CN112491468B - FBG sensing network node fault positioning method based on twin node auxiliary sensing - Google Patents
FBG sensing network node fault positioning method based on twin node auxiliary sensing Download PDFInfo
- Publication number
- CN112491468B CN112491468B CN202011305192.9A CN202011305192A CN112491468B CN 112491468 B CN112491468 B CN 112491468B CN 202011305192 A CN202011305192 A CN 202011305192A CN 112491468 B CN112491468 B CN 112491468B
- Authority
- CN
- China
- Prior art keywords
- node
- sensing
- twin
- fbg
- cnn
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
- H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
- H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
- H04B10/0791—Fault location on the transmission path
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method for positioning a node fault of an FBG (fiber Bragg Grating) sensing network based on twin node auxiliary sensing, which comprises the following steps of: step S1, acquiring load original data of the FBG sensing network; step S2, preprocessing the load original data and constructing a characteristic data set; s3, constructing a CNN twin node prediction model and training; step S4, inputting the characteristic data into the trained CNN twin node prediction model to obtain a prediction data set; s5, constructing and training a CNN load positioning model; step S6, if the FBG sensor network to be tested has node failure, inputting the sensing node value in the corresponding neighborhood into a CNN twin node prediction model to predict the sensing node value, and obtaining a twin node wavelength prediction value; and step S7, inputting the obtained complete sensing information into the CNN load positioning model according to the normally operated entity node data set and the twin node wavelength predicted value to realize the load position detection. The invention realizes the detection of the load position with higher precision, thereby achieving the purpose of fault tolerance.
Description
Technical Field
The invention relates to the field of optical fiber sensing load positioning, in particular to a FBG sensing network node fault positioning method based on twin node auxiliary sensing.
Background
With the continuous development of society and economy, the construction of basic engineering such as various civil engineering buildings, bridges and roads is continuously developed and perfected, and a Structural Health Monitoring System (SHM) is used as an important infrastructure part in the large-scale basic construction, and the stable and reliable operation of the system can provide necessary information for the safety evaluation of the engineering buildings. The Fiber Bragg Grating (FBG) sensor has the characteristics of passive sensing, small volume, high sensitivity, strong anti-electromagnetic interference capability, corrosion resistance, reusability and the like, and is widely applied to acquisition of key structure information in the SHM system.
The FBG sensor is used as the most basic acquisition facility in the SHM system and has the characteristics of wide coverage, large quantity, long service time and the like. Although the FBG sensor has strong stability, due to the increasing intelligent and diversified requirements of the application scene, the complexity and the severity of the application environment are also continuously improved, and the FBG sensor inevitably has faults such as fracture or performance degradation under the long-term load action and the limitation of the construction process. Once a link or node fault occurs in the sensing network, the load information cannot be accurately sensed, and further the evaluation of the structure health state by the SHM is directly influenced. However, two difficulties exist in replacing a failed sensing node: firstly, the position of a fault node is difficult to locate, and once the fault node breaks down, the whole optical fiber link is often required to be replaced, so that the cost is too high; second, the FBG sensing network is usually buried in the monitored object as a part of the facility at the initial stage of construction, and the replacement of the failed node needs to destroy its external structure, which is likely to cause new potential safety hazard and secondary damage. Therefore, the method has important significance for accurately monitoring the structural state under the condition that partial nodes of the FBG sensing network fail and enhancing the reliability research of the FBG sensing network.
At present, in order to enhance the reliability of the FBG sensing network, researchers at home and abroad have conducted many researches, and the main research direction mainly focuses on enhancing the fault tolerance of the FBG sensing network node under the condition of a fault by using a special network topology or adding a backup link. Although these methods can also effectively improve the reliability of the network, the use of a special topology and the additional link increase undoubtedly increase the complexity of the network and the laying cost of the system, and at the same time, the method also causes a waste of sensing resources, which is not allowed in the application scenario that a sensor needs to be laid on a large scale. Therefore, it is very meaningful to use limited sensor resources to perform reliable sensing of the monitored object.
Disclosure of Invention
In view of this, the present invention provides a method for locating a node fault of an FBG sensing network based on twin node assisted sensing, which achieves reliable sensing of a monitored structural object without changing a network structure and adding an additional link, thereby effectively improving fault tolerance of the FBG sensing network.
In order to achieve the purpose, the invention adopts the following technical scheme:
a FBG sensing network node fault locating method based on twin node auxiliary sensing comprises the following steps:
step S1, acquiring load original data of the FBG sensing network;
step S2, preprocessing the load original data and constructing a characteristic data set;
s3, constructing a CNN twin node prediction model, and training based on the feature data set to obtain a trained CNN twin node prediction model;
step S4, inputting the characteristic data into the trained CNN twin node prediction model to obtain a prediction data set;
s5, constructing a CNN load positioning model, and training according to the characteristic data set and the prediction data set to obtain a trained CNN load positioning model;
step S6, if the FBG sensor network to be tested has node failure, inputting the sensing node value in the corresponding neighborhood into a CNN twin node prediction model to predict the sensing node value, and obtaining a twin node wavelength prediction value;
and step S7, inputting the obtained complete sensing information into the CNN load positioning model according to the normally operated entity node data set and the twin node wavelength predicted value to realize the load position detection.
Further, the step S1 is specifically: the load original data of the monitored structure is obtained by laying a preset number of FBG sensors in a distributed manner on the monitored structure.
Further, the step S2 is specifically:
step S21, removing and reducing the weak correlation characteristic quantity in the original data;
and step S22, normalizing the reduced data and the like to obtain a characteristic data set.
Further, in the step S21, a neighborhood discrimination method is specifically adopted to remove weak correlation feature quantities of the original feature data set to obtain the feature data set.
Further, the characteristic data set comprises horizontal and vertical coordinate position information of each FBG node, a corresponding FBG central wavelength change value and load point position information.
Further, the CNN twin node prediction model adopts a convolutional neural network, which includes a convolutional layer, a pooling layer and an output layer, taking an FBG sensor network with 4 × 4 nodes for a load positioning system as an example (including but not limited to a 4 × 4 sensor network), and assuming that the number of features in a neighborhood is 9, the structure of the CNN twin node prediction model is shown in table 1.
Further, the CNN load localization model adopts a convolutional neural network, which includes a convolutional layer, a pooling layer and an output layer, and also takes an FBG sensing network with 4 × 4 nodes for a load localization system as an example (including but not limited to a 4 × 4 sensing network), and the structure of the CNN load localization model is shown in table 2.
Compared with the prior art, the invention has the following beneficial effects:
the invention realizes the reliable sensing of the monitored structural object under the conditions of not changing the network structure and not increasing additional links, thereby effectively improving the fault tolerance of the FBG sensing network.
Drawings
FIG. 1 is a load positioning system of an FBG sensor network in an embodiment of the present invention;
fig. 2 is a basic structure diagram of a CNN network according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating a convolution operation process according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a pooling operation according to an embodiment of the present invention;
FIG. 5 is a schematic of the process of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
Referring to fig. 5, the invention introduces a twin node to replace and compensate a failed entity node, and then jointly predicts load point information by combining sensing information of the twin node and a normal node, and particularly provides a FBG sensing network node fault location method based on twin node auxiliary sensing, which comprises the following steps:
step S1, laying a preset number of FBG sensors on the monitored structure in a distributed manner to obtain load original data of the monitored structure; when the FBG sensor is subjected to the action of external load, the FBG sensors generate wavelength drift of different degrees, information such as central wavelength deviation of the FBG sensors in different quantities and different positions is obtained through the acquisition system,
step S2, preprocessing load original data, wherein the original data can not be directly used for model training after being obtained, primary processing is required, weak related characteristic quantities in the original data are firstly removed and reduced in order to ensure that the extracted characteristic quantities can fully reflect concentrated load position information, and meanwhile, the sample data are required to be similar in scale and obvious in difference, normalization and other processing is required to construct a characteristic data set;
s3, constructing a CNN twin node prediction model, and training based on the feature data set to obtain a trained CNN twin node prediction model;
in this embodiment, preferably, with the central position of the twin node as a center of a circle, the sensing values (including the nodes on the boundary) of all entity nodes in the circle with the radius r (r is a constant between the minimum distance and the maximum distance between the twin node and the entity node) are input to the convolutional neural network together for training to obtain the corresponding CNN twin node prediction model.
Preferably, the characteristic data set includes information on the horizontal and vertical coordinate positions of each FBG node, a change value of the FBG center wavelength, and information on the load point position
Step S4, inputting the characteristic data into the trained CNN twin node prediction model,
predicting a data set;
s4, constructing a CNN load positioning model, and training according to the characteristic data set and the prediction data set to obtain a trained CNN load positioning model;
step S5, if the FBG sensor network to be tested has node failure, inputting the sensing node value in the corresponding neighborhood into a CNN twin node prediction model to predict the sensing node value, and obtaining a twin node wavelength prediction value;
and step S6, inputting the obtained complete sensing information into the CNN load positioning model according to the normally operated entity node data set and the twin node wavelength predicted value to realize the load position detection.
In this embodiment, the convolutional neural network is composed of a plurality of network structures, mainly including a convolutional layer, a pooling layer, and an output layer, as shown in fig. 2, a two-dimensional image data is input, the output of the previous layer is used as the input of the next layer, the deep-level features of the original image information are extracted after the operations of the convolutional layer and the pooling layer, and finally the obtained feature information is transmitted to the full-connection layer to form the output layer.
Convolutional layers are core network layers in convolutional networks, the network structure of which is generally composed of a plurality of convolutional cores. The convolution kernel performs feature extraction on an input image through convolution operation, fig. 3 specifically demonstrates the convolution operation process, the size of the input image is set to be 5 × 5, the size of the convolution kernel is set to be 3 × 3, the convolution kernel starts to slide on the image according to a fixed step length, the sliding step length is set to be 1, feature information of a corresponding image can be extracted through convolution operation once the convolution kernel slides, and when the convolution kernel completes the traversal process of the whole image, a new feature image can be finally obtained. The formula of the convolution operation is shown as follows:
in the formula (I), the compound is shown in the specification,represents the jth characteristic diagram of the first convolutional layer,convolution kernel, M, representing the jth feature map in the jth convolutional layerjRepresents a collection of input-layer feature maps,represents the offset vector corresponding to the jth characteristic map of the ith convolutional layer, and f () represents an activation function. The result of the convolution operation is brought into the activation function, mainly to enhance the learning ability of the network facing the non-linear problem.
The pooling layer is an operation of the convolutional neural network for reducing the number of parameters and reducing the dimension of the input characteristic image, the input information of an original image is possibly large, and the characteristic information extracted through convolutional layer convolutional operation is still large, so that not only is the burden brought to calculation, but also the problem of overfitting is caused. Therefore, a pooling layer is generally inserted between convolutional layers of the convolutional neural network, and the pooling layer further extracts more representative feature information in the feature image output by the convolutional layers, and simultaneously reduces the dimension of the feature information, and the process is regarded as a pooling operation. Common pooling operations include maximum pooling and average pooling, as shown in fig. 4, a 2 × 2 pooling kernel is used to slide on an input image, the sliding step is set to 2, when the pooling kernel slides to a certain region, the maximum value of the region is taken as a pooled output result by the maximum pooling operation, the average pooling operation is taken as a pooled output result by the average pooling operation, after pooling calculation, the size of an output feature image is 2 × 2, which is only one fourth of the size of the original input image, thereby reducing the dimension of the feature image, and simultaneously, keeping effective feature information.
The output layer, which is the last layer of the convolutional neural network, typically takes the form of a conventional full connection, with each neuron in the layer being fully connected to each neuron in the previous layer. After the input original image is subjected to a plurality of continuous convolution and pooling operations, a plurality of different values are obtained through a full connection layer to form a one-dimensional characteristic vector y (y)1,y2,y3,...,yn) And n is the number of categories, and is recorded as:
in the formula (I), the compound is shown in the specification,is the weight of the jth neuron in the Lth connected layer,is the weight of the jth neuron in the Lth fully-connected layer. For the regression problem, the output value y of the convolutional neural networkiIs the final result, the output value y of the convolutional neural network for the classification problemiThen it needs to be substituted into the Softmax function to count the probability size of each category.
In this embodiment, the CNN twin node prediction model adopts a convolutional neural network, which includes a convolutional layer, a pooling layer and an output layer, and takes an FBG sensor network with 4 × 4 nodes for a load positioning system as an example (including but not limited to a 4 × 4 sensor network), and assuming that the number of features in a neighborhood is 9, the structure of the CNN twin node prediction model is shown in table 1.
In this embodiment, the CNN load localization model adopts a convolutional neural network, which includes a convolutional layer, a pooling layer and an output layer, and also takes an FBG sensing network with 4 × 4 nodes for a load localization system as an example (including but not limited to a 4 × 4 sensing network), and the structure of the CNN load localization model is shown in table 2.
Table 1 shows the CNN twin node prediction model structure in one embodiment of the present invention
Table 2 shows the structure of the CNN load localization model in an embodiment of the present invention;
the above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (6)
1. A FBG sensing network node fault locating method based on twin node auxiliary sensing is characterized by comprising the following steps:
step S1, acquiring load original data of the FBG sensing network;
step S2, preprocessing the load original data and constructing a characteristic data set;
s3, constructing a CNN twin node prediction model, and training based on the feature data set to obtain a trained CNN twin node prediction model;
step S4, inputting the characteristic data into the trained CNN twin node prediction model to obtain a prediction data set;
s5, constructing a CNN load positioning model, and training according to the characteristic data set and the prediction data set to obtain a trained CNN load positioning model;
step S6, if the FBG sensor network to be tested has node failure, inputting the sensing node value in the corresponding neighborhood into a CNN twin node prediction model to predict the sensing node value, and obtaining a twin node wavelength prediction value;
and step S7, inputting the obtained complete sensing information into the CNN load positioning model according to the normally operated entity node data set and the twin node wavelength predicted value to realize the load position detection.
2. The method for locating the node fault of the FBG sensing network based on the auxiliary sensing of the twin node according to claim 1, wherein the step S1 is specifically as follows: the load original data of the monitored structure is obtained by laying a preset number of FBG sensors in a distributed manner on the monitored structure.
3. The method for locating the node fault of the FBG sensing network based on the auxiliary sensing of the twin node according to claim 1, wherein the step S2 is specifically as follows:
step S21, removing and reducing the weak correlation characteristic quantity in the original data;
and step S22, carrying out normalization processing on the reduced data to obtain a characteristic data set.
4. The FBG sensing network node fault locating method based on the twin node auxiliary sensing according to claim 3, characterized in that the step S21 is specifically to remove feature quantities of weak correlation features of original data by using a neighborhood discrimination method to obtain a feature data set.
5. The FBG sensing network node fault locating method based on twin node auxiliary sensing as claimed in claim 1, wherein the characteristic data set comprises horizontal and vertical coordinate position information of each FBG node, a corresponding FBG center wavelength variation value and load point position information.
6. The FBG sensing network node fault location method based on twin node auxiliary sensing as claimed in claim 1, wherein the CNN twin node prediction model and the CNN load location model both adopt convolutional neural networks, including convolutional layers, pooling layers and output layers.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011305192.9A CN112491468B (en) | 2020-11-20 | 2020-11-20 | FBG sensing network node fault positioning method based on twin node auxiliary sensing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011305192.9A CN112491468B (en) | 2020-11-20 | 2020-11-20 | FBG sensing network node fault positioning method based on twin node auxiliary sensing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112491468A CN112491468A (en) | 2021-03-12 |
CN112491468B true CN112491468B (en) | 2022-04-01 |
Family
ID=74932183
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011305192.9A Active CN112491468B (en) | 2020-11-20 | 2020-11-20 | FBG sensing network node fault positioning method based on twin node auxiliary sensing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112491468B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117498931B (en) * | 2023-10-31 | 2024-06-11 | 国网山东省电力公司德州供电公司 | Efficient optical cable inspection system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108197014A (en) * | 2017-12-29 | 2018-06-22 | 东软集团股份有限公司 | Method for diagnosing faults, device and computer equipment |
CN110225075A (en) * | 2019-03-25 | 2019-09-10 | 北京快电科技有限公司 | A kind of building energy internet wisdom operation cloud operating system |
CN111562769A (en) * | 2019-02-14 | 2020-08-21 | 罗克韦尔自动化技术公司 | AI extension and intelligent model validation for industrial digital twinning |
CN111898632A (en) * | 2020-06-19 | 2020-11-06 | 广州杰赛科技股份有限公司 | Equipment fault diagnosis method and system |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6513025B1 (en) * | 1999-12-09 | 2003-01-28 | Teradyne, Inc. | Multistage machine learning process |
US20120031193A1 (en) * | 2009-04-01 | 2012-02-09 | Purdue Research Foundation | Identification of loads acting on an object |
CN104091035B (en) * | 2014-07-30 | 2017-02-08 | 中国科学院空间应用工程与技术中心 | Health monitoring method for effective loads of space station based on data-driven algorithm |
CN106528945B (en) * | 2016-10-17 | 2019-04-19 | 南京航空航天大学 | Based on the load normalized hardened structure load position of angle and size discrimination method |
US10321344B2 (en) * | 2017-02-17 | 2019-06-11 | Cisco Technology, Inc. | System and method to facilitate troubleshooting and predicting application performance in wireless networks |
KR101984730B1 (en) * | 2018-10-23 | 2019-06-03 | (주) 글루시스 | Automatic predicting system for server failure and automatic predicting method for server failure |
CN109543746B (en) * | 2018-11-20 | 2019-09-10 | 河海大学 | A kind of sensor network Events Fusion and decision-making technique based on node reliability |
CN109815892A (en) * | 2019-01-22 | 2019-05-28 | 武汉理工大学 | The signal recognition method of distributed fiber grating sensing network based on CNN |
CN109743103B (en) * | 2019-02-01 | 2021-07-27 | 福州大学 | FBG sensing network node fault repairing method based on ELM |
CN110020651B (en) * | 2019-04-19 | 2022-07-08 | 福州大学 | License plate detection and positioning method based on deep learning network |
CN110324081B (en) * | 2019-07-17 | 2022-08-05 | 福州大学 | Distributed multi-node cooperative fiber Bragg grating sensor fault positioning method |
CN110807400A (en) * | 2019-10-29 | 2020-02-18 | 北京师范大学 | Twin network-based collapse hidden danger characteristic information extraction method |
CN111400930B (en) * | 2020-04-09 | 2022-04-15 | 武汉大学 | Power equipment small sample fault diagnosis method and system based on virtual and real twin space |
CN111737909B (en) * | 2020-06-10 | 2021-02-09 | 哈尔滨工业大学 | Structural health monitoring data anomaly identification method based on space-time graph convolutional network |
CN111931819A (en) * | 2020-07-13 | 2020-11-13 | 江苏大学 | Machine fault prediction and classification method based on deep learning |
-
2020
- 2020-11-20 CN CN202011305192.9A patent/CN112491468B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108197014A (en) * | 2017-12-29 | 2018-06-22 | 东软集团股份有限公司 | Method for diagnosing faults, device and computer equipment |
CN111562769A (en) * | 2019-02-14 | 2020-08-21 | 罗克韦尔自动化技术公司 | AI extension and intelligent model validation for industrial digital twinning |
CN110225075A (en) * | 2019-03-25 | 2019-09-10 | 北京快电科技有限公司 | A kind of building energy internet wisdom operation cloud operating system |
CN111898632A (en) * | 2020-06-19 | 2020-11-06 | 广州杰赛科技股份有限公司 | Equipment fault diagnosis method and system |
Also Published As
Publication number | Publication date |
---|---|
CN112491468A (en) | 2021-03-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11562224B2 (en) | 1D-CNN-based distributed optical fiber sensing signal feature learning and classification method | |
Wu et al. | Critical review of data-driven decision-making in bridge operation and maintenance | |
Xia et al. | Long‐term performance monitoring and assessment of concrete beam bridges using neutral axis indicator | |
CN113971463A (en) | Heat supply pipeline risk distribution analysis method and routing inspection path planning system | |
Tian et al. | Relationship modeling between vehicle‐induced girder vertical deflection and cable tension by BiLSTM using field monitoring data of a cable‐stayed bridge | |
CN113011763B (en) | Bridge damage identification method based on space-time diagram convolution attention | |
CN115618273A (en) | Railway track state evaluation method and system based on parallel graph convolution neural network | |
CN112491468B (en) | FBG sensing network node fault positioning method based on twin node auxiliary sensing | |
CN107295537A (en) | A kind of method and system for wireless sensor network reliability of testing and assessing | |
CN113256019A (en) | Geological disaster hidden danger sensitivity prediction method based on unsupervised characterization learning | |
CN115654381A (en) | Water supply pipeline leakage detection method based on graph neural network | |
CN113591371B (en) | Bridge cluster structure damage positioning method based on space-time correlation model | |
CN116909788A (en) | Multi-mode fault diagnosis method and system with unchanged task direction and visual angle | |
Ma et al. | Nonlinear uncertainty modeling between bridge frequencies and multiple environmental factors based on monitoring data | |
CN111340975A (en) | Abnormal data feature extraction method, device, equipment and storage medium | |
Fan et al. | Multi-agent system for tunnel-settlement monitoring: A case study in Shanghai | |
CN112215197A (en) | Underground cable fault early warning method and system based on artificial intelligence | |
CN116259161B (en) | Power failure early warning system | |
CN116862076A (en) | Drainage pipe network flow prediction method, device and storage medium | |
CN115660198B (en) | Method for predicting residual service life of rolling bearing | |
Lei et al. | Fiber Bragg grating smart material and structural health monitoring system based on digital twin drive | |
Kumar et al. | Multiple layer radial basis neural network with remora regression tree optimum feature extraction for structural health monitoring | |
Chun et al. | An application of data fusion technology in structural health monitoring and damage identification | |
CN117407824B (en) | Health detection method, equipment and medium of power time synchronization device | |
Pamuncak | Deep learning 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |