive evaluation is a comprehensive evaluation of the superstructure of the target bridge, which is composed of two sub-levels of detection report and monitoring information, the evaluation based on the detection report is weighted according to three indexes of upper bearing member, general bearing member and support, and the evaluation based on the monitoring information is according to direct index and core index.

5. The method for monitoring and evaluating small and medium span bridge structures according to claim 1, characterized in that, In the bearing capacity evaluation process, the video image data is obtained by the method of binocular stereo vision, including image acquisition, camera calibration, feature extraction, feature matching to generate disparity, depth perception and interpolation reconstruction process, the image acquisition process takes the binocular images collected by the binocular camera as the input item, adopts the video acquisition method, and considers the contrast of the target background, the performance index of the camera under the light condition and the shooting scene; the camera calibration process calculates the inherent parameters of the camera and the rotation and translation matrix between the binocular cameras by algorithm; the feature extraction process extracts the corner points in the collected binocular images and the extreme points under the action of different scale blurring; The feature matching to generate disparity process is to match the same features between the corner points and extreme points extracted in the feature extraction process to find the homonymous points in the binocular images; the depth perception process is to calculate the three-dimensional coordinates of the objects in space according to the triangular approximation measurement principle; the interpolation reconstruction process is to fit the complete surface information of the object by interpolation when the three-dimensional coordinates calculated in the depth perception process are incomplete.

6. The method for monitoring and evaluating small and medium span bridge structures according to claim 5, characterized in that, The vehicle type recognition in the bearing capacity evaluation adopts the Faster-RCNN deep learning algorithm based on the ZF framework to classify and recognize the vehicle types passing through the bridge, and divides the passing vehicles into large passenger cars, medium passenger cars, small passenger cars, large cars, medium cars and small cars.

7. The method for monitoring and evaluating small and medium span bridge structures according to claim 5, characterized in that, The theoretical load in the bearing capacity evaluation process is compared with the measured sensor data to obtain a verification coefficient When the verification coefficient is lower than a preset limit value, it is determined that the bridge is within a safe range, and the calculation formula of the verification coefficient is ， In the above formula, is the measured sensor data; is the theoretical load.

8. The method for monitoring and evaluating small and medium span bridge structures according to claim 1, characterized in that, The sensor data in the data acquisition and processing process also includes artificial inspection data.

Images