CN104504265A - Method for safety evaluation of monitoring information of in-service bridge - Google Patents
Method for safety evaluation of monitoring information of in-service bridge Download PDFInfo
- Publication number
- CN104504265A CN104504265A CN201410815239.4A CN201410815239A CN104504265A CN 104504265 A CN104504265 A CN 104504265A CN 201410815239 A CN201410815239 A CN 201410815239A CN 104504265 A CN104504265 A CN 104504265A
- Authority
- CN
- China
- Prior art keywords
- bridge
- state
- monitoring information
- network
- evolution
- 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.)
- Granted
Links
Landscapes
- Alarm Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for safety evaluation of monitoring information of an in-service bridge. The method comprises the following steps: analyzing the internal linkage between a structural response signal and state evolution of a bridge structure, excavating essential connotation of the monitoring information of the bridge, and constructing the time sequence of the monitoring information of the bridge; establishing an echo state network bridge monitoring information characteristic model as a bridge structure state characteristic system; constructing damage sensitive indexes of the bridge structure and obtaining critical parameters of damage evolution; constructing a bridge state evolution mode; performing safety evaluation on the bridge. According to the method disclosed by the invention, by integrating information science, data mining, mode identification and other disciplinary knowledge, intersection and integration of multiple disciplines are realized, beneficial explosion and technical support are provided for further improving the application quality of a bridge health monitoring system and widening the research direction of state evaluation of the bridge.
Description
Technical field
The invention belongs to bridge information monitoring field, be specifically related to a kind of new technology of the bridge security state estimation based on bridge structural health monitoring information.
Background technology
Along with the forties in 20th century computing machine appearance and the fifties artificial intelligence rise, people wish general-purpose computers to replace or expand the part brainwork of the mankind, pattern-recognition developed rapidly in early 1960s becomes a more complete ambit of ratio, it is also the intersection in multidisciplinary field, relates to statistics, fuzzy set theory, engineering science, artificial intelligence, computer science etc.Diagnosing structural damage is the continuation of pattern recognition theory in engineering field and application, its essence is " state recognition ", and namely solve the classification problem of same structure system different conditions, therefore the application of SHM statistical-simulation spectrometry recently been achieved more concern.
Due to civil structure bulky complex, residing bad environments, make sensor in actual acquisition process, inevitably be subject to the impact of the various interference of surrounding environment, and these interference often cause erroneous judgement to structural damage, hinder the application of damnification recognition method in Practical Project, so, be necessary that in non-destructive tests, introduce statistical theory carries out statistical study to non-destructive tests result, traditional to structural damage qualitative discrimination really with replacing the statistic discriminance of structural damage.Great majority research lays particular emphasis on the application of the statistical-simulation spectrometry based on SHM, these statistical models are by analytical characteristic vector monitoring abnormal data, comprising damage sensitive features value, service time, series model distinguished the Damage Evolution of structure with statistics method for detecting abnormality.The advantage of this method only needs to extract data from not damaged structure in the training stage exactly, and unlike supervised learning, wherein the data of training pattern will from can't harm and extracting damaged structure.Currently come, statistical pattern recognition method remains in certain defect and a difficult problem urgently to be resolved hurrily, is set forth principle and the method for diagnosing structural damage, become domestic and international study hotspot by the theoretical frame of pattern-recognition.
Summary of the invention
An object of the present invention is the circumscribed problem solving the assessment of traditional bridge safty, provides a kind of Corpus--based Method pattern recognition theory to utilize bridge structural health monitoring information to carry out the method for bridge security state estimation.
The invention provides a kind of method of servicing bridges monitoring information safety assessment, comprise the steps:
S1: the inner link that analytical structure response signal and bridge structural state develop, excavates the essential connotation of bridge monitoring information, builds bridge monitoring information time sequence;
S2: the bridge monitoring information time sequence built according to S1 and the bridge structural state evolution Feature comprised thereof set up echo state network bridge monitoring information characteristics model, in this, as bridge structural state feature architecture;
S3: the Bridge Structural Damage susceptibility index constructing the echo state network bridge monitoring information characteristics model set up based on S2, and obtain the critical parameters of Damage Evolution;
S4: the bridge state evolution pattern building the echo state network bridge monitoring information characteristics model set up based on S2;
S5: carry out bridge security assessment according to the bridge state evolution pattern that S4 builds.
Further, described step S2 specifically comprises: set up echo state network equation:
x(k+1)＝f(W
inu(k+1)+W
x(k)+W
fby(k))
Wherein W
in, W and W
fbbe respectively input weight matrix, inner connection weight value matrix and output feedack weight matrix, u (k) is input signal, the state that x (k) is echo state network, and y (k) is object vector;
Set up described echo state network model bridges girder construction monitoring information characteristic index construction step as follows:
1) parameter of deposit pond network is selected, network state dimension N, the inner degree of rarefication c connecting weights;
2) according to above-mentioned parameter stochastic production one dynamically deposit pond network;
3) this deposit pond network is carried out the identification of bridge monitoring infosystem, described echo state network model exports weights and is bridge monitoring information characteristics index, and by this characteristic index as structural state system.
Further, described step S3 specifically comprises: by based on echo state network model bridges girder construction monitoring information characteristic index, calculate n object mahalanobis distance d between any two
ij, then the mahalanobis distance between arbitrary object to all the other objects is sued for peace, definition D
sPRfor damage sensitive features index, namely
wherein d
ij=((x
i-x
j)
ts
-1(x
i-x
j))
1/2;
The Threshold Analysis of statistical model index under different structure state, namely obtains the critical parameters of structure generation Damage Evolution by statistical simulation methods.
Further, described step S4 specifically comprises: use the Outlier Detection method based on mahalanobis distance to pass judgment on the evolutionary pattern of the bridge structural state that described echo state network characterizes, under the framework of statistical-simulation spectrometry, described bridge structural state analysis is in completion system modeling, after feature architecture builds, set up discriminant function to the characteristic quantity extracted in described feature architecture to classify, to distinguish the further evolution-information of characteristic index system under normal condition and abnormality and abnormality, by checking that the principal character of a group objects determines isolated point, the differentiation of bridge structural state is analyzed by the bridge structure mahalanobis distance under different conditions and the departure degree under nondestructive state.
Advantageous Effects of the present invention is, statistical theory and recurrent neural network method is utilized to build bridge health monitoring information characteristics system, by the parameter that the method determination configuration state of emulation develops, the evolution of configuration state is differentiated by statistic pattern recognition theory, the engineering practice of significant increase bridge health monitoring system is worth, breach the limitation of traditional bridge safty assessment, statistic pattern recognition theory and the application in science of bridge building thereof are expanded, fuse information science, data mining, the multi-subject knowledges such as pattern-recognition, achieve multi-disciplinary intersection and fusion, for promoting bridge health monitoring system application quality and Bridge State Assessment research direction further, provide useful exploration and technical support.
Accompanying drawing explanation
Figure 1 shows that appraisal procedure process flow diagram of the present invention.
Embodiment
Hereafter will describe the specific embodiment of the invention in detail in conjunction with concrete accompanying drawing.It should be noted that the combination of technical characteristic or the technical characteristic described in following embodiment should not be considered to isolated, they can mutually be combined thus be reached better technique effect.In the accompanying drawing of following embodiment, the identical label that each accompanying drawing occurs represents identical feature or parts, can be applicable in different embodiment.
As shown in Figure 1, the invention provides a kind of method of servicing bridges monitoring information safety assessment, comprise the steps:
S1: the inner link that analytical structure response signal and bridge structural state develop, excavates the essential connotation of bridge monitoring information, builds bridge monitoring information time sequence;
S2: the bridge monitoring information time sequence built according to S1 and the bridge structural state evolution Feature comprised thereof set up echo state network bridge monitoring information characteristics model, in this, as bridge structural state feature architecture;
S3: the Bridge Structural Damage susceptibility index constructing the echo state network bridge monitoring information characteristics model set up based on S2, and obtain the critical parameters of Damage Evolution;
S4: the bridge state evolution pattern building the echo state network bridge monitoring information characteristics model set up based on S2;
S5: carry out bridge security assessment according to the bridge state evolution pattern that S4 builds.
Further, described step S2 specifically comprises: set up echo state network equation:
x(k+1)＝f(W
inu(k+1)+W
x(k)+W
fby(k))
Wherein W
in, W and W
fbbe respectively input weight matrix, inner connection weight value matrix and output feedack weight matrix, u (k) is input signal, the state that x (k) is echo state network, and y (k) is object vector;
Set up described echo state network model bridges girder construction monitoring information characteristic index construction step as follows:
1) parameter of deposit pond network is selected, network state dimension N, the inner degree of rarefication c connecting weights;
2) according to above-mentioned parameter stochastic production one dynamically deposit pond network;
3) this deposit pond network is carried out the identification of bridge monitoring infosystem, described echo state network model exports weights and is bridge monitoring information characteristics index, and by this characteristic index as structural state system.
Further, described step S3 specifically comprises: by based on echo state network model bridges girder construction monitoring information characteristic index, calculate n object mahalanobis distance d between any two
ij, then the mahalanobis distance between arbitrary object to all the other objects is sued for peace, definition D
sPRfor damage sensitive features index, namely
wherein d
ij=((x
i-x
j)
ts
-1(x
i-x
j))
1/2;
The Threshold Analysis of statistical model index under different structure state, namely obtains the critical parameters of structure generation Damage Evolution by statistical simulation methods.
Further, described step S4 specifically comprises: use the Outlier Detection method based on mahalanobis distance to pass judgment on the evolutionary pattern of the bridge structural state that described echo state network characterizes, under the framework of statistical-simulation spectrometry, described bridge structural state analysis is in completion system modeling, after feature architecture builds, set up discriminant function to the characteristic quantity extracted in described feature architecture to classify, to distinguish the further evolution-information of characteristic index system under normal condition and abnormality and abnormality, by checking that the principal character of a group objects determines isolated point, the differentiation of bridge structural state is analyzed by the bridge structure mahalanobis distance under different conditions and the departure degree under nondestructive state.
The present invention utilizes statistical theory and recurrent neural network method to build bridge health monitoring information characteristics system, by the parameter that the method determination configuration state of emulation develops, the evolution of configuration state is differentiated by statistic pattern recognition theory, the engineering practice of significant increase bridge health monitoring system is worth, breach the limitation of traditional bridge safty assessment, statistic pattern recognition theory and the application in science of bridge building thereof are expanded, fuse information science, data mining, the multi-subject knowledges such as pattern-recognition, achieve multi-disciplinary intersection and fusion, for promoting bridge health monitoring system application quality and Bridge State Assessment research direction further, provide useful exploration and technical support.
Although give some embodiments of the present invention, it will be understood by those of skill in the art that without departing from the spirit of the invention herein, can change embodiment herein.Above-described embodiment is exemplary, should using embodiment herein as the restriction of interest field of the present invention.
Claims (4)
1. a method for servicing bridges monitoring information safety assessment, is characterized in that, comprises the steps:
S1: the inner link that analytical structure response signal and bridge structural state develop, excavates the essential connotation of bridge monitoring information, builds bridge monitoring information time sequence;
S2: the bridge monitoring information time sequence built according to S1 and the bridge structural state evolution Feature comprised thereof set up echo state network bridge monitoring information characteristics model, in this, as bridge structural state feature architecture;
S3: the Bridge Structural Damage susceptibility index constructing the echo state network bridge monitoring information characteristics model set up based on described step S2, and obtain the critical parameters of Damage Evolution;
S4: the bridge state evolution pattern building the echo state network bridge monitoring information characteristics model set up based on described step S2;
S5: carry out bridge security assessment according to the bridge state evolution pattern that described step S4 builds.
2. the method for a kind of servicing bridges monitoring information safety assessment as claimed in claim 1, it is characterized in that, described step S2 specifically comprises: set up echo state network equation:
x(k+1)＝f(W
inu(k+1)+W
x(k)+W
fby(k))
Wherein W
in, W and W
fbbe respectively input weight matrix, inner connection weight value matrix and output feedack weight matrix, u (k) is input signal, the state that x (k) is echo state network, and y (k) is object vector;
Set up described echo state network model bridges girder construction monitoring information characteristic index construction step as follows:
1) parameter of deposit pond network is selected, network state dimension N, the inner degree of rarefication c connecting weights;
2) according to above-mentioned parameter stochastic production one dynamically deposit pond network;
3) this deposit pond network is carried out the identification of bridge monitoring infosystem, described echo state network model exports weights and is bridge monitoring information characteristics index, and by this characteristic index as structural state system.
3. the method for a kind of servicing bridges monitoring information safety assessment as claimed in claim 1, it is characterized in that, described step S3 specifically comprises:
By based on echo state network model bridges girder construction monitoring information characteristic index, calculate n object mahalanobis distance d between any two
ij, then the mahalanobis distance between arbitrary object to all the other objects is sued for peace, definition D
sPRfor damage sensitive features index, namely
wherein d
ij=((x
i-x
j)
ts
-1(x
i-x
j))
1/2;
The Threshold Analysis of statistical model index under different structure state, namely obtains the critical parameters of structure generation Damage Evolution by statistical simulation methods.
4. the method for a kind of servicing bridges monitoring information safety assessment as claimed in claim 1, it is characterized in that, described step S4 specifically comprises: use the Outlier Detection method based on mahalanobis distance to pass judgment on the evolutionary pattern of the bridge structural state that described echo state network characterizes, under the framework of statistical-simulation spectrometry, described bridge structural state analysis is in completion system modeling, after feature architecture builds, set up discriminant function to the characteristic quantity extracted in described feature architecture to classify, to distinguish the further evolution-information of characteristic index system under normal condition and abnormality and abnormality, by checking that the principal character of a group objects determines isolated point, the differentiation of bridge structural state is analyzed by the bridge structure mahalanobis distance under different conditions and the departure degree under nondestructive state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410815239.4A CN104504265B (en) | 2014-12-24 | 2014-12-24 | A kind of method of servicing bridges monitoring information security evaluation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410815239.4A CN104504265B (en) | 2014-12-24 | 2014-12-24 | A kind of method of servicing bridges monitoring information security evaluation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104504265A true CN104504265A (en) | 2015-04-08 |
| CN104504265B CN104504265B (en) | 2017-09-05 |
Family
ID=52945662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410815239.4A Expired - Fee Related CN104504265B (en) | 2014-12-24 | 2014-12-24 | A kind of method of servicing bridges monitoring information security evaluation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104504265B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133070A (en) * | 2017-09-19 | 2018-06-08 | 广州市建筑科学研究院有限公司 | A kind of appraisal procedure and system of the bridge health situation based on radial basis function neural network |
| CN108898292A (en) * | 2018-06-14 | 2018-11-27 | 合肥市城市生命线工程安全运行监测中心 | A kind of safety evaluation method of bridge health state |
| CN109614669A (en) * | 2018-11-23 | 2019-04-12 | 同济大学 | Net grade Bridge performance assessment prediction method |
| CN112818455A (en) * | 2021-02-22 | 2021-05-18 | 深圳市市政设计研究院有限公司 | Bridge structure response monitoring method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102565194A (en) * | 2012-02-09 | 2012-07-11 | 东南大学 | Method for carrying out early warning on damage to steel box girder of long span bridge in operation state |
| CN104200265A (en) * | 2014-07-28 | 2014-12-10 | 东北大学 | Improved bridge damage identification method based on neural network |
| CN104200005A (en) * | 2014-07-28 | 2014-12-10 | 东北大学 | Bridge damage identification method based on neural network |
-
2014
- 2014-12-24 CN CN201410815239.4A patent/CN104504265B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102565194A (en) * | 2012-02-09 | 2012-07-11 | 东南大学 | Method for carrying out early warning on damage to steel box girder of long span bridge in operation state |
| CN104200265A (en) * | 2014-07-28 | 2014-12-10 | 东北大学 | Improved bridge damage identification method based on neural network |
| CN104200005A (en) * | 2014-07-28 | 2014-12-10 | 东北大学 | Bridge damage identification method based on neural network |
Non-Patent Citations (5)
| Title |
|---|
| JIANXI YANG ET AL: ""PREDICTION OF BRIDGE MONITORING INFORMATION CHAOTIC USING TIME SERIES THEORY BY MULTI-STEP BP AND RBF NEURAL NETWORKS"", 《INTELLIGENT AUTOMATION AND SOFT COMPUTING》 * |
| YANG JIANXI ET AL: "《2009 World Congress on Computer Science and Information Engineering》", 2 April 2009 * |
| 李忠献 等: ""基于结构响应统计特征的神经网络损伤识别方法"", 《工程力学》 * |
| 范广露: ""基于回声状态网络的设备健康状态监测与预测方法"", 《中国优秀硕士学位论文全文数据库-工程科技II辑》 * |
| 魏锦辉: ""基于神经网络的桥梁损伤识别"", 《中国优秀硕士学位论文全文数据库-工程科技II辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133070A (en) * | 2017-09-19 | 2018-06-08 | 广州市建筑科学研究院有限公司 | A kind of appraisal procedure and system of the bridge health situation based on radial basis function neural network |
| CN108898292A (en) * | 2018-06-14 | 2018-11-27 | 合肥市城市生命线工程安全运行监测中心 | A kind of safety evaluation method of bridge health state |
| CN108898292B (en) * | 2018-06-14 | 2022-04-01 | 合肥泽众城市智能科技有限公司 | Safety assessment method for bridge health state |
| CN109614669A (en) * | 2018-11-23 | 2019-04-12 | 同济大学 | Net grade Bridge performance assessment prediction method |
| CN112818455A (en) * | 2021-02-22 | 2021-05-18 | 深圳市市政设计研究院有限公司 | Bridge structure response monitoring method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104504265B (en) | 2017-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Avci et al. | A review of vibration-based damage detection in civil structures: From traditional methods to Machine Learning and Deep Learning applications | |
| Eftekhar Azam et al. | Damage detection in structural systems utilizing artificial neural networks and proper orthogonal decomposition | |
| CN111274737A (en) | Method and system for predicting remaining service life of mechanical equipment | |
| CN107862331A (en) | It is a kind of based on time series and CNN unsafe acts recognition methods and system | |
| Guo et al. | Structural health monitoring by using a sparse coding-based deep learning algorithm with wireless sensor networks | |
| Yang et al. | A data-driven structural damage detection framework based on parallel convolutional neural network and bidirectional gated recurrent unit | |
| CN105716814B (en) | A kind of real-time monitoring system and its method for assessing truss structure damage | |
| Liu et al. | Data anomaly detection for structural health monitoring using a combination network of GANomaly and CNN | |
| CN113076975A (en) | Dam safety monitoring data anomaly detection method based on unsupervised learning | |
| CN111505424A (en) | Large experimental device power equipment fault diagnosis method based on deep convolutional neural network | |
| Jiang et al. | A hybrid data-fusion system using modal data and probabilistic neural network for damage detection | |
| Story et al. | A structural impairment detection system using competitive arrays of artificial neural networks | |
| CN106198749B (en) | A kind of data fusion method of the multiple sensor based on Metal Crack monitoring | |
| CN111724290B (en) | Environment-friendly equipment identification method and system based on depth layering fuzzy algorithm | |
| CN104504265A (en) | Method for safety evaluation of monitoring information of in-service bridge | |
| Bui-Tien et al. | Damage detection in structural health monitoring using hybrid convolution neural network and recurrent neural network | |
| Liao et al. | A channel-spatial-temporal attention-based network for vibration-based damage detection | |
| Khoshnoudian et al. | A new damage index using FRF data, 2D-PCA method and pattern recognition techniques | |
| Ye et al. | A deep learning-based method for automatic abnormal data detection: Case study for bridge structural health monitoring | |
| CN116911571A (en) | Mine operation and maintenance monitoring system | |
| CN111709393A (en) | Structural damage identification method combining convolution and cyclic neural network | |
| CN113312696B (en) | Bridge health condition dynamic prediction alarm method based on ARIMA algorithm | |
| Lin et al. | Optimal CNN-based semantic segmentation model of cutting slope images | |
| Che et al. | Few-shot structural repair decision of civil aircraft based on deep meta-learning | |
| CN108459933A (en) | A kind of big data computer system fault detection method based on depth Recursive Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170905 Termination date: 20181224 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |