CN116187177A - Steel structure damage prediction method based on Beidou system - Google Patents
Steel structure damage prediction method based on Beidou system Download PDFInfo
- Publication number
- CN116187177A CN116187177A CN202310058445.4A CN202310058445A CN116187177A CN 116187177 A CN116187177 A CN 116187177A CN 202310058445 A CN202310058445 A CN 202310058445A CN 116187177 A CN116187177 A CN 116187177A
- Authority
- CN
- China
- Prior art keywords
- steel structure
- damage
- curvature
- training
- learning
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B7/00—Measuring arrangements characterised by the use of electric or magnetic techniques
- G01B7/16—Measuring arrangements characterised by the use of electric or magnetic techniques for measuring the deformation in a solid, e.g. by resistance strain gauge
-
- 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
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a steel structure damage prediction method based on a Beidou system, which comprises the steps of firstly constructing a curvature damage model according to deflection changes of constructed or in-service steel structure members, then obtaining self vibration mode characteristics of different steel structure members under different load excitation and damage conditions, building an intelligent classification evaluation mode for the damage state of the steel structure through artificial intelligence learning and training, monitoring deflection changes of corresponding measuring points by the Beidou system, constructing the curvature damage model according to monitored data, further positioning the damage position of the structure, enabling the position detection accuracy of damage points of the steel structure members to be high, and predicting further development of damage inside the structure after the steel structure members are collected by a system, and early warning is carried out in advance. The method solves the problem that the existing steel structure damage judging method can not truly and effectively reflect the damage state.
Description
Technical Field
The invention relates to the technical field, in particular to a steel structure damage prediction method based on a Beidou system.
Background
Compared with a concrete structure and a masonry structure, the steel structure has the advantages of high strength, less materials, good ductility and plasticity and the like, and is therefore commonly used for large-span plants, high-rise buildings (signal towers and television towers), bridges and the like. The steel structure has poor fire resistance and corrosion resistance. In the long-term use process, the steel structure is inevitably damaged and destroyed, which leads to the shortening of the normal service life of the structure and even threatens to property and life safety. During the service period of the steel structure, the damage identification and evaluation are effectively carried out on the steel structure, and the real state of the steel structure is detected in real time, so that the method has great significance for ensuring the safety of the steel structure and avoiding the occurrence of serious disastrous accidents.
Most of the current damage judging methods of the steel structure are to extract the changes of stress, strain or displacement through health monitoring. However, these indicators may fluctuate in response to the external environment, but such changes do not account for the damaged condition of the structure.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art.
Disclosure of Invention
In order to overcome the defects existing in the prior art, a steel structure damage prediction method based on a Beidou system is provided, so that the problem that the existing steel structure damage judging method cannot truly and effectively reflect the damage state is solved.
In order to achieve the above purpose, the steel structure damage prediction method based on the Beidou system comprises the following steps:
a. a curvature damage model of the steel structure is constructed,
the curvature damage model is
In the formula (1), h is the height of the steel structure;
p is the curvature radius of the steel structure;
y is beam bending deflection of the steel structure;
delta is the distance between the measuring point of the steel structure and the neutral line;
m is the section position of the steel structure;
b. acquiring displacement and curvature at known damage points of a plurality of steel structures in service as learning samples;
c. an artificial intelligence learning training method of a support vector machine is adopted for the learning sample to obtain a displacement prediction threshold value for identifying the damage state of the steel structure;
d. acquiring displacement monitoring data of a measuring point of a steel structure to be monitored through a Beidou system;
e. calculating and determining the position of a damage point of the steel structure to be monitored based on the curvature damage model and the displacement monitoring data;
f. identifying whether the steel structure is damaged or not based on the displacement monitoring data and the displacement prediction threshold, generating a first alarm signal when the steel structure is damaged, acquiring the displacement monitoring data in a first preset time period when the steel structure is not damaged, predicting a curvature mode of the steel structure in a second preset time period by adopting a BP neural network algorithm, and generating a second alarm signal when the curvature mode exceeds the displacement prediction threshold;
g. and d-g, repeating the steps when the curvature mode is smaller than the displacement prediction threshold.
Further, the curvature damage model is constructed based on the bending static force and the structural deformation of the steel structure.
Further, the step a includes:
a first expression of the bending static force of the steel structure is obtained,
the first expression formula is
In the first expression formula, m is the section position of the steel structure, and p m Radius of curvature, M, of steel structure m Is a section bending moment E m I m Is the bending rigidity of the section;
obtaining a second expression formula of structural deformation of the steel structure,
the second expression formula is
In the second expression formula, x is the length of the steel structure, and y is the beam bending deflection;
and substituting the first expression formula into the second expression formula based on three measuring points which are continuously and equidistantly arranged on the steel structure to obtain the curvature damage model.
Further, the step of obtaining displacement monitoring data in a first preset time period to predict a curvature mode of the steel structure in a second preset time period by adopting a BP neural network algorithm includes:
monitoring the displacementThe measured data is used as a training set to determine the hidden layer node number q, the learning rate eta and the training termination condition of the neural network, wherein X is the input vector (X 1 ,x 2 ,…,x n ) T Y is the output vector y= (Y) of the training set 1 ,y 2 ,…,y n ) M groups of training samples;
determining hidden layers and weights, wherein a weight matrix between the input layer and the hidden layers is (v) 1 ,v 2 ,…,v m ) T The weight matrix between the hidden layer and the output layer is (w, w) 2 ,…,w m ) T Automatically adjusting the weight matrix after determining the number of hidden layers;
the Sigmoid function is selected as a transfer function, the quasi-newton method is adopted as a training function, and Trainbfg is selected as a training function. Then creating a BP network by using a feed forward net function;
and inputting the collected sample data into a model, calculating a learning and training sample through a function, and in the learning and training process, when the output weight Y value is adjusted to the output result Y according to the learning condition, stopping the learning and training and outputting the curvature mode predicted value of the steel structure when the confidence coefficient is more than 95%.
The steel structure damage prediction method based on the Beidou system has the advantages that firstly, a curvature damage model is built according to deflection changes of constructed or in-service steel structure members, then, self vibration mode characteristics of different steel structure members are obtained under different load excitation and damage conditions, an intelligent classification and grading evaluation mode for the steel structure damage state is built through artificial intelligent learning and training, deflection changes of corresponding measuring points are monitored by the Beidou system, the curvature damage model is built according to monitored data, the structure damage position is further positioned, so that the position detection accuracy of the steel structure member damage point is high, further development of internal damage of the structure is predicted after enough data are collected by the system, and early warning is carried out.
Detailed Description
The present application is described in further detail below with reference to examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail with reference to examples.
The invention provides a steel structure damage prediction method based on a Beidou system, which comprises the following steps of:
a. and constructing a curvature damage model of the steel structure.
The curvature damage model is
In the formula (1), h is the height of the steel structure;
p is the curvature radius of the steel structure;
y is beam bending deflection of the steel structure;
delta is the distance between the measuring point of the steel structure and the neutral axis of the steel structure;
m is the section position of the steel structure.
In this embodiment, the curvature damage model is constructed based on the bending static force and the structural deformation of the steel structure.
Specifically, the step a includes:
a1, acquiring a first expression formula of the bending static force of the steel structure.
The first expression formula is
In the first expression, m is the section position of the steel structure, p m Radius of curvature, M, of steel structure m Is a section bending moment E m I m Is the bending rigidity of the section;
a2, obtaining a second expression formula of structural deformation of the steel structure.
The second expression formula is
In the second expression, x is the length of the steel structure and y is the beam bending deflection.
a3, substituting the first expression formula into the second expression formula based on three continuous and equidistant measuring points on the steel structure to obtain the curvature damage model.
b. The displacement and curvature at known damage points of a plurality of steel structures in service are collected as learning samples.
c. And obtaining a displacement prediction threshold value for identifying the damage state of the steel structure by adopting an artificial intelligence learning training method of a support vector machine for the learning sample.
Step c comprises the steps of:
c1, setting a linear sample set
Wherein x is i Is a sample, y i Is a class symbol, i is the number of samples, a classification equation wx+b=0 is constructed and the samples are classified into two classes, the classification equation satisfies y i [(w·x)+b]-1 is greater than or equal to 0; i=1, 2, …, l, where w is the participation matrix of the sample and b is the class threshold.
c2 classification interval of D =2/||w|, the split interval maximally corresponds to the norm of the participation matrix of the sample being minimal and satisfies y i [(w·x)+b]-1 is greater than or equal to 0; i=1, 2, …, the smallest classification plane is the optimal classification plane.
c3 based on
Definition of Lagrangian function->
Wherein a is i Is a lagrange multiplier.
c4, converting the problems into dual problems
/>
The problem is a quadratic linear programming problem, a unique solution exists, and only a small part of a exists in the solution i If the sample is not zero, the corresponding sample is the support vector, and the optimal classification function is obtained finally>
Wherein n is<1 is the number of support vectors; x is x i Is a support vector; a, a i Is the corresponding lagrangian multiplier; b is a classification threshold, which can be obtained by taking the median value from any pair of support vectors in two classes.
d. And acquiring displacement monitoring data of the measuring points of the steel structure to be monitored through the Beidou system.
e. And calculating and determining the position of the damage point of the steel structure to be monitored based on the curvature damage model and the displacement monitoring data.
f. Based on displacement monitoring data and a displacement prediction threshold value, whether the steel structure is damaged is identified, when the steel structure is damaged, a first alarm signal is generated, when the steel structure is not damaged, the displacement monitoring data in a first preset time period are obtained to predict the curvature mode of the steel structure in a second preset time period by adopting a BP neural network algorithm, and when the curvature mode exceeds the displacement prediction threshold value, a second alarm signal is generated.
Specifically, the step of obtaining displacement monitoring data in the first preset time period in the step f to predict the curvature mode in the second preset time period of the steel structure by adopting the BP neural network algorithm includes:
f1, taking displacement monitoring data as a training set D, and determining hidden layer node number q, learning rate eta and training termination condition of the neural network, wherein X is input vector (X 1 ,x 2 ,…,x n ) T Y is the output vector y= (Y) of the training set 1 ,y 2 ,…,y n ) M training samples are used.
f2, determining the hidden layer and the weight, wherein the weight matrix between the input layer and the hidden layer is (v) 1 ,v 2 ,…,v m ) T The weight matrix between the hidden layer and the output layer is (w, w) 2 ,…,v m ) T The weight matrix is automatically adjusted after determining the number of hidden layers.
f3, selecting a Sigmoid function as a transfer function, adopting a quasi-Newton method as a training function, and selecting Trainbfg as a training function. The BP network is then created using the Feedforwardnet function.
And f4, inputting the collected sample data into a model, calculating learning and training samples through a function, and in the learning and training process, when the output weight Y value is adjusted to the output result Y according to the learning condition, stopping the learning and training and outputting the curvature mode predicted value of the steel structure when the confidence coefficient is more than 95%.
g. And when the curvature mode is smaller than the displacement prediction threshold value, repeating the steps d-g.
According to the steel structure damage prediction method based on the Beidou system, firstly, a curvature damage model is built according to deflection changes of constructed or in-service steel structure members, then, self vibration mode characteristics are obtained under different load excitation and damage conditions through different steel structure members, an intelligent classification and grading evaluation mode for the steel structure damage state is built through artificial intelligent learning and training, the deflection changes of corresponding measuring points are monitored through the Beidou system, the curvature damage model is built according to monitored data, the damage positions of the structure are further positioned, so that the position detection accuracy of the damage points of the steel structure members is high, and further development of internal damage of the structure is predicted after enough data are collected by the system, so that early warning is performed in advance.
The foregoing description is only of the preferred embodiments of the present application and is presented as a description of the principles of the technology being utilized. It will be appreciated by persons skilled in the art that the scope of the invention referred to in this application is not limited to the specific combinations of features described above, but it is intended to cover other embodiments in which any combination of features described above or equivalents thereof is possible without departing from the spirit of the invention. Such as the above-described features and technical features having similar functions (but not limited to) disclosed in the present application are replaced with each other.
Claims (4)
1. The steel structure damage prediction method based on the Beidou system is characterized by comprising the following steps of:
a. a curvature damage model of the steel structure is constructed,
the curvature damage model is
In the formula (1), h is the height of the steel structure;
p is the curvature radius of the steel structure;
y is beam bending deflection of the steel structure;
delta is the distance between the measuring point of the steel structure and the neutral line;
m is the section position of the steel structure;
b. acquiring displacement and curvature at known damage points of a plurality of steel structures in service as learning samples;
c. an artificial intelligence learning training method of a support vector machine is adopted for the learning sample to obtain a displacement prediction threshold value for identifying the damage state of the steel structure;
d. acquiring displacement monitoring data of a measuring point of a steel structure to be monitored through a Beidou system;
e. calculating and determining the position of a damage point of the steel structure to be monitored based on the curvature damage model and the displacement monitoring data;
f. identifying whether the steel structure is damaged or not based on the displacement monitoring data and the displacement prediction threshold, generating a first alarm signal when the steel structure is damaged, acquiring the displacement monitoring data in a first preset time period when the steel structure is not damaged, predicting a curvature mode of the steel structure in a second preset time period by adopting a BP neural network algorithm, and generating a second alarm signal when the curvature mode exceeds the displacement prediction threshold;
g. and d-g, repeating the steps when the curvature mode is smaller than the displacement prediction threshold.
2. The Beidou system-based steel structure damage prediction method of claim 1, wherein the curvature damage model is constructed based on bending static force and structural deformation of a steel structure.
3. The steel structure damage prediction method based on the beidou system according to claim 2, wherein the step a includes:
a first expression of the bending static force of the steel structure is obtained,
the first expression formula is
In the first expression formula, m is the section position of the steel structure, and p m Radius of curvature, M, of steel structure m Is a section bending moment E m I m Is the bending rigidity of the section;
obtaining a second expression formula of structural deformation of the steel structure,
the second expression formula is
In the second expression formula, x is the length of the steel structure, and y is the beam bending deflection;
and substituting the first expression formula into the second expression formula based on three measuring points which are continuously and equidistantly arranged on the steel structure to obtain the curvature damage model.
4. The method for predicting damage to a steel structure based on a beidou system according to claim 1, wherein the step of obtaining displacement monitoring data in a first preset time period to predict a curvature mode of the steel structure in a second preset time period by adopting a BP neural network algorithm includes:
the displacement monitoring data is used as a training set, and the hidden layer node number q, the learning rate eta and the training termination condition of the neural network are determined, wherein X is the input vector (X 1 ,x 2 ,…,x n ) T Y is the output vector y= (Y) of the training set 1 ,y 2 ,…,y n ) M groups of training samples;
determining hidden layers and weights, wherein a weight matrix between the input layer and the hidden layers is (v) 1 ,v 2 ,…,v m ) T The weight matrix between the hidden layer and the output layer is (w, w) 2 ,…,w m ) T Automatically adjusting the weight matrix after determining the number of hidden layers;
the Sigmoid function is selected as a transfer function, the quasi-newton method is adopted as a training function, and Trainbfg is selected as a training function. Then creating a BP network by using a feed forward net function;
and inputting the collected sample data into a model, calculating a learning and training sample through a function, and in the learning and training process, when the output weight Y value is adjusted to the output result Y according to the learning condition, stopping the learning and training and outputting the curvature mode predicted value of the steel structure when the confidence coefficient is more than 95%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310058445.4A CN116187177A (en) | 2023-01-18 | 2023-01-18 | Steel structure damage prediction method based on Beidou system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310058445.4A CN116187177A (en) | 2023-01-18 | 2023-01-18 | Steel structure damage prediction method based on Beidou system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116187177A true CN116187177A (en) | 2023-05-30 |
Family
ID=86433835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310058445.4A Pending CN116187177A (en) | 2023-01-18 | 2023-01-18 | Steel structure damage prediction method based on Beidou system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116187177A (en) |
-
2023
- 2023-01-18 CN CN202310058445.4A patent/CN116187177A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6454422B2 (en) | Damage information extraction apparatus, damage information extraction method, and damage information extraction program | |
| CN110807562B (en) | Regional bridge risk prediction method and system | |
| CN101504333B (en) | Progressive health monitoring method for cable structure median cord system based on strain monitoring | |
| JP6454790B2 (en) | Soundness determination device, soundness determination method, and soundness determination program | |
| CN210924883U (en) | Bridge structure health monitoring system | |
| KR102064328B1 (en) | Apparatus for providing earthquake damage prediction information of building and method thereof | |
| 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 | |
| KR102098888B1 (en) | System and method for disaster prediction and analysis of structures, and a recording medium having computer readable program for executing the method | |
| CN111625988A (en) | Bridge health management analysis and prediction system and method based on deep learning | |
| Bakhary et al. | Structure damage detection using neural network with multi-stage substructuring | |
| CN113378967B (en) | Structural health monitoring multivariate data anomaly diagnosis method based on convolutional neural network and transfer learning | |
| CN111783212B (en) | Typical damage identification method for cable-stayed bridge | |
| Tang et al. | Autoregressive model-based structural damage identification and localization using convolutional neural networks | |
| WO2024041233A1 (en) | Method for evaluating fatigue damage and life of bridge structure under multi-factor coupling action | |
| Dogan et al. | Detection of damages caused by earthquake and reinforcement corrosion in RC buildings with Deep Transfer Learning | |
| CN116029555A (en) | Bridge risk identification early warning system based on lightweight neural network and application method | |
| CN106446384B (en) | A kind of damnification recognition method of bridging crane main beam structure | |
| Li et al. | Long-term structural health monitoring for bridge based on back propagation neural network and long and short-term memory | |
| CN114722858A (en) | Safety assessment method for prestressed concrete structure | |
| CN113158362B (en) | Elevator residual life prediction method integrating physical failure and data driving | |
| Peil | Assessment of bridges via monitoring | |
| CN116187177A (en) | Steel structure damage prediction method based on Beidou system | |
| CN116591768A (en) | Tunnel monitoring method, system and device based on distributed network | |
| Kontoni et al. | Damage detection in reinforced concrete structures using advanced automatic systems: An overview | |
| Saju | Smart bridges based on bridge information modelling |
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 |