CN108090295A - A kind of long-span cablestayed bridges Damages in Stay Cables recognition methods - Google Patents
A kind of long-span cablestayed bridges Damages in Stay Cables recognition methods Download PDFInfo
- Publication number
- CN108090295A CN108090295A CN201711450716.1A CN201711450716A CN108090295A CN 108090295 A CN108090295 A CN 108090295A CN 201711450716 A CN201711450716 A CN 201711450716A CN 108090295 A CN108090295 A CN 108090295A
- Authority
- CN
- China
- Prior art keywords
- damages
- recognition methods
- long
- drag
- bridge
- 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
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01D—CONSTRUCTION OF BRIDGES, ELEVATED ROADWAYS OR VIADUCTS; ASSEMBLY OF BRIDGES
- E01D19/00—Structural or constructional details of bridges
- E01D19/16—Suspension cables; Cable clamps for suspension cables ; Pre- or post-stressed cables
Abstract
The invention belongs to bridge technology fields, disclose a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, including:Establish cable-stayed bridge finite element model;Simulated wind load is applied to the finite element model, obtains the wind speed time history curve of drag-line;The stress time-histories of wind action downhaul is calculated based on the wind speed time history curve;The stress time-histories and bridge cable are subjected to Lifting Wavelet bag decomposition without the stress time-histories hindered under state, extract wavelet packet signal component energy and identify suspension cable damage position using energy accumulation variation value as characteristic value;It is poor to obtain Lifting Wavelet bag component energy, and BP AdaBoost Model Identification structural damage degrees are established with this, and drag-line Fatigue Reliability is calculated based on structural damage degree.The present invention provides a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, possesses good adaptability and identification certainty.
Description
Technical field
The present invention relates to bridge technology field, more particularly to a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods.
Background technology
Modern Cable-Stayed Bridge Structure is mainly made of components such as Sarasota, girder, suspension cables, key of the suspension cable as cable-stayed bridge
Part has the characteristics that quality is small, flexibility is big, low resistance, heavily stressed, and a large amount of examples of runing show that suspension cable is used in full-bridge
Service life minimum and compared with cracky component.In the entire service phase of structure, component is chronically at the coupled vibrations under wind load effect
State, thus caused component fatigue damage problem can not ignore.
However, the identification to suspension cable damage position in the prior art, decomposition algorithm is relative complex, to the demand of memory
Measure larger, the speed of service is slow, causes to calculate the time also longer;In addition, its calculation error is larger, so as to tend not to realize signal
With the Accurate Reconstruction of image.Thus in the online health monitoring of the structure of large-scale Practical Project, calculating cycle is longer, and it is right
It cannot accurately be identified in the damage of Loads of Long-span Bridges suspension cable.When the degree of injury of suspension cable is identified, convergence rate
It is relatively slow, different local minimums is easily converged on, generalization ability is poor, thus with larger limitation.
The content of the invention
The present invention provides a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, solves non-destructive tests positioning accurate in the prior art
It spends low, damage extent identification over-fitting and easily the technical issues of the defects of local optimum occurs.
In order to solve the above technical problems, the present invention provides a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, including:
Establish cable-stayed bridge finite element model;
Simulated wind load is applied to the finite element model, obtains the wind speed time history curve of drag-line;
The stress time-histories of wind action downhaul is calculated based on the wind speed time history curve;
The stress time-histories and bridge cable are subjected to Lifting Wavelet bag decomposition without the stress time-histories hindered under state, extraction is small
Ripple bag signal component energy and using energy accumulation variation value as characteristic value, identifies suspension cable damage position;
It is poor to obtain Lifting Wavelet bag component energy, and BP-AdaBoost Model Identification structural damage degrees are established with this, and
Drag-line Fatigue Reliability is calculated based on structural damage degree.
Further, the Lifting Wavelet bag decomposition includes:
Wavelet Properties are improved by lift frame, the wavelet basis function with desired characteristic is constructed, becomes construction second
For the basic tool of small echo;
Original signal is decomposed using the wavelet packet analysis based on lift frame, and WAVELET PACKET DECOMPOSITION coefficient is carried out
Reconstruct, obtains the energy of i node of jth layer, i.e. wavelet packet component energy;
Wherein, when structure is damaged, the wavelet packet component energy will change.
Further, the BP-AdaBoost Model Identification structural damage degrees of establishing include:
Using wavelet packet component energy difference as training sample, 10 networks of Adaboost technique drills are utilized;
Training sample in initialization is normalized, sample value is made to be distributed in [0,1] section, and to training set
Each sample assigns identical weights;
Be distributed according to weights, randomly select the 1st network N N1 of sample training in training set, network structure for 8 × 6 ×
34 BP networks, learning algorithm are error backpropagation algorithm, and maximum study number is 20000 times, learning rate lr=
0.1, learning objective error sum of squares is Err_goal=4x10-4;
10 groups of data in training network are randomly selected as test sample, using BP-AdaBoost models to suspension cable
Damage be identified.
Further, the cable-stayed bridge finite element model of establishing includes:
The model of bridge is established using ANSYS softwares, beam4 units is selected to carry out girder simulation, shell63 simulation highways
Floorings select link10 unit simulation suspension cables.
Further, during drag-line is simulated, drag-line sag is considered using revised equivalent elastic modulus
It influences;It is modified using Ernst formula.
Further, wind load is simulated using harmony superposition, and using spectral factorization and trigonometrical number superposition come
Simulate random process sample;
With based on harmony superposition formed bridge simulation of wind formula be:
In formula, ωkFor double index frequencies, Δ ω=ωμp/ N, ωμpFor circular frequency, N counts for frequency sampling,For
Section [0,2 π] equally distributed independent phase angle.
Further, the wind speed time history curve bag that simulated wind load is applied to the finite element model, obtains drag-line
It includes:
In simulation process, some dummy activity point as wind load, king-tower edge are taken every a unit distance on girder
Vertical direction also sets a simulation point every a unit distance;
Then according to the design basic wind speed in definite bridge location, the design basis wind speed of girder, surface roughness;
Cut-off circular frequency 2 π rad/s, N=1024 are taken, the wind speed time history curve of corresponding simulation point is obtained using Matlab softwares.
The one or more technical solutions provided in the embodiment of the present application, have at least the following technical effects or advantages:
The long-span cablestayed bridges Damages in Stay Cables recognition methods provided in the embodiment of the present application, greatly across oblique pull under being acted on for wind load
A kind of method for Damages in Stay Cables identifying and diagnosing that bridge drag-line proposes.Specifically, using Lifting Wavelet bag analyzing and positioning and BP-
AdaBoost models carry out damage extent identification to drag-line.It is simple to promote parcel wave analysis decomposition algorithm, can realize to signal
With the Accurate Reconstruction of image, the speed of service is fast, and calculating speed is much higher than the decomposition of Traditional Wavelet bag, and can be shaken according to drag-line
Dynamic signal characteristic carries out adaptive analysis, realizes the quick positioning of high-precision.On this basis, BP-AdaBoost models are passed through
The defects of over-fitting characteristic of BP neural network had both been overcome to the recognition methods of suspension cable degree of injury and has been absorbed in local optimum,
The generalization ability of BP neural network is enhanced again, is enhanced the adaptability to the damage of different suspension cables, can be identified different oblique pulls
The degree of injury of rope.This method has preferable applicability and practicability to the damage diagnosis of Loads of Long-span Bridges drag-line.
Description of the drawings
Fig. 1 is long-span cablestayed bridges Damages in Stay Cables recognition methods principle process schematic diagram provided in an embodiment of the present invention;
Fig. 2 determines damage position principle schematic for Lifting Wavelet bag decomposition method provided in an embodiment of the present invention;
Fig. 3 is the BP-Adaboost modular concept schematic diagrames of Adaboost algorithm provided in an embodiment of the present invention.
Specific embodiment
The embodiment of the present application is damaged in the prior art by providing a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, solution
It identifies that positioning accuracy is low, damage extent identification over-fitting and easily the technical issues of local optimum defect occurs.
In order to better understand the above technical scheme, in conjunction with appended figures and specific embodiments to upper
It states technical solution to be described in detail, it should be understood that the specific features in the embodiment of the present invention and embodiment are to the application skill
The detailed description of art scheme rather than the restriction to technical scheme, in the case where there is no conflict, the embodiment of the present application
And the technical characteristic in embodiment can be mutually combined.
Referring to Fig. 1, a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, core thinking is mainly completed in two steps.
First, the damage position of Loads of Long-span Bridges drag-line is determined by promoting parcel wave analysis.Then, before and after to structural damage
Stress TIME HISTORY SIGNAL carry out Lifting Wavelet bag decomposition, using wavelet packet signal component energy accumulation variation value as characteristic value, with
This establishes BP-AdaBoost (Back Propagation neural network, Adaptive Boosting) model, utilizes
The degree of injury of long-span cablestayed bridges drag-line is identified in the method that AdaBoost algorithms and BP neural network are combined.So as to real
Now in high precision, while quick non-destructive tests, the adaptability to the oblique Damages in Stay Cables of varying environment is enhanced.
Specifically, the recognition methods includes:
Establish cable-stayed bridge finite element model;
Simulated wind load is applied to the finite element model, obtains the wind speed time history curve of drag-line;
The stress time-histories of wind action downhaul is calculated based on the wind speed time history curve;
The stress time-histories and bridge cable are subjected to Lifting Wavelet bag decomposition without the stress time-histories hindered under state, extraction is small
Ripple bag signal component energy simultaneously identifies suspension cable damage position using energy accumulation variation value as characteristic value;
It is poor to obtain Lifting Wavelet bag component energy, and BP-AdaBoost Model Identification structural damage degrees are established with this, and
Drag-line Fatigue Reliability is calculated based on structural damage degree.
It is described in detail below for above-mentioned various method steps.
Cable-stayed bridge finite element model is established, i.e., according to the practical structures of bridge, bridge finite element model is established, respectively to master
The structures such as beam, highway bridge, suspension cable are simulated.
It is operated in general, correlation model can be carried out by such as instruments such as ANSYS softwares and established with structural simulation.It selects
Beam4 units carry out girder simulation, and shell63 simulation highway bridge panels select link10 unit simulation suspension cables.
During drag-line is simulated, the influence of drag-line sag is considered using revised equivalent elastic modulus;Using
Ernst formula are modified it.
Simulated wind load is applied to the finite element model, obtains the wind speed time history curve of drag-line;Using harmony superposition
(WAWS) wind load is simulated, and random process sample is simulated using spectral factorization and trigonometrical number superposition.The base of utilization
It is in the bridge simulation of wind formula that harmony superposition is formed:
In formula, ωkFor double index frequencies, Δ ω=ωμp/ N, N count for frequency sampling,To be equal at section [0,2 π]
The independent phase angle of even distribution.
During fluctuating wind is described with statistical method, the power spectral density function of fluctuating wind speed embodies certain frequency
The energy size of fluctuating wind on domain, is most important statistical nature.And in each wind speed spectrum, Kaimal spectrums can preferably reflect
Wind spectrum density is with the variation of height, and therefore, in the simulation of fluctuating wind field, selection uses the power density functions of Kaimal spectrums.
In simulation process, some dummy activity as wind load is taken every a distance (about 20m-30m) on girder
Point.Equally, king-tower also sets a simulation point every a distance in the vertical direction.Then setting according to definite bridge location
Count basic wind speed V10, the design basis wind speed V of girderd, surface roughness Z0;Take cut-off circular frequency ωμp=2 π rad/s, N=
1024, Δ t=0.1s obtain the wind speed time history curve of corresponding simulation point using Matlab softwares.
The stress time-histories of wind action downhaul is calculated based on the wind speed time history curve;That is, according to acquisition
The Wind Velocity History of bridge simulation point is calculated influencing the wind load under bridge buffeting effect;Then wind action is obtained
The stress time-histories of lower part suspension cable.
The stress time-histories and bridge cable are subjected to Lifting Wavelet bag decomposition without the stress time-histories hindered under state, extraction is small
Ripple bag signal component energy and using energy accumulation variation value as characteristic value, identifies suspension cable damage position.
Specifically, Wavelet Properties are improved by lift frame first, construct the wavelet basis function with desired characteristic,
As the basic tool of construction Second Generation Wavelets.Then original signal is divided using the wavelet packet analysis based on lift frame
Solution, and WAVELET PACKET DECOMPOSITION coefficient is reconstructed, obtain the energy of i node of jth layer.
When structure is damaged, there can be large effect to parcel wavelength-division energy.That is, before and after damage, it is small
Ripple bag component energy can generate variation, thus the stress time-histories and bridge cable are carried without the stress time-histories hindered under state
WAVELET PACKET DECOMPOSITION is risen, extracts wavelet packet signal component energy and using energy accumulation variation value as characteristic value, identification suspension cable damage
Hinder position.
Referring to Fig. 2, cross promotion parcel wave analysis and determine that the damage position of drag-line is divided into two steps:Lift frame and using energy as
Element construction feature vector.
Specific step is as follows:
1. the wavelet packet analysis of lift frame decomposes the original signal for vibrating stayed structure, and to WAVELET PACKET DECOMPOSITION
Coefficient is reconstructed, and the energy for obtaining i node of jth layer is:
Wherein it is the reconstruction signal of j layers of i node.
2. when drag-line damages, there can be larger impact to wavelet packet component energy, therefore be constructed by element of energy
Feature vector T:
T=[E (j, i)], i=1,2 ..., 2j
Definition:
T '=[E ' (j, i)]-[E " (j, i)], i=1,2 ..., 2j
Wherein, E ' is the Lifting Wavelet bag component energy for having damaged structure, and E " is the Lifting Wavelet bag component of intact structure
Energy.Feature vector T ' is poor for Lifting Wavelet bag component energy, can be as the subset of BP-AdaBoost models, so as to establish
BP-AdaBoost models carry out the non-destructive tests of structure.
After completing damage reason location, acquisition Lifting Wavelet bag component energy is poor, and establishes BP-AdaBoost Model Identifications with this
Structural damage degree, and drag-line Fatigue Reliability is calculated based on structural damage degree.Specifically:
Using wavelet packet component energy difference as training sample, 10 networks of Adaboost technique drills are utilized;
Training sample in initialization is normalized, sample value is made to be distributed in [0,1] section, and to training set
Each sample assigns identical weights;
Be distributed according to weights, randomly select the 1st network N N1 of sample training in training set, network structure for 8 × 6 ×
34 BP networks, learning algorithm are error backpropagation algorithm, and maximum study number is 20000 times, learning rate lr=
0.1, learning objective error sum of squares is Err_goal=4x10-4;
10 groups of data in training network are randomly selected as test sample, using BP-AdaBoost models to suspension cable
Damage be identified.
Referring to Fig. 3, specifically, the realization step of the BP neural network prediction model of Adaboost algorithm is:
1. it inputs:Give N number of training sample:S={ (x1,y1),…(xi,yi),(xn,yn), wherein class label yi∈ Y=
{ 1 ..., k }, i are specimen number, and k represents classification number.
2. it is distributed as 1/I on initialization training set;
3. using the major cycle process (t=1,2 ..., T) of AdaBoost method calls ANN, training set is trained,
And obtain estimated sequence ht;Distribution on training set is updated according to training result, and is trained according to new sample distribution;
4. after T Xun Huan, final classification device H is obtained:
The present embodiment determines the damage position of Loads of Long-span Bridges drag-line by promoting parcel wave analysis.Then, to structural damage
Front and rear stress TIME HISTORY SIGNAL carries out Lifting Wavelet bag decomposition, and wavelet packet signal component energy is accumulated variation value as feature
Value, BP-AdaBoost (Back Propagation neural network, Adaptive Boosting) mould is established with this
Type knows the degree of injury of long-span cablestayed bridges drag-line using the method that AdaBoost algorithms and BP neural network are combined
Not.
The method for promoting parcel wave analysis is that the original signal of inhaul cable vibration is decomposed to obtain low frequency signal and high-frequency signal,
And WAVELET PACKET DECOMPOSITION coefficient is reconstructed, obtain the energy of node.Then, using energy as element construction feature vector, to carry
The variation of wavelet packet component energy difference is risen to judge the damage position of drag-line.The decomposition algorithm of this method is simple, accelerates computing
Speed reduces the calculating time of suspension cable non-destructive tests.
BP-AdaBoost models construct the weak fallout predictors of multiclass BP, then by choosing different parameters to BP neural network
Use strong fallout predictor of the Adaboost algorithm by obtained multiple weak fallout predictor compositions newly.BP-AdaBoost models enhance BP
The generalization ability of neutral net enhances the adaptability to the oblique Damages in Stay Cables of varying environment.
The one or more technical solutions provided in the embodiment of the present application, have at least the following technical effects or advantages:
The long-span cablestayed bridges Damages in Stay Cables recognition methods provided in the embodiment of the present application, greatly across oblique pull under being acted on for wind load
A kind of method for Damages in Stay Cables identifying and diagnosing that bridge drag-line proposes.Specifically, using Lifting Wavelet bag analyzing and positioning and BP-
AdaBoost models carry out damage extent identification to drag-line.It is simple to promote parcel wave analysis decomposition algorithm, can realize to signal
With the Accurate Reconstruction of image, the speed of service is fast, and calculating speed is much higher than the decomposition of Traditional Wavelet bag, and can be shaken according to drag-line
Dynamic signal characteristic carries out adaptive analysis, realizes the quick positioning of high-precision.On this basis, BP-AdaBoost models are passed through
The defects of over-fitting characteristic of BP neural network had both been overcome to the recognition methods of suspension cable degree of injury and has been absorbed in local optimum,
The generalization ability of BP neural network is enhanced again, is enhanced the adaptability to the damage of different suspension cables, can be identified different oblique pulls
The degree of injury of rope.This method has preferable applicability and practicability to the damage diagnosis of Loads of Long-span Bridges drag-line.
It should be noted last that more than specific embodiment is merely illustrative of the technical solution of the present invention and unrestricted,
Although the present invention is described in detail with reference to example, it will be understood by those of ordinary skill in the art that, it can be to the present invention
Technical solution be modified or replaced equivalently, without departing from the spirit and scope of technical solution of the present invention, should all cover
Among scope of the presently claimed invention.
Claims (7)
1. a kind of long-span cablestayed bridges Damages in Stay Cables recognition methods, which is characterized in that including:
Establish cable-stayed bridge finite element model;
Simulated wind load is applied to the finite element model, obtains the wind speed time history curve of drag-line;
The stress time-histories of wind action downhaul is calculated based on the wind speed time history curve;
The stress time-histories and bridge cable are subjected to Lifting Wavelet bag decomposition without the stress time-histories hindered under state, extract wavelet packet
Signal component energy simultaneously identifies suspension cable damage position using energy accumulation variation value as characteristic value;
It is poor to obtain Lifting Wavelet bag component energy, and BP-AdaBoost Model Identification structural damage degrees are established with this, and is based on
Structural damage degree calculates drag-line Fatigue Reliability.
2. long-span cablestayed bridges Damages in Stay Cables recognition methods as described in claim 1, which is characterized in that the Lifting Wavelet bag point
Solution includes:
Wavelet Properties are improved by lift frame, construct the wavelet basis function with desired characteristic, it is small to become the construction second generation
The basic tool of ripple;
Original signal is decomposed using the wavelet packet analysis based on lift frame, and weight is carried out to WAVELET PACKET DECOMPOSITION coefficient
Structure obtains the energy of i node of jth layer, i.e. wavelet packet component energy;
Wherein, when structure is damaged, the wavelet packet component energy will change.
3. long-span cablestayed bridges Damages in Stay Cables recognition methods as described in claim 1, which is characterized in that described to establish BP-
AdaBoost Model Identification structural damage degrees include:
Using wavelet packet component energy difference as training sample, 10 networks of Adaboost technique drills are utilized;
Training sample in initialization is normalized, sample value is made to be distributed in [0,1] section, and to training set various kinds
The identical weights of this tax;
It is distributed according to weights, the 1st network N N1 of sample training is randomly selected in training set, network structure is 8 × 6 × 34
BP networks, learning algorithm are error backpropagation algorithm, and maximum study number is 20000 times, learning rate lr=0.1, is learned
It is Err_goal=4x10-4 to practise target error quadratic sum;
10 groups of data in training network are randomly selected as test sample, using BP-AdaBoost models to the damage of suspension cable
Wound is identified.
4. long-span cablestayed bridges Damages in Stay Cables recognition methods as described in claim 1, which is characterized in that the cable-stayed bridge of establishing has
Limit meta-model includes:
The model of bridge is established using ANSYS softwares, beam4 units is selected to carry out girder simulation, shell63 simulation highway bridges
Plate selects link10 unit simulation suspension cables.
5. long-span cablestayed bridges Damages in Stay Cables recognition methods as claimed in claim 4, which is characterized in that in the process of simulation drag-line
In, the influence of drag-line sag is considered using revised equivalent elastic modulus;It is modified using Ernst formula.
6. long-span cablestayed bridges Damages in Stay Cables recognition methods as claimed in claim 5, which is characterized in that using harmony superposition pair
Wind load is simulated, and simulates random process sample using spectral factorization and trigonometrical number superposition;
With based on harmony superposition formed bridge simulation of wind formula be:
In formula, ωkFor double index frequencies,ωμpFor circular frequency, N counts for frequency sampling,For in section [0,2
π] equally distributed independent phase angle.
7. long-span cablestayed bridges Damages in Stay Cables recognition methods as claimed in claim 6, which is characterized in that described to the finite element
Model applies simulated wind load, and obtaining the wind speed time history curve of drag-line includes:
In simulation process, some dummy activity point as wind load is taken every a unit distance on girder, king-tower is vertically
Direction also sets a simulation point every a unit distance;
Then according to the design basic wind speed V in definite bridge locationo, the design basis wind speed V of girderd, surface roughness
Zo;Cut-off circular frequency is taken, N=1024 obtains the wind speed time history curve of corresponding simulation point using Matlab softwares.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711450716.1A CN108090295A (en) | 2017-12-27 | 2017-12-27 | A kind of long-span cablestayed bridges Damages in Stay Cables recognition methods |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711450716.1A CN108090295A (en) | 2017-12-27 | 2017-12-27 | A kind of long-span cablestayed bridges Damages in Stay Cables recognition methods |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108090295A true CN108090295A (en) | 2018-05-29 |
Family
ID=62180403
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711450716.1A Pending CN108090295A (en) | 2017-12-27 | 2017-12-27 | A kind of long-span cablestayed bridges Damages in Stay Cables recognition methods |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108090295A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109682460A (en) * | 2018-09-19 | 2019-04-26 | 广西交通科学研究院有限公司 | Drag-line fracture of wire recognition methods based on vibration signal |
CN109975136A (en) * | 2019-04-09 | 2019-07-05 | 西南交通大学 | A kind of steel-frame structure damnification recognition method based on wavelet packet analysis |
CN110046379A (en) * | 2019-03-04 | 2019-07-23 | 浙江大学 | A kind of structure entirety damnification recognition method based on space-frequency information |
CN110231403A (en) * | 2019-06-10 | 2019-09-13 | 中设设计集团股份有限公司 | Long-span bridge online real-time damage identification method supported by mobile load response wavelet packet analysis |
CN110321616A (en) * | 2019-06-26 | 2019-10-11 | 国网辽宁省电力有限公司葫芦岛供电公司 | The three crossing pole tower method for diagnosing faults based on wavelet transformation and attention model |
CN111175068A (en) * | 2020-02-24 | 2020-05-19 | 长沙理工大学 | Device and method for typical damage simulation of cable-stayed bridge |
CN110728089B (en) * | 2019-09-29 | 2020-05-29 | 济南城建集团有限公司 | Large-span bridge stay cable structure damage diagnosis method based on BOTDA technology |
CN111639450A (en) * | 2020-06-03 | 2020-09-08 | 太原科技大学 | Tower crane damage identification method based on support vector machine |
CN111783212A (en) * | 2020-07-09 | 2020-10-16 | 长沙理工大学 | Typical damage identification method for cable-stayed bridge |
CN111856058A (en) * | 2019-09-09 | 2020-10-30 | 东南大学 | Anemograph support capable of being fixed on inhaul cable |
CN112001110A (en) * | 2019-05-27 | 2020-11-27 | 浙江大学 | Structural damage identification monitoring method based on vibration signal space real-time recursive graph convolutional neural network |
CN112529842A (en) * | 2020-11-23 | 2021-03-19 | 天津大学 | Multi-excitation fusion plate structure damage identification method based on wavelet packet energy |
CN113312834A (en) * | 2021-04-19 | 2021-08-27 | 桂林理工大学 | Method for picking up abnormal cable force of stayed cable based on convolutional neural network |
CN114239364A (en) * | 2021-12-20 | 2022-03-25 | 石家庄铁道大学 | Cable-stayed bridge damage identification method and device based on improved wavelet packet energy curvature |
CN114969934A (en) * | 2022-05-31 | 2022-08-30 | 湖南工商大学 | Stay cable damage degree identification method and model construction method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013188341A2 (en) * | 2012-06-11 | 2013-12-19 | Parsons Corporation | Redundancy truss for cable-stayed and other bridge types |
DE102013007635A1 (en) * | 2013-05-21 | 2014-11-27 | Peter Wagner | Cable-stayed bridge with hybrid stiffening beam |
-
2017
- 2017-12-27 CN CN201711450716.1A patent/CN108090295A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013188341A2 (en) * | 2012-06-11 | 2013-12-19 | Parsons Corporation | Redundancy truss for cable-stayed and other bridge types |
DE102013007635A1 (en) * | 2013-05-21 | 2014-11-27 | Peter Wagner | Cable-stayed bridge with hybrid stiffening beam |
Non-Patent Citations (5)
Title |
---|
刘文会等: "斜拉桥的脉动风荷载数值模拟", 《吉林建筑工程学院学报》 * |
查大奎等: "风荷载作用下斜拉桥拉索疲劳可靠性分析", 《建材世界》 * |
谭冬梅等: "大跨斜拉桥拉索断裂的损伤诊断研究", 《武汉理工大学学报》 * |
谭冬梅等: "提升小波包和BP-AdaBoost在大跨斜拉桥拉索损伤识别中的应用", 《噪声与振动控制》 * |
谭冬梅等: "用于大跨斜拉桥拉索疲劳可靠度分析荷载的研究", 《武汉理工大学学报》 * |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109682460A (en) * | 2018-09-19 | 2019-04-26 | 广西交通科学研究院有限公司 | Drag-line fracture of wire recognition methods based on vibration signal |
CN110046379A (en) * | 2019-03-04 | 2019-07-23 | 浙江大学 | A kind of structure entirety damnification recognition method based on space-frequency information |
CN110046379B (en) * | 2019-03-04 | 2021-01-19 | 浙江大学 | Structure overall damage identification method based on space-frequency information |
CN109975136A (en) * | 2019-04-09 | 2019-07-05 | 西南交通大学 | A kind of steel-frame structure damnification recognition method based on wavelet packet analysis |
CN109975136B (en) * | 2019-04-09 | 2021-06-25 | 西南交通大学 | Steel frame structure damage identification method based on wavelet packet analysis |
CN112001110A (en) * | 2019-05-27 | 2020-11-27 | 浙江大学 | Structural damage identification monitoring method based on vibration signal space real-time recursive graph convolutional neural network |
CN112001110B (en) * | 2019-05-27 | 2023-04-28 | 浙江大学 | Structural damage identification monitoring method based on vibration signal space real-time recurrent graph convolutional neural network |
CN110231403A (en) * | 2019-06-10 | 2019-09-13 | 中设设计集团股份有限公司 | Long-span bridge online real-time damage identification method supported by mobile load response wavelet packet analysis |
CN110321616A (en) * | 2019-06-26 | 2019-10-11 | 国网辽宁省电力有限公司葫芦岛供电公司 | The three crossing pole tower method for diagnosing faults based on wavelet transformation and attention model |
CN110321616B (en) * | 2019-06-26 | 2022-11-25 | 国网辽宁省电力有限公司葫芦岛供电公司 | Three-span tower fault diagnosis method based on wavelet transformation and attention model |
CN111856058A (en) * | 2019-09-09 | 2020-10-30 | 东南大学 | Anemograph support capable of being fixed on inhaul cable |
CN111856058B (en) * | 2019-09-09 | 2022-04-29 | 东南大学 | Anemograph support capable of being fixed on inhaul cable |
CN110728089B (en) * | 2019-09-29 | 2020-05-29 | 济南城建集团有限公司 | Large-span bridge stay cable structure damage diagnosis method based on BOTDA technology |
CN111175068A (en) * | 2020-02-24 | 2020-05-19 | 长沙理工大学 | Device and method for typical damage simulation of cable-stayed bridge |
CN111639450A (en) * | 2020-06-03 | 2020-09-08 | 太原科技大学 | Tower crane damage identification method based on support vector machine |
CN111783212B (en) * | 2020-07-09 | 2022-07-12 | 长沙理工大学 | Typical damage identification method for cable-stayed bridge |
CN111783212A (en) * | 2020-07-09 | 2020-10-16 | 长沙理工大学 | Typical damage identification method for cable-stayed bridge |
CN112529842A (en) * | 2020-11-23 | 2021-03-19 | 天津大学 | Multi-excitation fusion plate structure damage identification method based on wavelet packet energy |
CN112529842B (en) * | 2020-11-23 | 2023-02-07 | 天津大学 | Multi-excitation fusion plate structure damage identification method based on wavelet packet energy |
CN113312834B (en) * | 2021-04-19 | 2022-04-26 | 桂林理工大学 | Method for picking up abnormal cable force of stayed cable based on convolutional neural network |
CN113312834A (en) * | 2021-04-19 | 2021-08-27 | 桂林理工大学 | Method for picking up abnormal cable force of stayed cable based on convolutional neural network |
CN114239364A (en) * | 2021-12-20 | 2022-03-25 | 石家庄铁道大学 | Cable-stayed bridge damage identification method and device based on improved wavelet packet energy curvature |
CN114239364B (en) * | 2021-12-20 | 2023-09-12 | 石家庄铁道大学 | Cable-stayed bridge damage identification method and device based on improved wavelet packet energy curvature |
CN114969934A (en) * | 2022-05-31 | 2022-08-30 | 湖南工商大学 | Stay cable damage degree identification method and model construction method |
CN114969934B (en) * | 2022-05-31 | 2023-09-05 | 湖南工商大学 | Stay cable damage degree identification method and model construction method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108090295A (en) | A kind of long-span cablestayed bridges Damages in Stay Cables recognition methods | |
CN105973594B (en) | A kind of rolling bearing fault Forecasting Methodology based on continuous depth confidence network | |
CN108038300B (en) | Optical fiber state evaluation method based on improved membership function combined with neural network | |
Yu et al. | Structural health monitoring based on continuous ACO method | |
CN104091074B (en) | A kind of MEDIUM OR LONG RANGE HYDROLOGIC FORECAST METHOD based on empirical mode decomposition | |
CN101404071B (en) | Electronic circuit fault diagnosis neural network method based on grouping particle swarm algorithm | |
CN107085763A (en) | A kind of driving motor for electric automobile system performance evaluation method | |
CN109242149A (en) | A kind of student performance early warning method and system excavated based on educational data | |
US11709979B1 (en) | Bridge damage identification method considering uncertainty | |
CN104376214A (en) | Fluctuating wind velocity simulation method based on data driving | |
CN105678343A (en) | Adaptive-weighted-group-sparse-representation-based diagnosis method for noise abnormity of hydroelectric generating set | |
CN115577436B (en) | Combined deep learning method for solving wind-induced vibration response of uncertain structure | |
CN111523778A (en) | Power grid operation safety assessment method based on particle swarm algorithm and gradient lifting tree | |
CN111368863A (en) | Intelligent health state assessment method for key mechanism of shore bridge | |
CN111783212B (en) | Typical damage identification method for cable-stayed bridge | |
CN107295537A (en) | A kind of method and system for wireless sensor network reliability of testing and assessing | |
CN102034111A (en) | Method for identifying and detecting aircraft structural damage conditions in diversified way | |
CN115587309A (en) | Method, device and equipment for extracting key features of short-circuit resistance of transformer | |
Li et al. | Deep learning models for time-history prediction of vehicle-induced bridge responses: A comparative study | |
Lakshmi | Detection and quantification of damage in bridges using a hybrid algorithm with spatial filters under environmental and operational variability | |
CN103984856A (en) | System and method for predicting longitudinal crack width of wall body based on neural network | |
CN110147827A (en) | A kind of failure prediction method based on IAALO-SVM and similarity measurement | |
CN108805419A (en) | Grid nodes importance calculation method based on internet startup disk and support vector regression | |
Akintunde et al. | Singular value decomposition and unsupervised machine learning for virtual strain sensing: Application to an operational railway bridge | |
Kovalchuk et al. | Towards ensemble simulation of complex systems |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180529 |