CN113553749A - Bridge health monitoring method, system, computer and storage medium - Google Patents

Bridge health monitoring method, system, computer and storage medium Download PDF

Info

Publication number
CN113553749A
CN113553749A CN202111106958.5A CN202111106958A CN113553749A CN 113553749 A CN113553749 A CN 113553749A CN 202111106958 A CN202111106958 A CN 202111106958A CN 113553749 A CN113553749 A CN 113553749A
Authority
CN
China
Prior art keywords
damage
bridge
data
module
sensor
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111106958.5A
Other languages
Chinese (zh)
Inventor
黄笑犬
刘恒
丘建栋
游博雅
周子益
李梦蝶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Shenyan Urban Transportation Co ltd
Shenzhen Urban Transport Planning Center Co Ltd
Original Assignee
Shanghai Shenyan Urban Transportation Co ltd
Shenzhen Urban Transport Planning Center Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Shenyan Urban Transportation Co ltd, Shenzhen Urban Transport Planning Center Co Ltd filed Critical Shanghai Shenyan Urban Transportation Co ltd
Priority to CN202111106958.5A priority Critical patent/CN113553749A/en
Publication of CN113553749A publication Critical patent/CN113553749A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/08—Probabilistic or stochastic CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Abstract

The invention provides a bridge health monitoring method, a bridge health monitoring system, a computer and a storage medium, and belongs to the technical field of artificial intelligence monitoring. Firstly, establishing a bridge finite element numerical simulation model; secondly, performing wavelet packet decomposition on the time-course data obtained by the simulation model, and converting the time-course data into frequency band energy data; secondly, constructing a relative energy ratio index according to the frequency band energy, and calculating an energy ratio deviation damage index and an energy ratio variance damage index; finally, summarizing the damage indexes to generate multi-sensor characteristic vectors, training and learning the bridge structure by adopting an ensemble learning-based method, and further carrying out damage diagnosis on the sub-regions of the structure. The problem that structural damage information acquired by a large number of sensors for health monitoring in the prior art is lack of effective utilization is solved. The invention effectively utilizes all sensor information of the bridge arrangement to carry out data mining and searches the internal association between the structural damage and the dynamic response.

Description

Bridge health monitoring method, system, computer and storage medium
Technical Field
The application relates to bridge health monitoring, in particular to a bridge health monitoring method, a bridge health monitoring system, a bridge health monitoring computer and a storage medium, and belongs to the technical field of artificial intelligence monitoring.
Background
With the continuous development and progress of the traffic infrastructure construction industry, more and more long-span bridges are constructed. However, the design benchmark period of the bridge structure is long, and in the long-term operation process, the bridge structure may suffer from coupling effects of environmental corrosion, material aging, vehicle overload, traffic flow increase and the like, and the effects may cause damage accumulation on the structure, so that the resistance of the structure is reduced, the function is degraded, and if the structure cannot be found and repaired in time, the normal operation function of the structure may be greatly influenced. How to acquire health condition information in the bridge operation period in time and perform safety assessment on the self state becomes important through a health monitoring system, and the key is that the health condition of the structure is estimated by acquiring and acquiring corresponding structure information in real time from the bridge structure in the operation state.
The ideal bridge health monitoring system not only can play a role in guaranteeing the construction quality and safety of the bridge, but also can perform real-time monitoring and damage early warning on the bridge structure in the operation stage, evaluate the bearing capacity, estimate the remaining service life of the structure, provide important basis for decisions such as traffic organization management, operation maintenance, regular maintenance and reinforcement and the like of a bridge management department, and prevent the occurrence of bridge collapse accidents.
However, the existing health monitoring system is mostly applied to damage identification of a single sensor and a monitoring index, structural damage information acquired by a large number of sensors for health monitoring is not effectively utilized, all sensor information arranged on a bridge is not well utilized for data mining, real-time analysis and study and judgment are carried out, damage identification of the bridge structure cannot achieve high precision, and the requirement of actual engineering is still difficult to meet.
Disclosure of Invention
The following presents a simplified summary of the invention in order to provide a basic understanding of some aspects of the invention. It should be understood that this summary is not an exhaustive overview of the invention. It is not intended to determine the key or critical elements of the present invention, nor is it intended to limit the scope of the present invention. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later.
In view of this, the present invention provides a bridge health monitoring method, including the following steps:
s1, establishing a bridge finite element numerical simulation model, dividing the finite element numerical simulation model into a plurality of nodes and units, dividing the full-bridge girder into a plurality of sub-regions, and setting damage conditions with different damage degrees in each region;
s2, performing wavelet packet decomposition on the time-course data obtained by the simulation model, and converting the time-course data into frequency band energy data;
s3, constructing a relative energy ratio index according to the frequency band energy, and calculating an energy ratio deviation damage index and an energy ratio variance damage index;
s4 summarizing the damage indexes in the step S3 to generate multi-sensor feature vectors, training and learning the bridge structure by adopting an ensemble learning-based method, and further carrying out damage diagnosis on the sub-region of the structure.
Preferably, the step S1 of establishing a bridge finite element numerical simulation model specifically includes the following steps:
s11, establishing a finite element numerical simulation model according to engineering design data, and correcting the model according to data acquired by the data acquisition module;
s12, obtaining a traffic flow probability distribution model according to actual traffic flow data of the bridge, loading the traffic flow probability distribution model in a finite element numerical simulation model, and simulating load input of the actual bridge;
s13, extracting the structural response data of the corresponding position of the bridge from the finite element numerical simulation model according to the position of the actual sensor layout of the bridge, and constructing a bridge damage simulation database.
Preferably, in step S2, the specific method for performing wavelet packet decomposition on the time-course data obtained by the simulation model and converting the time-course data into the frequency-band energy data is as follows:
the time course original signal S (t) measured by the sensor is expressed as follows after wavelet packet decomposition:
Figure 480007DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 865989DEST_PATH_IMAGE002
the energy of the signal at the j position of the k layer obtained by wavelet packet decomposition is
Figure 724355DEST_PATH_IMAGE003
；
Figure 447460DEST_PATH_IMAGE004
Is the signal energy in the j frequency band;
the wavelet packet coefficients are:
Figure 730674DEST_PATH_IMAGE005
wherein the content of the first and second substances,
Figure 133230DEST_PATH_IMAGE006
is a wavelet packet with a scale index k, a position index j, and a frequency index i;
Figure 869105DEST_PATH_IMAGE007
wavelet packet coefficients of a k-th layer j position frequency index i obtained by wavelet packet decomposition;
the total energy of the signal is equal to the sum of the energy of the signals of all frequency bands
Figure 395901DEST_PATH_IMAGE008
And if so, the sum E of the wavelet packet node energies is as follows:
Figure 143408DEST_PATH_IMAGE009
。
preferably, in step S3, the specific method for constructing the relative energy ratio index according to the band energy is:
Figure 464668DEST_PATH_IMAGE010
wherein the content of the first and second substances,
Figure 422260DEST_PATH_IMAGE011
signal energy of the ith frequency band;
the energy ratio change of each frequency band before and after the damage is:
Figure 736436DEST_PATH_IMAGE012
wherein the content of the first and second substances,
Figure 853296DEST_PATH_IMAGE013
and
Figure 486403DEST_PATH_IMAGE014
before and after structural damage, respectivelyiEnergy ratio of each frequency band;
the ERVD damage index and the ERVV damage index are calculated according to the energy ratio deviation and the energy ratio difference as follows:
Figure 806657DEST_PATH_IMAGE015
Figure 409677DEST_PATH_IMAGE016
wherein the content of the first and second substances,
Figure 621522DEST_PATH_IMAGE017
variation of energy ratio for all characteristic frequency bands
Figure 425530DEST_PATH_IMAGE018
ERVD is the energy ratio deviation and ERVV is the energy ratio variance.
Preferably, in step S4, the specific method for collecting the damage indicators in step S3 to generate multi-sensor feature vectors, performing training and learning on the bridge structure by using an ensemble learning-based method, and further performing damage diagnosis on sub-regions of the structure is:
s41 summarizing the energy ratio deviation damage index and energy ratio deviation damage index data of each damage working condition to obtain a multi-sensor damage feature vector { a1, a2, a3, …, an } under the same damage working condition;
s42, representing the damage information of the bridge structure represented by each damage working condition by using a damage target vector { b }1,b2,b3,bi,…,bn}; s43, inputting the damage characteristic vector and the damage target vector into a damage simulation database as sample data;
s44, dividing the sample data into a training set and a testing set, and inputting the training set into an ensemble learning algorithm to establish an ensemble learning model; inputting the test set into a trained ensemble learning model for testing, and outputting a test result;
s45, inputting the real-time sensor data into the ensemble learning model, and outputting a damage diagnosis result;
s46 repeats the training test at intervals to iterate a new round of the model.
A bridge health monitoring system is used for realizing a bridge health monitoring method and comprises a data acquisition module, a data transmission module, a data storage module, a comprehensive early warning and evaluation module and a terminal module; the data acquisition module, the data transmission module, the data storage module, the comprehensive early warning and evaluation module and the terminal module are sequentially in communication connection;
the data acquisition module is used for acquiring bridge information;
the data transmission module is used for transmitting the data acquired by the data acquisition module to the data storage module;
the data storage module is used for storing the acquired bridge information;
the comprehensive early warning and evaluation module is used for carrying out comprehensive analysis and damage judgment on the stored bridge information through an integrated learning method to determine the health condition of the bridge;
and the terminal module is used for transmitting and displaying the result information of the comprehensive early warning and evaluation module, monitoring the health condition of the bridge in real time and transmitting the early warning information to the bridge manager terminal.
Preferably, the data acquisition module comprises an acceleration sensor, a pressure transmitter, a cable force sensor, a telescopic sensor, a temperature sensor, a resistance type strain gauge and a temperature and humidity instrument;
the acceleration sensor is used for monitoring the acceleration response of the bridge structure;
the pressure transmitter is used for monitoring the deflection change of the bridge structure;
the cable force sensor is used for monitoring the change of the stayed cable force;
the telescopic sensor is used for monitoring the change of the displacement of the support;
the temperature sensor is used for monitoring the change of the temperature of the structure;
the resistance strain gauge is used for monitoring the change of structural strain;
the humiture instrument is used for monitoring the change of environment humiture.
Preferably, the comprehensive early warning and evaluation module comprises an abnormal information processing module, a data analysis module, a comprehensive evaluation module and an early warning data storage module;
the abnormal information processing module is used for marking data with abnormal values, and immediately transmitting the data to the terminal to send out early warning information if the indexes exceed a safety threshold or the sensor is abnormal, so that background management personnel can analyze, study and judge and check safety in the field;
the data analysis module is used for carrying out data analysis on the summarized multi-sensor information;
the comprehensive evaluation module is used for evaluating the analysis result of the data analysis module to obtain the health condition information of the bridge;
the early warning data storage module is used for storing the result data of the abnormal information processing module, the data analysis module and the comprehensive evaluation module.
A computer comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the bridge health monitoring method when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out a bridge health monitoring method.
The invention has the following beneficial effects: according to the invention, wavelet packet decomposition is carried out by using time-course data of the full-bridge girder multi-sensor, and a damage characteristic vector is constructed by extracting normalized relative energy difference of damage characteristic frequency bands, so that information redundancy among frequency band characteristics is reduced, and the sensitivity of structural damage is improved. Compared with the traditional machine learning algorithm, the method has the advantages that the calculation cost is high, the algorithm is over-trained, the feature extraction blindness is realized, the multivariable multi-classification problem precision is low, and the like, the bridge is divided into sub-regions with smaller distance, the problem of whether a certain region of the bridge is damaged is converted into the problem of multi-classification of the damaged region category by means of the integrated learning method, and the problem of diagnosing the damaged region of the bridge even a large-span bridge is effectively solved. The invention effectively utilizes all sensor information of the bridge layout to carry out data mining, searches for the internal correlation between the structural damage and the dynamic response, particularly judges and evaluates the health condition of the bridge according to the mass monitoring data acquired by the optimal layout of the sensors, and can be well adapted to the optimal layout method of the sensors.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a flow diagram of a method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a system according to an embodiment of the present invention;
fig. 3 is a schematic diagram illustrating a division of a main girder sub-region of a cable-stayed bridge according to an embodiment of the present invention;
FIG. 4 is a schematic layout diagram of a bridge acceleration sensor according to an embodiment of the present invention;
fig. 5 is a schematic flow chart of a Bagging algorithm according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a sample training set flow in the Boosting algorithm according to an embodiment of the present invention;
FIG. 7 is a diagram illustrating updating an iterative model (classification error rate) to improve accuracy of output results according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating an embodiment of updating an iterative model (log loss) to improve accuracy of output results;
FIG. 9 is a diagram illustrating an accuracy result of the damage identification according to the embodiment of the present invention;
FIG. 10 is a diagram illustrating the accuracy of lesion identification, recall, and F1-Score results according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions and advantages of the embodiments of the present application more apparent, the following further detailed description of the exemplary embodiments of the present application with reference to the accompanying drawings makes it clear that the described embodiments are only a part of the embodiments of the present application, and are not exhaustive of all embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
Embodiment 1, referring to fig. 1, a bridge health monitoring system according to this embodiment is described, and includes a data acquisition module, a data transmission module, a data storage module, a comprehensive early warning and evaluation module, and a terminal module; the system comprises a data acquisition module, a data transmission module, a data storage module, a comprehensive early warning and evaluation module and a terminal module which are sequentially in communication connection;
the data acquisition module is used for acquiring bridge information; the data acquisition module comprises an acceleration sensor, a pressure transmitter, a cable force sensor, a telescopic sensor, a temperature sensor, a resistance type strain gauge and a temperature and humidity instrument;
specifically, the acceleration sensor is used for monitoring the acceleration response of the bridge structure; the pressure transmitter is used for monitoring the deflection change of the bridge structure; the cable force sensor is used for monitoring the change of the stayed cable force; the telescopic sensor is used for monitoring the change of the displacement of the support; the temperature sensor is used for monitoring the change of the temperature of the structure; the resistance strain gauge is used for monitoring the change of structural strain; the humiture instrument is used for monitoring the change of environment humiture.
Specifically, there are a plurality of acceleration sensor, pressure transmitter, cable force sensor, telescopic sensor, temperature sensor, resistance type strain gauge and humiture apparatus.
The data transmission module is used for transmitting the data acquired by the data acquisition module to the data storage module;
the data storage module is used for storing the acquired bridge information;
the comprehensive early warning and evaluation module is used for carrying out comprehensive analysis and damage judgment on the stored bridge information through an integrated learning method to determine the health condition of the bridge; the comprehensive early warning and evaluation module comprises an abnormal information processing module, a data analysis module, a comprehensive evaluation module and an early warning data storage module;
specifically, the abnormal information processing module is used for marking data with abnormal values, and if the index exceeds a safety threshold or the sensor is abnormal, the abnormal data is immediately transmitted to the terminal to send out early warning information for background management personnel to analyze, study and judge and check the safety in the field;
the data analysis module is used for carrying out data analysis on the summarized multi-sensor information;
the comprehensive evaluation module is used for evaluating the analysis result of the data analysis module to obtain the health condition information of the bridge;
the early warning data storage module is used for storing the result data of the abnormal information processing module, the data analysis module and the comprehensive evaluation module.
The terminal module is used for transmitting and displaying result information of the comprehensive early warning and evaluation module, can monitor the health condition of the bridge in real time, and sends early warning information to the bridge manager terminal.
Specifically, the warning information may be sent to the bridge manager terminal in the form of a short message or an email.
Embodiment 2, referring to fig. 2 to fig. 10, illustrates this embodiment, and the bridge health monitoring method of this embodiment includes the following steps:
s1, establishing a bridge finite element numerical simulation model, dividing the finite element numerical simulation model into a plurality of nodes and units, dividing the full-bridge girder into a plurality of sub-regions, and setting damage conditions with different damage degrees in each region; the method specifically comprises the following steps:
s11, establishing a finite element numerical simulation model according to the bridge engineering design data, and correcting the model according to the data collected by the data collection module to obtain an accurate finite element model; dividing a finite element numerical simulation model into a plurality of nodes and units, dividing a full-bridge main beam into a plurality of sub-regions, and setting damage working conditions with different damage degrees in each region;
specifically, the method for setting the damage conditions of different damage degrees is to simulate the damage by adopting a rigidity reduction method for a finite element numerical simulation model.
Specifically, the damage condition is that a certain area of the bridge is damaged to a certain extent; for example, the damage occurs to the extent of 10% in the region 1, and the damage occurs to the extent of 20% in the region 2.
S12, carrying out data statistics according to the actual traffic flow of the bridge to obtain a traffic flow probability distribution model, loading the traffic flow probability distribution model in a finite element numerical simulation model, and simulating the load input of the actual bridge; the vehicle load adopts the vehicle axle weight in JTG D60-2015 Highway bridge and culvert design general standard as the simulated vehicle load input.
S13, according to the position of the actual sensor arrangement of the bridge, extracting the structural response data (time-course data) of the corresponding position of the bridge from the finite element numerical simulation model, and constructing a bridge damage simulation database, wherein the time-course data are the acceleration, stress and strain data of different positions under different damage working conditions.
S2, performing wavelet packet decomposition on the time-course data obtained by the simulation model by using a dbN wavelet basis function, and converting the time-course data into frequency band energy data, wherein the frequency band energy data are signal energy distribution bands corresponding to different frequencies; the specific calculation formula is as follows:
the time course original signal s (t) measured by the sensor (in the present embodiment, the acceleration signal is taken as an example for explanation), which is expressed as:
Figure 482348DEST_PATH_IMAGE019
wherein S (t) is an acceleration signal,
Figure 30004DEST_PATH_IMAGE020
the energy of the signal at the j position of the k layer obtained by wavelet packet decomposition is
Figure 606610DEST_PATH_IMAGE021
；
Figure 706153DEST_PATH_IMAGE022
Is the signal energy in the j frequency band;
the wavelet packet coefficients are:
Figure 125633DEST_PATH_IMAGE023
wherein the content of the first and second substances,
Figure 319723DEST_PATH_IMAGE024
is a wavelet packet with a scale index k, a position index j, and a frequency index i;
Figure 265682DEST_PATH_IMAGE007
wavelet packet coefficients of a k-th layer j position frequency index i obtained by wavelet packet decomposition;
the total energy of the frequency band signal is equal to the sum of the energy of the signals of the respective frequency bands,
Figure 677072DEST_PATH_IMAGE025
and if so, the sum E of the wavelet packet node energies is as follows:
Figure 193635DEST_PATH_IMAGE026
specifically, the purpose of performing wavelet packet decomposition on the time-course data obtained by the simulation model by using dbN wavelet basis functions is to obtain an index with higher damage sensitivity to the bridge structure.
S3, constructing a relative energy ratio index according to the frequency band energy, and calculating an energy ratio deviation damage index and an energy ratio variance damage index; the specific calculation method of the relative energy ratio index is as follows:
Figure 942148DEST_PATH_IMAGE027
wherein the content of the first and second substances,
Figure 883559DEST_PATH_IMAGE028
for the signal energy of the ith frequency band,
Figure 577102DEST_PATH_IMAGE029
is the total energy of the frequency band signal;
the energy ratio of each frequency band signal before and after the damage is changed as follows:
Figure 830229DEST_PATH_IMAGE030
wherein the content of the first and second substances,
Figure 133165DEST_PATH_IMAGE031
and
Figure 663504DEST_PATH_IMAGE032
before and after structural damage, respectivelyiEnergy ratio of each frequency band;
the energy ratio deviation (ERVD) and energy ratio variance (ERVV) impairment indicators are constructed as follows:
Figure 275751DEST_PATH_IMAGE033
Figure 265441DEST_PATH_IMAGE034
wherein the content of the first and second substances,
Figure 496702DEST_PATH_IMAGE035
variation of energy ratio for all characteristic frequency bands
Figure 6181DEST_PATH_IMAGE036
Average value of (a).
S4, summarizing the energy ratio deviation damage index and energy ratio difference damage index data, and calculating the corresponding energy ratio deviation damage index and energy ratio variance damage index data from the data collected by each sensor, so as to generate multi-sensor feature vectors according to the damage index data calculated by different sensors; training and learning the bridge structure by adopting an ensemble learning-based method, and further performing damage diagnosis on a sub-region of the structure; the method specifically comprises the following steps:
s41 summarizing the energy ratio deviation damage index and energy ratio deviation damage index data of each damage working condition to obtain a multi-sensor damage feature vector { a1, a2, a3, …, an } under the same damage working condition;
s42, generating damage target vectors { b1, b2, b3, bi, …, bn } from the bridge structure damage information represented by each damage working condition;
specifically, whether a certain sub-area of the bridge is damaged or not is used as the purpose of identifying the damaged area.
Specifically, bi is only 0 or 1, bi is 1, which means that the ith area is damaged, and bi is 0, which means that the ith area is not damaged. For example: the bridge monitoring sensors are acceleration sensors, the number of the acceleration sensors is 30, the fusion indexes under the same damage working condition comprise 2 Í 30, the bridge is divided into 30 areas, and if the first area is judged to be a damaged area, a target output 1 Í 30 vector {1,0,0, …,0} represents that the position area 1 is damaged.
S43, inputting the damage characteristic vector and the damage target vector into a damage simulation database as sample data;
s44, dividing the sample data into a training set and a testing set, and inputting the training set into an ensemble learning algorithm to establish an ensemble learning model; inputting the test set into a trained ensemble learning model for testing, and outputting a test result;
specifically, the sample data training set and the test set may be divided in a custom ratio, for example, the training set: the test set was 7: 3.
Specifically, the damage degree can be customized according to the damage condition, for example, the damage condition data of 10%, 15% and 20% of the damage degree are training sets, and the damage condition data of other damage degrees are test sets.
Specifically, the ensemble learning algorithm comprises a Bagging algorithm and a Boosting algorithm;
the Bagging algorithm is as follows:
randomly extracting n training samples from an original sample set, carrying out T times of replaced random sampling in total to obtain T different sampling sets, then independently training the T training sets to obtain T weak classifiers, and finally obtaining the final strong classifier by an aggregation strategy (such as a voting strategy) for the T weak classifiers.
The Boosting algorithm is:
randomly extracting n training samples from an original sample set, endowing a sample training set with wrong classification with higher weight aiming at the result of each iteration, carrying out next training, training a weak classifier 1 by using the initial weight from the training set, updating the weight of the training sample according to the learning error rate performance of weak learning, leading the weight of the training sample point with high learning error rate of the weak classifier 1 to be higher, then training a weak learner 2 based on the training set after the weight is adjusted, repeating the training of T weak classifiers in the way, and obtaining the final strong learner through an integration strategy.
Specifically, the weak classifier includes: decision trees, support vector machines, naive Bayes, Logistic regression, artificial neural networks and other machine learning algorithms.
Specifically, in order to obtain an ensemble learning model with high accuracy and generalization capability, the XGBoost algorithm in the ensemble learning model is, for example, optimized with respect to the hyper-parameters, specifically, the Log loss function and the classification error rate of the sample test set are compared to optimize the parameters:
referring to fig. 8, the Log loss function is illustrated:
Figure 540062DEST_PATH_IMAGE038
wherein Y is an output variable, X is an input variable, L is a Log loss function, N is an input sample size, M is a possible number of classes,
Figure 643147DEST_PATH_IMAGE039
is a binary index representing whether the class j is an input instance xiThe true category of (a) of (b),
Figure 537154DEST_PATH_IMAGE040
predicting input instance x for a model or classifieriProbability of belonging to category j.
Referring to fig. 7, the classification error rate is illustrated:
Figure 418915DEST_PATH_IMAGE041
whereiny i The predicted class of the output of the classifier,Y i are actual categories.
S45, inputting the real-time sensor data into the ensemble learning model, and outputting a damage diagnosis result;
s46 repeats the training test at intervals to iterate a new round of the model.
Specifically, monitoring data acquired by an acquisition module in real time are added into a damage simulation database at specific intervals for training and testing, a new model is iterated, real-time damage diagnosis and early warning are carried out on the bridge through the new model and real-time sensor data, in order to avoid that the monitoring and early warning real-time performance is influenced by overlong training time caused by overlarge capacity of damage simulation data, the damage simulation database is required to be periodically cleaned from historical data, meanwhile, the timeliness of the data is ensured, the health condition of the bridge structure is reflected by the latest monitoring data, and the iterative training model is continuously updated.
Example 3, referring to fig. 2 to 10, this embodiment will be described, and the concrete implementation steps described in example 2 of the present invention will be described for a cable-stayed bridge:
the method comprises the steps of firstly, modeling the bridge by means of finite element software based on engineering design data of the bridge, considering acceleration response information of a main beam under the action of moving load as damage characteristic information of structural damage identification, and considering reduction of structural rigidity to simulate damage, wherein the full-bridge main beam is divided into 30 areas as shown in figure 3, and the arrangement position of a main beam sensor is as shown in figure 4.
And step two, setting the damage degree of 10-70% on selected partial positions of 30 sub-regions of the full-bridge girder, wherein the damage working conditions comprise 210, and increasing 5 different random interferences of 10%, 15%, 20% and 25% on each working condition to obtain 1260 groups of damage sample data sets. 30 groups of different acceleration signals can be measured by the acceleration sensor arranged on the main beam under each damage working condition, the normalized relative energy difference of the characteristic frequency bands of the signals before and after damage is extracted as a damage index, and the obtained 60 characteristic variables are constructed into a multi-sensor damage characteristic vector
Figure 107386DEST_PATH_IMAGE042
And whether the damage occurs to the sub-region is represented by a vector {1,0,0, …,0} of 1 Í 30, and a damage simulation database is constructed.
And step three, training and testing the damage simulation database by adopting an XGboost algorithm in the integrated learning method, wherein the algorithm flow is shown in a figure 6. The XGboost objective function is:
Figure 697767DEST_PATH_IMAGE043
in the formula
Figure 146197DEST_PATH_IMAGE044
Is a loss function describing the residual between the true and predicted values,
Figure 630268DEST_PATH_IMAGE045
model objects for regularization terms to reduce model complexity and prevent overfitting
Figure 630585DEST_PATH_IMAGE046
Is to minimize the objective function. Selecting and optimizing parameters by comparing Log loss and classification error rate of a sample test set:
Figure 82164DEST_PATH_IMAGE048
wherein Y is an output variable, X is an input variable, and L is a Log loss function. N is the input sample size, M is the number of possible classes,
Figure 583552DEST_PATH_IMAGE049
is a binary indicator that indicates whether the class j is the true class of the input instance xi.
Figure 407283DEST_PATH_IMAGE050
The probability that the input instance xi belongs to the class j is predicted for the model or classifier. Refer to the drawings7-FIG. 8, the algorithm min _ child _ weight parameter optimization iteration graph.
And step four, inputting the sample test data set into a model trained by the integrated learning XGboost algorithm, and carrying out damage diagnosis on the damaged area of the bridge.
Specifically, for the model in the bridge health monitoring method provided by the invention, the Accuracy of the model output is evaluated by using four indexes of Precision (Precision), Recall (Recall), F1-score and Accuracy (Accuracy), and refer to fig. 9-10.
The working principle of the invention is as follows: dynamic response data generated by a vehicle to a bridge structure in the actual operation process of the bridge is monitored and collected in a data storage module in real time through a distributed sensor, in a comprehensive early warning and evaluation module, corresponding index data is obtained according to the monitoring data and is added into a damage simulation database to judge a damage area, if the judgment is no damage, the monitoring data is directly added into a damage simulation data set to carry out iterative training at specific intervals, if damage is predicted to occur, the health condition of the bridge is monitored in real time, early warning information is sent to a bridge management personnel terminal in a short message or mail mode, and a management department can take corresponding treatment measures in time to avoid safety accidents.
The invention can carry out real-time monitoring and damage prediction evaluation on bridge structure data under the actual vehicle operation state, carry out real-time monitoring and damage early warning on abnormal bridge operation conditions, evaluate the bearing capacity and estimate the remaining service life of the structure, thereby providing very important basis for decision making of traffic organization management, operation maintenance, regular maintenance reinforcement and the like of a bridge management department and preventing bridge collapse accidents from happening. The method can also be extended to the bridge monitoring data with similar local environmental and structural characteristics and combined with the structural regional bridge health monitoring big data to perform damage early warning and identification.
The computer device of the present invention may be a device including a processor, a memory, and the like, for example, a single chip microcomputer including a central processing unit and the like. And the processor is used for implementing the steps of the recommendation method capable of modifying the relationship-driven recommendation data based on the CREO software when executing the computer program stored in the memory.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Computer-readable storage medium embodiments
The computer readable storage medium of the present invention may be any form of storage medium that can be read by a processor of a computer device, including but not limited to non-volatile memory, ferroelectric memory, etc., and the computer readable storage medium has stored thereon a computer program that, when the computer program stored in the memory is read and executed by the processor of the computer device, can implement the above-mentioned steps of the CREO-based software that can modify the modeling method of the relationship-driven modeling data.
The computer program comprises computer program code which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this description, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as described herein. Furthermore, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the appended claims. The present invention has been disclosed in an illustrative rather than a restrictive sense, and the scope of the present invention is defined by the appended claims.

Claims (10)

1. A bridge health monitoring method is characterized by comprising the following steps:
s1, establishing a bridge finite element numerical simulation model, dividing the finite element numerical simulation model into a plurality of nodes and units, dividing the full-bridge girder into a plurality of sub-regions, and setting damage conditions with different damage degrees in each region;
s2, performing wavelet packet decomposition on the time-course data obtained by the simulation model, and converting the time-course data into frequency band energy data;
s3, constructing a relative energy ratio index according to the frequency band energy, and calculating an energy ratio deviation damage index and an energy ratio variance damage index;
s4 summarizing the damage indexes in the step S3 to generate multi-sensor feature vectors, training and learning the bridge structure by adopting an ensemble learning-based method, and further carrying out damage diagnosis on the sub-region of the structure.
2. The method according to claim 1, wherein the step S1 of establishing the bridge finite element numerical simulation model specifically comprises the steps of:
s11, establishing a finite element numerical simulation model according to engineering design data, and correcting the model according to data acquired by the data acquisition module;
s12, obtaining a traffic flow probability distribution model according to actual traffic flow data of the bridge, loading the traffic flow probability distribution model in a finite element numerical simulation model, and simulating load input of the actual bridge;
s13, extracting the structural response data of the corresponding position of the bridge from the finite element numerical simulation model according to the position of the actual sensor layout of the bridge, and constructing a bridge damage simulation database.
3. The method according to claim 2, wherein the step S2 is to perform wavelet packet decomposition on the time-course data obtained from the simulation model, and convert the time-course data into the frequency-band energy data by:
the time course original signal S (t) measured by the sensor is expressed as follows after wavelet packet decomposition:
Figure 906491DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 620369DEST_PATH_IMAGE002
the energy of the signal at the j position of the k layer obtained by wavelet packet decomposition is
Figure 462423DEST_PATH_IMAGE003
；
Figure 716686DEST_PATH_IMAGE004
Is the signal energy in the j frequency band;
the wavelet packet coefficients are:
Figure 327796DEST_PATH_IMAGE005
wherein the content of the first and second substances,
Figure 884680DEST_PATH_IMAGE006
is a wavelet packet with a scale index k, a position index j, and a frequency index i;
Figure 948451DEST_PATH_IMAGE007
wavelet packet coefficients of a k-th layer j position frequency index i obtained by wavelet packet decomposition;
the total energy of the signal is equal to the sum of the energy of the signals of all frequency bands
Figure 6405DEST_PATH_IMAGE008
And if so, the sum E of the wavelet packet node energies is as follows:
Figure 472022DEST_PATH_IMAGE009
。
4. the method according to claim 3, wherein the specific method for constructing the relative energy ratio index according to the band energy in step S3 is as follows:
Figure 199806DEST_PATH_IMAGE010
wherein the content of the first and second substances,
Figure 750873DEST_PATH_IMAGE011
signal energy of the ith frequency band;
the energy ratio change of each frequency band before and after the damage is:
Figure 346940DEST_PATH_IMAGE012
wherein the content of the first and second substances,
Figure 667063DEST_PATH_IMAGE013
and
Figure 831328DEST_PATH_IMAGE014
before and after structural damage, respectivelyiEnergy ratio of each frequency band;
the ERVD damage index and the ERVV damage index are calculated according to the energy ratio deviation and the energy ratio difference as follows:
Figure 666429DEST_PATH_IMAGE015
Figure 738290DEST_PATH_IMAGE016
wherein the content of the first and second substances,
Figure 850602DEST_PATH_IMAGE017
variation of energy ratio for all characteristic frequency bands
Figure 44823DEST_PATH_IMAGE018
ERVD is the energy ratio deviation and ERVV is the energy ratio variance.
5. The method of claim 4, wherein in step S4, the damage indicators in step S3 are collected to generate multi-sensor feature vectors, and the specific method for performing training learning on the bridge structure by using the ensemble learning-based method and further performing damage diagnosis on the sub-regions of the structure is:
s41 summarizing the energy ratio deviation damage index and energy ratio deviation damage index data of each damage working condition to obtain a multi-sensor damage feature vector { a1, a2, a3, …, an } under the same damage working condition;
s42, representing the damage information of the bridge structure represented by each damage working condition by using a damage target vector { b }1,b2,b3,bi,…,bn}；
S43, inputting the damage characteristic vector and the damage target vector into a damage simulation database as sample data;
s44, dividing the sample data into a training set and a testing set, and inputting the training set into an ensemble learning algorithm to establish an ensemble learning model; inputting the test set into a trained ensemble learning model for testing, and outputting a test result;
s45, inputting the real-time sensor data into the ensemble learning model, and outputting a damage diagnosis result;
s46 repeats the training test at intervals to iterate a new round of the model.
6. A bridge health monitoring system is characterized in that the bridge health monitoring system is used for realizing the bridge health monitoring method of any one of claims 1 to 5, and comprises a data acquisition module, a data transmission module, a data storage module, a comprehensive early warning and evaluation module and a terminal module; the data acquisition module, the data transmission module, the data storage module, the comprehensive early warning and evaluation module and the terminal module are sequentially in communication connection;
the data acquisition module is used for acquiring bridge information;
the data transmission module is used for transmitting the data acquired by the data acquisition module to the data storage module;
the data storage module is used for storing the acquired bridge information;
the comprehensive early warning and evaluation module is used for carrying out comprehensive analysis and damage judgment on the stored bridge information through an integrated learning method to determine the health condition of the bridge;
and the terminal module is used for transmitting and displaying the result information of the comprehensive early warning and evaluation module, monitoring the health condition of the bridge in real time and transmitting the early warning information to the bridge manager terminal.
7. The monitoring system of claim 6, wherein the data acquisition module comprises an acceleration sensor, a pressure transmitter, a cable force sensor, a telescopic sensor, a temperature sensor, a resistive strain gauge and a temperature and humidity meter;
the acceleration sensor is used for monitoring the acceleration response of the bridge structure;
the pressure transmitter is used for monitoring the deflection change of the bridge structure;
the cable force sensor is used for monitoring the change of the stayed cable force;
the telescopic sensor is used for monitoring the change of the displacement of the support;
the temperature sensor is used for monitoring the change of the temperature of the structure;
the resistance strain gauge is used for monitoring the change of structural strain;
the humiture instrument is used for monitoring the change of environment humiture.
8. The monitoring system of claim 7, wherein the comprehensive early warning and assessment module comprises an abnormal information processing module, a data analysis module, a comprehensive assessment module and an early warning data storage module;
the abnormal information processing module is used for marking data with abnormal values, and immediately transmitting the data to the terminal to send out early warning information if the indexes exceed a safety threshold or the sensor is abnormal, so that background management personnel can analyze, study and judge and check safety in the field;
the data analysis module is used for carrying out data analysis on the summarized multi-sensor information;
the comprehensive evaluation module is used for evaluating the analysis result of the data analysis module to obtain the health condition information of the bridge;
the early warning data storage module is used for storing the result data of the abnormal information processing module, the data analysis module and the comprehensive evaluation module.
9. A computer comprising a memory storing a computer program and a processor, the processor implementing the steps of a bridge health monitoring method according to any one of claims 1 to 5 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a bridge health monitoring method according to any one of claims 1 to 5.
CN202111106958.5A 2021-09-22 2021-09-22 Bridge health monitoring method, system, computer and storage medium Pending CN113553749A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111106958.5A CN113553749A (en) 2021-09-22 2021-09-22 Bridge health monitoring method, system, computer and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111106958.5A CN113553749A (en) 2021-09-22 2021-09-22 Bridge health monitoring method, system, computer and storage medium

Publications (1)

Publication Number Publication Date
CN113553749A true CN113553749A (en) 2021-10-26

Family

ID=78106515

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111106958.5A Pending CN113553749A (en) 2021-09-22 2021-09-22 Bridge health monitoring method, system, computer and storage medium

Country Status (1)

Country Link
CN (1) CN113553749A (en)

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114252283A (en) * 2021-11-26 2022-03-29 宜春市公路管理局 System and method for monitoring whole life of small and medium-span bridge
CN114429032A (en) * 2021-12-23 2022-05-03 贵州大学 Bridge health online intelligent monitoring system
CN114563150A (en) * 2021-12-23 2022-05-31 贵州大学 Bridge health online detection module generation method, detection method, tool box and device
CN114608432A (en) * 2022-05-16 2022-06-10 深圳市城市交通规划设计研究中心股份有限公司 Bridge deformation monitoring device and method
CN115032270A (en) * 2022-06-01 2022-09-09 北京科技大学 Method and device for quantitatively identifying damage state of building curtain wall based on machine learning algorithm
CN115440010A (en) * 2022-08-12 2022-12-06 无锡小淞交通科技发展有限公司 Bridge remote monitoring early warning device based on thing networking
CN115685786A (en) * 2023-01-01 2023-02-03 北京云庐科技有限公司 Bridge online simulation and evaluation system based on temperature monitoring load
CN115993097A (en) * 2023-03-23 2023-04-21 长安大学 Monitoring and early warning method and system for cable-stayed bridge stay cable broken wire
CN116090838A (en) * 2023-04-07 2023-05-09 协筑科技(厦门)有限公司 Automatic building site system of patrolling and examining based on camera
CN116123465A (en) * 2023-04-11 2023-05-16 东莞先知大数据有限公司 Pipeline leakage early warning method, electronic equipment and storage medium
CN116150680A (en) * 2023-04-18 2023-05-23 中铁第四勘察设计院集团有限公司 Underground transportation junction damage state identification method and system based on big data
CN116839659A (en) * 2023-06-21 2023-10-03 中交三航局第三工程有限公司 Method and system for identifying damage of girder of bridge girder erection machine
CN116881819A (en) * 2023-09-07 2023-10-13 成都理工大学 Stay cable working state monitoring method based on isolated forest
CN116975989A (en) * 2023-09-25 2023-10-31 深圳市城市交通规划设计研究中心股份有限公司 Method for evaluating bridge running state based on finite element guided deep learning agent model
CN117216701A (en) * 2023-09-13 2023-12-12 广州桐富科技发展有限公司 Intelligent bridge monitoring and early warning method and system
CN117592341A (en) * 2024-01-18 2024-02-23 苏州大学 Finite element prediction method and system for T-shaped bridge health state

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106556498A (en) * 2016-10-31 2017-04-05 石家庄铁道大学 Damage Identification Methods for Bridge Structures and system
CN110633855A (en) * 2019-09-17 2019-12-31 江南大学 Bridge health state detection and management decision making system and method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106556498A (en) * 2016-10-31 2017-04-05 石家庄铁道大学 Damage Identification Methods for Bridge Structures and system
CN110633855A (en) * 2019-09-17 2019-12-31 江南大学 Bridge health state detection and management decision making system and method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
黄笑犬: "基于传感器优化布置与集成学习的 大跨桥梁损伤识别研究", 《万方学位论文》 *

Cited By (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114252283A (en) * 2021-11-26 2022-03-29 宜春市公路管理局 System and method for monitoring whole life of small and medium-span bridge
CN114429032A (en) * 2021-12-23 2022-05-03 贵州大学 Bridge health online intelligent monitoring system
CN114563150A (en) * 2021-12-23 2022-05-31 贵州大学 Bridge health online detection module generation method, detection method, tool box and device
CN114608432A (en) * 2022-05-16 2022-06-10 深圳市城市交通规划设计研究中心股份有限公司 Bridge deformation monitoring device and method
CN114608432B (en) * 2022-05-16 2022-08-23 深圳市城市交通规划设计研究中心股份有限公司 Bridge deformation monitoring device and method
CN115032270A (en) * 2022-06-01 2022-09-09 北京科技大学 Method and device for quantitatively identifying damage state of building curtain wall based on machine learning algorithm
CN115440010A (en) * 2022-08-12 2022-12-06 无锡小淞交通科技发展有限公司 Bridge remote monitoring early warning device based on thing networking
CN115685786A (en) * 2023-01-01 2023-02-03 北京云庐科技有限公司 Bridge online simulation and evaluation system based on temperature monitoring load
CN115993097A (en) * 2023-03-23 2023-04-21 长安大学 Monitoring and early warning method and system for cable-stayed bridge stay cable broken wire
CN116090838A (en) * 2023-04-07 2023-05-09 协筑科技(厦门)有限公司 Automatic building site system of patrolling and examining based on camera
CN116123465A (en) * 2023-04-11 2023-05-16 东莞先知大数据有限公司 Pipeline leakage early warning method, electronic equipment and storage medium
CN116150680A (en) * 2023-04-18 2023-05-23 中铁第四勘察设计院集团有限公司 Underground transportation junction damage state identification method and system based on big data
CN116150680B (en) * 2023-04-18 2023-07-21 中铁第四勘察设计院集团有限公司 Underground transportation junction damage state identification method and system based on big data
CN116839659A (en) * 2023-06-21 2023-10-03 中交三航局第三工程有限公司 Method and system for identifying damage of girder of bridge girder erection machine
CN116881819A (en) * 2023-09-07 2023-10-13 成都理工大学 Stay cable working state monitoring method based on isolated forest
CN116881819B (en) * 2023-09-07 2023-11-14 成都理工大学 Stay cable working state monitoring method based on isolated forest
CN117216701A (en) * 2023-09-13 2023-12-12 广州桐富科技发展有限公司 Intelligent bridge monitoring and early warning method and system
CN117216701B (en) * 2023-09-13 2024-03-15 华夏安信物联网技术有限公司 Intelligent bridge monitoring and early warning method and system
CN116975989A (en) * 2023-09-25 2023-10-31 深圳市城市交通规划设计研究中心股份有限公司 Method for evaluating bridge running state based on finite element guided deep learning agent model
CN116975989B (en) * 2023-09-25 2024-02-27 深圳市城市交通规划设计研究中心股份有限公司 Method for evaluating bridge running state based on finite element guided deep learning agent model
CN117592341A (en) * 2024-01-18 2024-02-23 苏州大学 Finite element prediction method and system for T-shaped bridge health state
CN117592341B (en) * 2024-01-18 2024-03-29 苏州大学 Finite element prediction method and system for T-shaped bridge health state

Similar Documents

Publication Publication Date Title
CN113553749A (en) Bridge health monitoring method, system, computer and storage medium
WO2023060672A1 (en) Bridge scouring power identification method based on health monitoring data
Han et al. Intelligent decision model of road maintenance based on improved weight random forest algorithm
CN114357594B (en) Bridge abnormity monitoring method, system, equipment and storage medium based on SCA-GRU
Zhao et al. A fuzzy system for concrete bridge damage diagnosis
CN110796859A (en) Real-time traffic state identification and accident risk early warning method based on traffic flow
CN106157132A (en) Credit risk monitoring system and method
CN112200237B (en) Time sequence monitoring data abnormality diagnosis method for structural health monitoring system
Parisi et al. Automated location of steel truss bridge damage using machine learning and raw strain sensor data
CN110910641A (en) Neural network-based real-time risk assessment method for highway traffic operation
CN114563150B (en) Bridge health online detection module generation method, detection method, tool box and device
Shtayat et al. An overview of pavement degradation prediction models
Son et al. Deep learning-based anomaly detection to classify inaccurate data and damaged condition of a cable-stayed bridge
CN112488142A (en) Radar fault prediction method and device and storage medium
Bandara et al. Structural health assessment of timber utility poles using stress wave propagation and artificial neural network techniques
KR102185737B1 (en) Estimating system of road condition and estimating method thereof
Carrillo et al. Design of efficient deep learning models for determining road surface condition from roadside camera images and weather data
CN117151513A (en) Method, device, equipment and storage medium for evaluating traffic safety
CN114723042B (en) Bridge damage position and damage degree identification method based on optimized neural network
CN113450565B (en) Method and system for reducing noise of asphalt pavement
CN116757303A (en) Sedimentation prediction method, system and equipment
Raza et al. Application of extreme learning machine algorithm for drought forecasting
Giglioni et al. A domain adaptation approach to damage classification with an application to bridge monitoring
CN113570111A (en) On-chip monitoring method for bridge health state based on lightweight network
Kumar et al. Multiple layer radial basis neural network with remora regression tree optimum feature extraction for structural health monitoring

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20211026