CN114429032A - Bridge health online intelligent monitoring system - Google Patents
Bridge health online intelligent monitoring system Download PDFInfo
- Publication number
- CN114429032A CN114429032A CN202111592341.9A CN202111592341A CN114429032A CN 114429032 A CN114429032 A CN 114429032A CN 202111592341 A CN202111592341 A CN 202111592341A CN 114429032 A CN114429032 A CN 114429032A
- Authority
- CN
- China
- Prior art keywords
- bridge
- data
- damage
- module
- analyzed
- 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
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 45
- 230000036541 health Effects 0.000 title claims abstract description 44
- 238000013145 classification model Methods 0.000 claims abstract description 64
- 230000001133 acceleration Effects 0.000 claims abstract description 52
- 238000004458 analytical method Methods 0.000 claims abstract description 48
- 238000007405 data analysis Methods 0.000 claims abstract description 29
- 238000012800 visualization Methods 0.000 claims abstract description 20
- 230000000007 visual effect Effects 0.000 claims abstract description 8
- 230000002159 abnormal effect Effects 0.000 claims description 47
- 238000000034 method Methods 0.000 claims description 37
- 238000012549 training Methods 0.000 claims description 33
- 239000000463 material Substances 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 21
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000013500 data storage Methods 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000004891 communication Methods 0.000 claims description 8
- 238000006073 displacement reaction Methods 0.000 claims description 8
- 238000003066 decision tree Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 abstract description 20
- 238000001514 detection method Methods 0.000 abstract description 13
- 230000006870 function Effects 0.000 description 15
- 238000013079 data visualisation Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 9
- 230000005540 biological transmission Effects 0.000 description 8
- 238000013523 data management Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 108010076504 Protein Sorting Signals Proteins 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013526 transfer learning Methods 0.000 description 2
- 241000274582 Pycnanthus angolensis Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005511 kinetic theory Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 239000011087 paperboard Substances 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24564—Applying rules; Deductive queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- 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
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Operations Research (AREA)
- Algebra (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides an online intelligent monitoring system for bridge health, which is provided with a data analysis module for loading a bridge damage identification classification model obtained based on machine learning, performs anomaly detection on data acquired by an acceleration sensor, and can automatically monitor the health state of a bridge in real time. And a damage visualization module is arranged, so that under the condition that the abnormity of the bridge to be analyzed is detected, the bridge to be analyzed is subjected to damage analysis based on near-field dynamics, visual display is carried out, and the health state of the bridge to be analyzed is presented more intuitively.
Description
Technical Field
The invention relates to the technical field of bridge damage assessment, in particular to an online intelligent bridge health monitoring system.
Background
By 12 months in 2020, the total number of Chinese bridges exceeds one million, wherein the total number of highway bridges reaches 87.83 ten thousand, 5716 grand bridges and 108344 grand bridges. With the rapid increase of the number of bridges in China and the gradual increase of the bridge ages of parts of bridges, the health and safety problems of bridges are also attracting more and more attention. If the bridge safety problem cannot be reflected and processed quickly in time, a great potential safety hazard p exists, and even great life and economic property losses are caused.
The monitoring technology is the key for guaranteeing the safe operation of the bridge, the damage condition and the dynamic characteristic of the bridge structure are obtained through real-time monitoring, so that the actual bearing condition of the bridge is analyzed, and a basis and a guide are provided for maintenance, repair and management decisions of the bridge through monitoring and evaluation of the condition of the bridge structure. The existing bridge monitoring system captures massive information data such as strain, deformation and the like, obtains the data from the data to be processed, then analyzes and evaluates the data, converts signals into information through a whole set of analysis process, and then analyzes the information to represent the service performance condition of the bridge. The common health monitoring system mainly comprises several parts of data acquisition, data analysis and data storage, wherein the most important and technically difficult part is the data analysis part. In recent years, with the rapid development of technologies such as sensing, communication and storage, the data acquisition capability in bridge monitoring is greatly improved, however, after the data are collected, the data are not well utilized and developed, the data analysis capability of a health system of a bridge at the present stage is not matched with the data collection capability, in the aspect of data analysis, the problems that the data processing speed is slow and the accuracy is to be improved exist, the data still stay in the stage of simply classifying data through machine learning, in the aspect of structural safety evaluation, the capability of full-bridge state evaluation through analysis of bridge components is slowly developed, and the early warning capability is to be improved.
Disclosure of Invention
The embodiment of the invention provides an online intelligent monitoring system for bridge health, which is used for eliminating or improving one or more defects in the prior art and solving the problems of insufficient analysis capability of bridge monitoring data and complex data analysis operation in the prior art.
The technical scheme of the invention is as follows:
the invention provides an online intelligent monitoring system for bridge health, which comprises:
the data acquisition module comprises a plurality of sensors, the sensors at least comprise acceleration sensors, and the sensors are arranged on the bridge to be analyzed according to a preset layout;
the data analysis module inputs the acceleration information acquired by the acceleration sensor into a bridge damage identification classification model to identify abnormal data in the acceleration information and judge whether the bridge to be analyzed is damaged; the bridge damage identification classification model is selected and called from preset toolbox plate blocks;
the damage visualization module is used for acquiring the position information and the acceleration information acquired by the acceleration sensor under the condition that the data analysis module judges that the bridge to be analyzed has damage, constructing an integral or local model for the bridge to be analyzed, establishing a near-field dynamic motion equation and solving the damage of the bridge to be analyzed at each time step to obtain a bridge damage result at each time step; combining the data information acquired by each sensor and the bridge damage result to the model of the bridge to be analyzed for visual display;
and the data storage module stores data information acquired by each sensor, abnormal data identified by the data analysis module and bridge damage results of each time step solved by the damage visualization module by adopting a relational database.
In some embodiments, the toolbox plate integrates a plurality of bridge damage identification classification models, and selects one or more bridge damage identification classification models according to a preset rule to identify abnormal data in the acceleration information;
each bridge damage identification classification model is obtained by pre-training an initial classification model and comprises the following steps:
acquiring a sample training set, wherein the sample training set comprises a plurality of sample data, each sample data comprises sample acceleration signals acquired by a first set number of sensors uniformly distributed on a sample bridge and bridge body state information, the bridge body state information comprises two types of structure damage and structure damage, and the bridge body state information corresponding to each sample data is used as a label;
taking the sample acceleration signal as input and the label as output, and adopting the training sample set to respectively train a plurality of initial classification models to obtain a plurality of bridge damage identification classification models; the initial classification model at least comprises a support vector machine, a decision tree, a fully connected neural network, a long-short term memory neural network or a self-organizing mapping algorithm network.
In some embodiments, before the training of the plurality of initial classification models with the training sample set, the method further includes:
and denoising the sample acceleration signal by adopting Gaussian filtering, and completing missing data in the sample acceleration signal by adopting a K-neighbor algorithm.
In some embodiments, selecting one or more bridge damage recognition classification models according to a preset rule to identify abnormal data in the acceleration information includes:
calculating evaluation indexes of each bridge damage identification classification model, wherein the evaluation indexes comprise accuracy, precision, recall and F1 scores;
after normalization processing is carried out on the accuracy, the precision ratio, the recall ratio and the F1 score of each bridge damage identification classification model, weighting and summing are carried out to obtain a comprehensive score of each bridge damage identification classification model, and the comprehensive scores are arranged from high to low;
selecting the bridge damage identification classification model with the highest comprehensive score to identify abnormal data in the acceleration information; or selecting a second set number of bridge damage identification classification models with higher comprehensive scores to identify the bridge damages, wherein the second set number is an odd number, and taking the identification results with larger numbers as final abnormal data identification results.
In some embodiments, the toolbox block is further provided with an improved prediction model, the improved prediction model maps the output of each bridge damage identification classification model to the final abnormal data and has a result, and the improved prediction model is obtained by adopting a fully-connected neural network training.
In some embodiments, the damage visualization module constructs a whole or local model for the bridge to be analyzed, establishes a near-field dynamic motion equation, and solves the damage of the bridge to be analyzed at each time step, including:
the method comprises the steps of obtaining material parameters including elastic modulus, density, Poisson ratio and tensile critical elongation, load parameters including load force and load speed, time step, ratio of radius of a near field region to side length of a point unit, and solving parameters of limiting time step, limiting displacement or crack arrest condition of limiting solving end conditions, setting side length of the point unit, a fixed constraint region, a working condition loading region and a material distribution region, constructing a near field dynamic motion equation of the bridge to be analyzed, solving damage of the bridge to be analyzed, and dynamically storing solving results of the time steps as bridge damage analysis results.
In some embodiments, the damage visualization module establishes a near-field dynamics motion equation based on a shared memory thread-level parallel method of OpenMP and solves the damage of the bridge to be analyzed at each time step.
In some embodiments, the data acquisition module comprises:
the system comprises a plurality of sensors, a data processing system and a data processing system, wherein the sensors comprise a temperature sensor, a pressure sensor and an acceleration sensor, and are arranged on a bridge to be analyzed according to a preset layout;
the multi-channel analog switch is connected with each sensor;
a plurality of digital-to-analog converters respectively connected with the analog switches correspondingly;
and the data processing module is connected with each digital-to-analog converter to acquire data information acquired by each sensor.
In some embodiments, the data acquisition module further comprises:
the wireless transmitting module is used for wirelessly transmitting the data information acquired by each sensor to the data analysis module; the wireless transmitting module is a zigbee module, and the data processing module is an STM32 module; the STM32 module integrates the SPI communication protocol.
In some embodiments, the data storage module employs a MySQL relational database for data storage.
The invention has the beneficial effects that at least:
in the bridge health online intelligent monitoring system, a data analysis module is arranged to load a bridge damage identification classification model obtained based on machine learning, abnormal detection is carried out on data collected by an acceleration sensor, and the health state of a bridge can be monitored automatically in real time. And a damage visualization module is arranged, so that under the condition that the abnormity of the bridge to be analyzed is detected, the bridge to be analyzed is subjected to damage analysis based on near-field dynamics, visual display is carried out, and the health state of the bridge to be analyzed is presented more intuitively.
Furthermore, a plurality of bridge damage identification classification models are arranged for joint monitoring, so that the accuracy of abnormal detection of data acquired by the sensor can be obviously improved.
Furthermore, the damage visualization module uses an OpenMP-based shared memory type thread-level parallel method in the near-field dynamics analysis process, so that the calculation speed is greatly increased, and the acceleration effect of 20-25 times can be realized in the hardware environment used by the system.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
It will be appreciated by those skilled in the art that the objects and advantages that can be achieved with the present invention are not limited to the specific details set forth above, and that these and other objects that can be achieved with the present invention will be more clearly understood from the detailed description that follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principle of the invention. In the drawings:
fig. 1 is a schematic structural diagram of an online intelligent monitoring system for bridge health according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an online intelligent monitoring system for bridge health according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of a sensor network structure in the bridge health online intelligent monitoring system according to an embodiment of the present invention;
FIG. 4 is a main page diagram of the data management plane of FIG. 2;
FIG. 5 is a schematic view illustrating an anomaly detection process of a data analysis module in the bridge health online intelligent monitoring system according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating a process of performing a near field dynamics analysis by the bridge health online intelligent monitoring system according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of the main interface of the statistical chart of FIG. 2.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the following embodiments and accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
It should be noted that, in order to avoid obscuring the present invention with unnecessary details, only the structures and/or processing steps closely related to the scheme according to the present invention are shown in the drawings, and other details not so relevant to the present invention are omitted.
It should be emphasized that the term "comprises/comprising" when used herein, is taken to specify the presence of stated features, elements, steps or components, but does not preclude the presence or addition of one or more other features, elements, steps or components.
It is also noted that, unless otherwise specified, the term "coupled" is used herein to refer not only to a direct connection, but also to an indirect connection with an intermediate.
In view of the shortcomings of the prior art, the present invention provides a system for online intelligent monitoring of bridge health status based on machine learning and near field dynamics theory (PD theory), so as to solve the above problems of the existing bridge health monitoring system in the background art. In the data collected by the bridge health online monitoring system sensor, most data are in a normal state, special attention and analysis are not needed, only a small part of data are abnormal data, and the occurrence of the abnormal data means that the bridge has certain safety risk and needs to be comprehensively analyzed and processed. The invention can quickly and accurately find the abnormal data from the mass data, can quickly and accurately identify the abnormal data in the acquired data by using a machine learning abnormal detection algorithm, and can carry out accurate damage analysis by a near-field dynamics theory.
Specifically, the present invention provides an online intelligent monitoring system for bridge health, as shown in fig. 1, at least comprising: the system comprises a data acquisition module, a data analysis module, a damage visualization module and a data storage module which are connected with each other.
The data acquisition module comprises a plurality of sensors, the sensors at least comprise acceleration sensors, and the sensors are arranged on the bridge to be analyzed according to a preset layout. The sensors can be uniformly distributed on the whole bridge to be analyzed, and can also be locally deployed according to specific requirements. The category of sensors may also include pressure sensors that can sense the load of the deck or support structure and temperature sensors that can sense the deck temperature to introduce further sensed parameters for analysis of the bridge health.
In some embodiments, the data acquisition module comprises: the system comprises a plurality of sensors, a plurality of sensors and a control system, wherein the sensors comprise a temperature sensor, a pressure sensor and an acceleration sensor, and are arranged on a bridge to be analyzed according to a preset layout; the multi-channel analog switch is connected with each sensor; a plurality of digital-to-analog converters respectively connected with the analog switches correspondingly; and the data processing module is connected with each digital-to-analog converter to acquire data information acquired by each sensor.
In some embodiments, the data acquisition module further comprises: the wireless transmitting module is used for wirelessly transmitting the data information acquired by each sensor to the data analysis module; the wireless transmitting module is a zigbee module, and the data processing module is an STM32 module; the STM32 module integrates the SPI communication protocol.
The data analysis module inputs the acceleration information acquired by the acceleration sensor into the bridge damage identification classification model to identify abnormal data in the acceleration information and judge whether the bridge to be analyzed is damaged; and selecting and calling the bridge damage identification classification model from preset toolbox plate blocks.
The tool box plate is used for loading a plurality of bridge damage identification and classification models, and in some embodiments, before a health detection task of a bridge to be analyzed is performed on the tool box plate, a training sample set is adopted for training to obtain each bridge damage identification and classification model.
Each bridge damage identification classification model is obtained by pre-training an initial classification model, and comprises the following steps of S101-S102:
step S101: the method comprises the steps of obtaining a sample training set, wherein the sample training set comprises a plurality of sample data, each sample data comprises sample acceleration signals and bridge subject state information, the sample acceleration signals are collected by a first set number of sensors which are uniformly distributed on a sample bridge, the bridge subject state information comprises a structure damage type and a structure damage-free type, and the bridge subject state information corresponding to each sample data is used as a label.
Step S102: taking a sample acceleration signal as an input and a label as an output, and respectively training a plurality of initial classification models by adopting a training sample set to obtain a plurality of bridge damage identification classification models; the initial classification model at least comprises a support vector machine, a decision tree, a fully connected neural network, a long-short term memory neural network or a self-organizing mapping algorithm network.
Specifically, the sample training set is obtained by sampling on a sample bridge, and illustratively, 1000 samples can be set in the sample training set, wherein the first 500 are sensor data signal sequences acquired on the sample bridge in an undamaged state, and the last 500 are sensor data signal sequences acquired on the sample bridge in a damaged state. In other embodiments, the sensor data signal sequences generated by the sample bridge in the damaged and undamaged states can also be modeled and used for training. And training by adopting a plurality of initial classification models based on the obtained training sample set, and solving the two-classification problem of the bridge damage state. The machine learning method capable of completing the two-classification problem comprises multiple types, the final effects generated by different algorithm structures and different modes are different, and in the embodiment, multiple bridge damage identification classification models are loaded in the tool box module for selection.
In some embodiments, before the training of the plurality of initial classification models with the training sample set, the method further includes: and denoising the sample acceleration signal by adopting Gaussian filtering, and completing missing data in the sample acceleration signal by adopting a K-neighbor algorithm.
Specifically, the toolbox plate integrates a plurality of bridge damage identification classification models, and one or more bridge damage identification classification models are selected according to preset rules to identify abnormal data in the acceleration information. Selecting one or more bridge damage identification classification models according to a preset rule to identify abnormal data in the acceleration information, wherein the method comprises the following steps of S201-S203:
step S201: and calculating evaluation indexes of each bridge damage identification classification model, wherein the evaluation indexes comprise accuracy, precision, recall and F1 scores.
Step S202: after the accuracy, precision ratio, recall ratio and F1 score of each bridge damage identification classification model are normalized, the comprehensive score of each bridge damage identification classification model is obtained through weighting and summing, and the comprehensive scores are arranged from high to low according to the comprehensive scores.
Step S203: selecting a bridge damage identification classification model with the highest comprehensive score to identify abnormal data in the acceleration information; or selecting a second set number of bridge damage identification classification models with higher comprehensive scores to identify the bridge damages, wherein the second set number is an odd number, and taking the identification results with larger number as final abnormal data identification results.
In steps S201 to S203, methods for selecting a bridge damage identification classification model are provided, comparison and selection are performed by calculating a comprehensive score, and in other embodiments, a user may freely select a bridge damage identification classification model in a tool box plate.
In some embodiments, the tool box plate further comprises an improved prediction model, the improved prediction model maps the output of each bridge damage identification classification model to the final abnormal data and has a result, and the improved prediction model is obtained by adopting full-connection neural network training.
Specifically, based on data in a sample training set, the output of each bridge damage recognition classification model obtained through training is used as input, the label (namely the classification result of the sample) of the corresponding sample is used as output, an improved prediction model is obtained through construction and training, the recognition results of each bridge damage recognition classification model are subjected to combined processing, the final recognition result is obtained through further analysis and evaluation, and the accuracy of damage state recognition is improved.
In this embodiment, it is first determined whether sensor data acquired by a bridge to be analyzed is abnormal through the data analysis module, and if not, damage analysis is not required, and only when abnormal data is identified, that is, when it is determined that a bridge to be analyzed is damaged, damage analysis is further performed, so that computational power is saved and unnecessary data processing is reduced.
In some embodiments, when the neural networks such as the fully-connected neural network, the long-short term memory neural network or the self-organizing mapping algorithm network are used for training, a transfer learning mode can be used, and the model can obtain better prediction performance under the conditions of less data volume and shorter time.
The damage visualization module is used for acquiring the position information and the acceleration information acquired by the acceleration sensor under the condition that the data analysis module judges that the bridge to be analyzed has damage, constructing an integral or local model of the bridge to be analyzed, establishing a near-field dynamic motion equation and solving the damage of the bridge to be analyzed at each time step to obtain a bridge damage result at each time step; and combining the data information and the bridge damage result acquired by each sensor to a model of the bridge to be analyzed for visual display.
In some embodiments, the method for constructing a whole or local model of a bridge to be analyzed, establishing a near-field dynamic motion equation, and solving the damage of the bridge to be analyzed at each time step includes: the method comprises the steps of obtaining material parameters including elastic modulus, density, Poisson ratio and tensile critical elongation, load parameters including load force and load speed, time step, ratio of radius of a near field region to side length of a point unit, and solving parameters of limiting time step, limiting displacement or crack arrest condition of limiting solving end conditions, setting side length of the point unit, a fixed constraint region, a working condition loading region and a material distribution region, constructing a near field dynamic motion equation of the bridge to be analyzed, solving damage of the bridge to be analyzed, and dynamically storing solving results of the time steps as bridge damage analysis results.
Specifically, a near-field dynamics theory is introduced for damage analysis. And establishing an integral or local model according to the actual condition of the bridge to be analyzed for integral analysis or local analysis. The near-field kinetic theory based on bonds is to describe the constitutive relation of a particle of matter (point element in discretization modeling) in the material and a particle of matter (point element in discretization modeling) in its near-field region R. The relationship between a point unit i and a certain point unit j in the near field area of the point unit i is also called a key, the interaction force existing between the point unit i and the point unit j in the key is called a near field force and is expressed by f and is also called an constitutive force function, the point unit and all the point units in the near field area of the point unit form a plurality of keys, and the relationship of the keys forms the constitutive relationship of the material. This constitutive relation is expressed as a motion equation in the form of an integral, and satisfies newton's second law. And (3) constructing a motion equation of the constitutive relation between a certain point unit i in the material at the time t and a point unit j in a near field region of the point unit i, solving the displacement, damage, deformation energy density and energy release rate of the point unit, and performing visual presentation.
Illustratively, a near-field dynamics motion equation is constructed for a bridge to be analyzed, the damage of the bridge to be analyzed is solved, the solved result of each time step is dynamically stored as a bridge damage analysis result, the method comprises the steps of constructing the motion equation in a near-field region for a local model of the bridge to be analyzed, and at the time t, the motion equation of the constitutive relation between a certain point unit i in a material and a point unit j in the near-field region is as follows:
where ρ is the material density of the dot unit i, VjIs the volume of the dot cell j, xiIs the position of the dot cell i, xjIs the position of the dot cell j, u (x)iT) is the displacement of the point unit i, u (x)jT) is the displacement of the dot cell j; f (eta, xi) is the constitutive force between point unit i and point unit j, b (x)iT) is the external force density applied to the point unit i; xi represents the relative position of the point unit i and the point unit j, and eta represents the relative displacement of the point unit i and the point unit j; xi is xi-xj，η＝u(xj，t)-u(xi，t)；
The bond-based near-field dynamics theory is used to solve the problem of damage to micro-elastic brittle materials, which are generally similar to micro-elastic brittle materials, and thus are used extensively in damage analysis of brittle materials. The constitutive force function of the micro elastic brittle material based on the near field dynamics of the bond is as follows:
wherein the content of the first and second substances,
wherein c is a micro modulus constant, E is an elastic modulus, v is a Poisson's ratio, s is an elongation of a bond, and s is0Is the critical elongation of the bond, mu is the amount of label for whether the bond is broken or not; g0Represents the fracture energy of the material, δ represents the radius of the near field region R;
for the dot cell i, the damage is represented by evaluating the breakage of the bond in the near field region, and the damage D is represented as:
wherein R represents the near field region, μ represents the amount of a label for whether a bond is broken, μ is 1 when a bond is normal, and μ is 0 when a bond is broken; dVjRepresents the volume integral quantity of the point unit j; wherein the damage is more than or equal to 0 and less than or equal to 1, 0 represents that the material point is not damaged, and 1 represents that the material point is completely damaged.
For point unit i, the total energy due to deformation of the near field region
Expressed as:
wherein w represents the energy of the bond between the dot cell i and the dot cell j, R represents the near field region, dVjRepresents the volume integral quantity of the point unit j; c is a micro modulus constant, s is an elongation of the bond, and xi represents the relative position of the point unit i and the point unit j;
recording the displacement and damage of each point unit in the near field region, and calculating the total energy of each point unit
The deformation energy density and the energy release rate are calculated.
In some embodiments, the damage visualization module establishes a near-field dynamics motion equation based on a shared memory thread-level parallel method of OpenMP and solves the damage of the bridge to be analyzed at each time step.
And the data storage module stores the data information acquired by each sensor, abnormal data identified by the data analysis module and the bridge damage result of each time step solved by the damage visualization module by adopting a relational database.
In some embodiments, the data storage module employs a MySQL relational database for data storage.
The invention is illustrated below with reference to specific examples:
the overall structure of the bridge health online intelligent monitoring system under the fusion of near field dynamics and machine learning is shown in fig. 2, wherein the bridge health online intelligent monitoring system mainly comprises four core modules of data acquisition, data management, data analysis and data visualization, the data acquisition function is realized by arranging a sensor network, the sensor network carries out grid deployment on various sensors in a monitored target, various data capable of reflecting the health condition of the target bridge structure are acquired in all directions in real time, and transmission among the data is displayed through multi-level nodes. The data management system carries out certain preprocessing on the received sensor data, including missing value processing, noise data processing and the like, carries out data fusion on heterogeneous data, stores historical data into a special database in a certain data organization mode, and further realizes searching, analysis and the like on the historical data by utilizing the database. The data analysis system is an abnormal data detection method based on machine learning, and in the data visualization system, the data alarm module is used for carrying out structural damage and crack propagation analysis through near-field dynamics, so that powerful theoretical support is provided for structural health prediction, and the method is also the core content of the embodiment. The data visualization system performs visualization presentation through the statistical chart, and data visualization enables a user to have overall assurance on the safety condition of the bridge through visualization processing on an analysis result, so that bridge abnormity can be found and processed in time.
In the aspect of communication among the modules, the modules are tightly connected to form an open-loop control network, so that the whole dynamic monitoring system is formed. The sensor network sends the acquired data to the database, and the data management module preprocesses the data and stores the sorted standard data into the database; the data analysis module finds out abnormal data through a machine learning abnormal detection algorithm; because the PD analysis method not only has good processing capacity on data, but also has very visual effect on the result of data processing, the system places the PD analysis method in a data visualization module; the data visualization module displays the analyzed abnormal state of the structure and the simulated structural damage condition and transmits the abnormal state of the structure and the simulated structural damage condition to the database for storage, the historical early warning data are stored while the real-time early warning of the abnormal state is realized, and the development of a new and better prediction algorithm is facilitated while the data are accumulated.
Besides the four core modules, in order to facilitate the operation and management of a user, the system is further provided with user setting, system setting and more three sections, the user can modify the information and parameters of the user through the user setting sections, a manager can manage the authority of a user through the system setting sections, and a system developer can develop more system functions in more sections subsequently.
In the data acquisition module, the main functions of the sensor network comprise two parts of data acquisition and data transmission. In the development and research of the current sensor network part, in order to reflect the structural state and meet the requirements of a PD algorithm, the acquired physical quantity mainly comprises pressure, acceleration and the like in the structure, the data transmission mode adopts Zigbee wireless transmission, the data signal control and wireless transmission tasks in the data acquisition are uniformly processed by an STM32 microcontroller, and the whole design scheme is shown in FIG. 3. The data is generated by a multi-channel sensor, transmitted to the interior of the STM32 through a multi-channel analog switch and a digital-to-analog converter, and finally transmitted to the far end through a wireless transmitting device. The data acquisition task is realized through the multisensor, multichannel analog switch and digital analog converter in STM32 come together, the multisensor includes pressure sensor, acceleration sensor and temperature sensor three kinds, adopt DS18B20 temperature sensor in the design, piezoresistive pressure sensor, temperature sensor is used for compensating in order to improve measurement accuracy to pressure sensor, multichannel analog switch and digital analog converter are used for controlling acquisition switch and analog-to-digital conversion, STM32 is inside to have integrateed 12 digit digital analog converter and the channel number has reached 18, this not only practiced thrift hardware cost but also make system architecture more succinct. Besides data communication between a wireless receiving end and a wireless transmitting end, the wireless transmitting module also needs to be communicated with an STM32 microcontroller to realize data transmission, an SPI communication protocol integrated in an STM32 is adopted, high-speed data transmission in a full-duplex working mode can be realized, and a relatively standard data form and content are formed when data are transmitted externally to form an external data interface.
The main function of the data management module is to store historical data of the sensors, the data analysis module and the data visualization module, and a main interface of the data management plate is shown in fig. 4. For data sent by the sensor network, data preprocessing methods such as Gaussian filtering can be used for denoising the data and complementing missing data with K neighbors, and then the data are integrated and stored in a database. The stored sensor data mainly comprises monitoring target names, sensor types, sensor positions, acquisition time, specific numerical values and the like, and can be inquired according to the acquisition time, the sensor positions and the like. The results of the data analysis module and the data visualization module are stored in a certain form, and mainly comprise near-field dynamics analysis results and abnormal data identified by machine learning, the near-field dynamics analysis results mainly comprise the name of a sensor related to analysis, an analysis position, coordinates of object points in an analysis model, damage of the object points, formation energy of the object points, damage release energy and the like, the state data identified by machine learning comprise normal data and abnormal data, and the data are stored and are convenient for later inquiry and research. The database is realized by using a MySQL relational database, and the data tables mainly comprise sensor data tables of different monitoring targets, a near field dynamics analysis result table, a machine learning anomaly detection result table and the like.
Because the monitoring target is in a normal state in most of time, if the data acquired at any time is analyzed by near-field dynamics, great resource demand and resource waste are caused, and the instantaneity is difficult to guarantee. Therefore, if the abnormal state of the system can be accurately identified according to the sensor data, the near field dynamics method can be used for analysis more reasonably, a large number of unnecessary analysis processes are avoided, and the real-time performance of the system is ensured. If normal data and abnormal data are regarded as different categories, the abnormal detection becomes a classification problem in the field of machine learning, and the problem of sample imbalance can be solved by using a down-sampling method and an over-sampling method. The machine learning anomaly detection algorithm used by the system comprises a nearest neighbor method, a decision tree, a random forest, Bagging, a support vector machine, a neural network and the like, wherein the input of each algorithm is data collected by a sensor, and the output of each algorithm is a predicted data state at the moment, the neural network model has better performance in two aspects of prediction performance and prediction time, and in addition, the neural network can quickly and effectively adjust the model through a fine-tune method in transfer learning when transferring among different distributed sample domains, so that the neural network can keep better prediction performance, for example, fig. 5 shows the process of machine learning anomaly analysis in the system.
The data analysis module has the main functions of analyzing the data in the data storage module through a machine learning algorithm, finding out a small part of abnormal data in massive data, and transmitting an analysis result to the data visualization module, wherein the analysis result is realized by a tool box board block in the system.
In the data visualization module, a near field dynamics analysis process is as shown in fig. 6, real-time data of a monitoring target acquired by sensors such as pressure and acceleration are combined with an established simulation structure, so that a model capable of being analyzed by using near field dynamics is obtained, in order to accelerate the near field dynamics analysis speed and improve the real-time performance of a system, a shared memory type thread parallel method based on OpenMP is adopted to perform near field dynamics solving calculation, and finally, model result data and some visualization results are generated, and the analysis results are stored in a database and are output for a system user to refer to. The used near-field dynamics theory is mainly key-based near-field dynamics analysis of the brittle material, mechanical characteristics of a key in the brittle material comprise three states of health, damage and damage, a damage function is added into the constitutive function to describe damage conditions of the key, the damage function uses a commonly used damage function based on an exponential function and a newly proposed damage function based on a polynomial, the used near-field dynamics can well simulate and analyze damage and damage conditions of the material, but the calculation speed is low when large-scale calculation is carried out, and the real-time performance of the system can be reduced. Therefore, the OpenMP-based shared memory type thread-level parallel method is used in the near-field dynamics analysis process, the calculation speed is greatly improved, 20-25 times of acceleration effect can be achieved under the hardware environment used by the system, the model analysis time of 20w object particles can be controlled within 12min, and the real-time performance of the system is greatly improved.
The data visualization module has the functions of visually displaying not only the sensor data integrated in the data management block and the abnormal data found through machine learning, but also the machine learning analysis results of the abnormal data found in the tool kit and the PD method analysis, and also visually displaying the PD method analysis results. The former is realized by a statistical chart block, the latter is realized by a data alarm block, and the main interface of the data alarm block is shown in figure 7.
The embodiment combines machine learning and near field dynamics, realizes online intelligent monitoring of bridge health, and can accurately evaluate and predict the safety of the bridge structure.
The embodiment uses the near field dynamics in the bridge health monitoring system, and can deeply analyze the abnormal data of the bridge by utilizing the natural advantages of the near field dynamics, more accurately simulate the damage and crack development of the bridge structure and visually display the damage and crack development.
In conclusion, in the bridge health online intelligent monitoring system, the data analysis module is arranged to load the bridge damage identification classification model obtained based on machine learning, so that the data acquired by the acceleration sensor is subjected to abnormal detection, and the bridge health state can be monitored automatically in real time. And a damage visualization module is arranged, so that under the condition that the abnormity of the bridge to be analyzed is detected, the bridge to be analyzed is subjected to damage analysis based on near-field dynamics, visual display is carried out, and the health state of the bridge to be analyzed is presented more intuitively.
Furthermore, a plurality of bridge damage identification classification models are arranged for joint monitoring, so that the accuracy of abnormal detection of data acquired by the sensor can be obviously improved.
Furthermore, the damage visualization module uses a shared memory type thread-level parallel method based on OpenMP in the near-field dynamics analysis process, so that the calculation speed is greatly increased, and the acceleration effect of 20-25 times can be realized in the hardware environment used by the system.
Those of ordinary skill in the art will appreciate that the various illustrative components, systems, and methods described in connection with the embodiments disclosed herein may be implemented as hardware, software, or combinations of both. Whether this is done in hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranets, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Features that are described and/or illustrated with respect to one embodiment may be used in the same way or in a similar way in one or more other embodiments and/or in combination with or instead of the features of the other embodiments in the present invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The utility model provides a bridge health online intelligent monitoring system which characterized in that includes:
the data acquisition module comprises a plurality of sensors, the sensors at least comprise acceleration sensors, and the sensors are arranged on the bridge to be analyzed according to a preset layout;
the data analysis module inputs the acceleration information acquired by the acceleration sensor into a bridge damage identification classification model to identify abnormal data in the acceleration information and judge whether the bridge to be analyzed is damaged; the bridge damage identification classification model is selected and called from preset toolbox plate blocks;
the damage visualization module is used for acquiring the position information and the acceleration information acquired by the acceleration sensor under the condition that the data analysis module judges that the bridge to be analyzed has damage, constructing an integral or local model for the bridge to be analyzed, establishing a near-field dynamic motion equation and solving the damage of the bridge to be analyzed at each time step to obtain a bridge damage result at each time step; combining the data information acquired by each sensor and the bridge damage result to the model of the bridge to be analyzed for visual display;
and the data storage module stores data information acquired by each sensor, abnormal data identified by the data analysis module and bridge damage results of each time step solved by the damage visualization module by adopting a relational database.
2. The online intelligent bridge health monitoring system of claim 1, wherein the toolbox board integrates a plurality of bridge damage identification classification models, and one or more bridge damage identification classification models are selected according to preset rules to identify abnormal data in the acceleration information;
each bridge damage identification classification model is obtained by pre-training an initial classification model and comprises the following steps:
acquiring a sample training set, wherein the sample training set comprises a plurality of sample data, each sample data comprises sample acceleration signals acquired by a first set number of sensors uniformly distributed on a sample bridge and bridge body state information, the bridge body state information comprises two types of structure damage and structure damage, and the bridge body state information corresponding to each sample data is used as a label;
taking the sample acceleration signal as input and the label as output, and adopting the training sample set to respectively train a plurality of initial classification models to obtain a plurality of bridge damage identification classification models; the initial classification model at least comprises a support vector machine, a decision tree, a fully connected neural network, a long-short term memory neural network or a self-organizing mapping algorithm network.
3. The bridge health online intelligent monitoring system of claim 2, before the training of the plurality of initial classification models with the training sample set, further comprising:
and denoising the sample acceleration signal by adopting Gaussian filtering, and completing missing data in the sample acceleration signal by adopting a K-neighbor algorithm.
4. The bridge health online intelligent monitoring system of claim 2, wherein selecting one or more bridge damage recognition classification models according to preset rules to identify abnormal data in the acceleration information comprises:
calculating evaluation indexes of each bridge damage identification classification model, wherein the evaluation indexes comprise accuracy, precision, recall and F1 scores;
after normalization processing is carried out on the accuracy, the precision ratio, the recall ratio and the F1 score of each bridge damage identification classification model, weighting and summing are carried out to obtain a comprehensive score of each bridge damage identification classification model, and the comprehensive scores are arranged from high to low;
selecting the bridge damage identification classification model with the highest comprehensive score to identify abnormal data in the acceleration information; or selecting a second set number of bridge damage identification classification models with higher comprehensive scores to identify the bridge damage, wherein the second set number is an odd number, and taking the identification result with a larger number as a final abnormal data identification result.
5. The online intelligent bridge health monitoring system of claim 2, wherein the toolbox board further comprises an improved prediction model, the improved prediction model maps the output of each bridge damage identification classification model to the final abnormal data with the result, and the improved prediction model is obtained by adopting a fully-connected neural network for training.
6. The online intelligent bridge health monitoring system of claim 1, wherein the damage visualization module builds a whole or local model for the bridge to be analyzed, builds a near-field dynamic motion equation, and solves damage of the bridge to be analyzed at each time step, and comprises:
the method comprises the steps of obtaining material parameters including elastic modulus, density, Poisson ratio and tensile critical elongation, load parameters including load force and load speed, time step, ratio of radius of a near field region to side length of a point unit, and solving parameters of limiting time step, limiting displacement or crack arrest condition of limiting solving end conditions, setting side length of the point unit, a fixed constraint region, a working condition loading region and a material distribution region, constructing a near field dynamic motion equation of the bridge to be analyzed, solving damage of the bridge to be analyzed, and dynamically storing solving results of the time steps as bridge damage analysis results.
7. The online intelligent bridge health monitoring system of claim 1, wherein the damage visualization module establishes a near-field dynamics equation of motion based on a shared memory thread-level parallel method of OpenMP and solves damage of the bridge to be analyzed at each time step.
8. The bridge health online intelligent monitoring system of claim 1, wherein the data acquisition module comprises:
the system comprises a plurality of sensors, a data processing system and a data processing system, wherein the sensors comprise a temperature sensor, a pressure sensor and an acceleration sensor, and are arranged on a bridge to be analyzed according to a preset layout;
the multi-channel analog switch is connected with each sensor;
a plurality of digital-to-analog converters respectively connected with the analog switches correspondingly;
and the data processing module is connected with each digital-to-analog converter to acquire data information acquired by each sensor.
9. The online intelligent bridge health monitoring system according to claim 7, wherein the data acquisition module further comprises:
the wireless transmitting module is used for wirelessly transmitting the data information acquired by each sensor to the data analysis module; the wireless transmitting module is a zigbee module, and the data processing module is an STM32 module; the STM32 module integrates the SPI communication protocol.
10. The bridge health online intelligent monitoring system of claim 1, wherein the data storage module employs a MySQL relational database for data storage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111592341.9A CN114429032A (en) | 2021-12-23 | 2021-12-23 | Bridge health online intelligent monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111592341.9A CN114429032A (en) | 2021-12-23 | 2021-12-23 | Bridge health online intelligent monitoring system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114429032A true CN114429032A (en) | 2022-05-03 |
Family
ID=81312224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111592341.9A Pending CN114429032A (en) | 2021-12-23 | 2021-12-23 | Bridge health online intelligent monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114429032A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114894411A (en) * | 2022-05-24 | 2022-08-12 | 东营固泰尔建筑科技有限责任公司 | Bridge health monitoring method and system based on wireless sensor network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017202139A1 (en) * | 2016-05-26 | 2017-11-30 | 东南大学 | Bridge damage identification method based on long-gauge-length strain influence envelope |
| CN111486893A (en) * | 2020-04-07 | 2020-08-04 | 中铁西南科学研究院有限公司 | Bridge structure health monitoring and early warning system and early warning method |
| CN113553749A (en) * | 2021-09-22 | 2021-10-26 | 深圳市城市交通规划设计研究中心股份有限公司 | Bridge health monitoring method, system, computer and storage medium |
-
2021
- 2021-12-23 CN CN202111592341.9A patent/CN114429032A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017202139A1 (en) * | 2016-05-26 | 2017-11-30 | 东南大学 | Bridge damage identification method based on long-gauge-length strain influence envelope |
| CN111486893A (en) * | 2020-04-07 | 2020-08-04 | 中铁西南科学研究院有限公司 | Bridge structure health monitoring and early warning system and early warning method |
| CN113553749A (en) * | 2021-09-22 | 2021-10-26 | 深圳市城市交通规划设计研究中心股份有限公司 | Bridge health monitoring method, system, computer and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 石春香 等: "基于键型近场动力学的含裂纹板损伤模拟", 《应用技术学报》 * |
| 郭健 等: "桥梁健康监测中的关键性问题和损伤识别方法", 《公路》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114894411A (en) * | 2022-05-24 | 2022-08-12 | 东营固泰尔建筑科技有限责任公司 | Bridge health monitoring method and system based on wireless sensor network |
| CN114894411B (en) * | 2022-05-24 | 2024-03-22 | 东营固泰尔建筑科技有限责任公司 | Bridge health monitoring method and system based on wireless sensor network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Salehi et al. | Emerging artificial intelligence methods in structural engineering | |
| CN110807562B (en) | Regional bridge risk prediction method and system | |
| CN114563150B (en) | Bridge health online detection module generation method, detection method, tool box and device | |
| CN210924883U (en) | Bridge structure health monitoring system | |
| CN117494292B (en) | Engineering progress management method and system based on BIM and AI large model | |
| CN101799674A (en) | Method for analyzing service state of numerical control equipment | |
| CN112926749A (en) | Intelligent power grid information equipment monitoring system and method | |
| CN115575584B (en) | Atmospheric environment cadmium content monitoring and early warning method and system | |
| CN114429032A (en) | Bridge health online intelligent monitoring system | |
| CN117807914A (en) | Real-time bridge stress detection method and system | |
| CN116306062A (en) | Intelligent monitoring method, system, device and storage medium for construction engineering | |
| CN114812984A (en) | Movable rapid monitoring system for railway bridge | |
| CN113177362A (en) | Furnace temperature prediction method and device based on furnace temperature soft measurement model | |
| Carrillo et al. | Design of efficient deep learning models for determining road surface condition from roadside camera images and weather data | |
| Rahman et al. | Remote condition monitoring of rail tracks using distributed acoustic sensing (DAS): a deep CNN-LSTM-SW based model | |
| CN112287586A (en) | Rapid reliability assessment method based on medium and small bridge damage identification result | |
| CN111062827B (en) | Engineering supervision method based on artificial intelligence mode | |
| CN115828164A (en) | Electric nail gun fault type identification method based on data driving | |
| CN114004138A (en) | Building monitoring method and system based on big data artificial intelligence and storage medium | |
| CN116730226B (en) | Safety intelligent supervision system and method for cantilever crane | |
| CN113392576A (en) | Staying balloon main cable tension state assessment early warning method | |
| CN114491944B (en) | Bridge damage analysis method, analysis alarm system and device | |
| CN113567247A (en) | Bridge detection information management system | |
| Duan et al. | Data mining technology for structural health monitoring | |
| CN116821800B (en) | Structure state classification method and related equipment based on semi-supervised generation countermeasure network |
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 |
Application publication date: 20220503 |
|
| RJ01 | Rejection of invention patent application after publication |