CN115238365A - Tunnel post-disaster damage early warning method and system based on dynamic deep learning - Google Patents
Tunnel post-disaster damage early warning method and system based on dynamic deep learning Download PDFInfo
- Publication number
- CN115238365A CN115238365A CN202211091997.7A CN202211091997A CN115238365A CN 115238365 A CN115238365 A CN 115238365A CN 202211091997 A CN202211091997 A CN 202211091997A CN 115238365 A CN115238365 A CN 115238365A
- Authority
- CN
- China
- Prior art keywords
- tunnel
- disaster
- damage
- post
- deep 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides a tunnel post-disaster damage early warning method and system based on dynamic deep learning, which comprises the following steps: step 1: collecting the post-disaster information of the existing tunnel sample; step 2: training a softmax classifier for evaluating the post-disaster damage level of the tunnel; and step 3: collecting key index data, training and updating an LSTM deep learning model for predicting the development trend of the key indexes after the disaster in real time; and 4, step 4: predicting the development trend of key indexes after the disaster by adopting an LSTM model, and evaluating the damage level of a tunnel system at a future moment by adopting a softmax classifier according to the key index value and the tunnel structure design parameters at the future moment; and 5: if any key index exceeds a certain threshold or the tunnel system reaches a certain damage level, outputting corresponding alarm information and residual rescue time, otherwise, exiting. The method and the device can predict the damage development trend of the tunnel system after the disaster in advance and provide guidance for efficient rescue of the tunnel after the disaster.
Description
Technical Field
The invention relates to the field of railway engineering tunnel service safety, in particular to a tunnel post-disaster damage early warning method and system based on dynamic deep learning.
Background
When the tunnel passes through the movable fracture zone, the tunnel is affected by earthquake disasters and is easy to generate longitudinal deformation, lining deformation and other problems. In addition, the cracks of the rock body are rapidly increased under the action of earthquake force, a channel is provided for harmful gas accumulated in the stratum, and after the earthquake, the ventilation equipment is powered off, and other faults cause the harmful gas in the tunnel to be rapidly accumulated, so that secondary disasters are caused. The occurrence and development of the tunnel post-disaster problems seriously affect the railway operation safety, and cause great economic loss.
At present, the existing means can not predict the development of the tunnel post-disaster damage state, so that the optimal rescue opportunity is missed after the disaster, and the method becomes a problem to be solved urgently in the field of domestic railway engineering tunnel service safety, so that a tunnel post-disaster damage early warning method and system based on dynamic deep learning are urgently needed.
Disclosure of Invention
The technical problem solved by the invention is as follows: the method and the device solve the problems that the development trend of the post-disaster damage state of the tunnel cannot be predicted and accurate rescue after the tunnel disaster is difficult to be timely carried out in the prior art.
The technical solution of the invention is as follows: a tunnel post-disaster damage early warning method based on dynamic deep learning comprises the following steps:
step 1: and collecting the post-disaster information of the existing tunnel sample, wherein the information comprises tunnel structure design parameters and key indexes influencing tunnel risks after disasters.
And 2, step: setting multistage threshold values for key indexes, setting multiple damage levels for the damage degree of the post-disaster tunnel system, selecting the post-disaster information of the tunnel sample in the step 1, and classifying the post-disaster tunnel systems into corresponding damage levels through expert evaluation.
And 3, step 3: and (3) taking the post-disaster information of the tunnel sample and the corresponding damage level as a data set, and training a softmax classifier capable of determining the damage level based on the post-disaster information of the tunnel.
And 4, step 4: and installing a sensor for acquiring key indexes at a specific section position of the tunnel, acquiring data after the tunnel is subjected to a disaster, taking the acquired data as a data set, and training a dynamic LSTM deep learning model for predicting the future development trend of each key index after the disaster.
And 5: and (4) predicting the key index value at the future time after the disaster by adopting the dynamic LSTM deep learning model of each key index obtained in the step (4).
Step 6: and (5) transferring the key index value and the tunnel structure design parameter at the future moment predicted in the step (5) to a softmax classifier for damage level evaluation. When any key index exceeds a certain threshold or the damage of the tunnel system reaches a certain level, sending out the alarm information of the level and the residual rescue time reaching the level, making a rescue scheme according to the residual rescue time and taking corresponding rescue measures.
Further, the design parameters of the tunnel structure and the key indexes influencing the risk of the tunnel after the disaster in the step 1 are as follows:
the design parameters of the tunnel structure comprise the geometric dimension of the section of the tunnel, the thickness of a lining and the parameters of lining materials;
key indicators affecting tunnel risk after a disaster include tunnel settlement, tunnel section deformation, temperature, carbon dioxide concentration, methane concentration, hydrogen sulfide concentration, and nitrogen concentration.
Further, the multiple damage levels in step 2 include four levels: mild level, moderate level, severe level, damage level.
Further, the softmax classifier in the step 3 includes an input layer, a hidden layer and an output layer, the input parameters of the input layer are tunnel structure parameters before disaster and key indexes affecting the tunnel risk after disaster, the number of units of the hidden layer is determined by a grid search method, and the number of units of the output layer is four, which respectively represents four damage levels.
Further, the dynamic LSTM deep learning model for predicting the future development trend of each key index after the disaster is trained in step 4, and the process is as follows:
after a tunnel is subjected to disaster, high-frequency acquired data with equal time intervals of a key index i serve as a data set, data at n continuous moments serve as input parameters (n is less than the number of data in the data set), the data at the next moment serve as output parameters to train an LSTM deep learning model of the key index i, the acquired new data set is continuously expanded along with continuous increase of time, and therefore the LSTM deep learning model is continuously updated in real time through the new data set, and the dynamic LSTM deep learning model is obtained.
A tunnel post-disaster damage early warning system based on dynamic deep learning comprises an information acquisition module, a processing module and an early warning module;
the information acquisition module is used for acquiring post-disaster information of the existing tunnel sample and post-disaster information of the monitored tunnel;
the processing module is used for evaluating the damage level of the existing tunnel sample according to the information acquired by the information acquisition module and training a softmax model; training a dynamic LSTM deep learning model according to the post-disaster information of the monitored tunnel; loading and updating an LSTM model in real time, and predicting a key index value of the tunnel at a future moment after disaster; and loading a softmax model, and calculating the damage level of the tunnel system at a future moment.
The early warning module is used for judging whether the key indexes exceed a set threshold value and the damage level of the tunnel system, if any key index exceeds the threshold value or the tunnel system reaches a certain damage level, corresponding warning information and residual rescue time are output, a corresponding rescue scheme is formulated according to the residual rescue time, and rescue measures are taken.
Compared with the prior art, the invention has the advantages that:
according to the scheme provided by the embodiment of the invention, the dynamic LSTM deep learning model is adopted to predict the development trend of the key indexes influencing the tunnel risk after the disaster, and the softmax model for evaluating the damage level of the tunnel system after the disaster based on the tunnel structure design parameters and the key index numerical values at the future moment is constructed, so that the development trend of the damage state of the tunnel system after the disaster is predicted, the residual rescue time is determined, and accurate and effective measures are taken in time for tunnel rescue and rescue. The method provides a powerful means for predicting and evaluating the post-disaster damage degree of the tunnel system; the analysis method is clear and has strong reliability.
Drawings
FIG. 1 is a schematic flow chart of an early warning method according to an embodiment of the present invention
Fig. 2 is a frame diagram of an early warning system according to an embodiment of the present invention.
Detailed Description
Those skilled in the art will appreciate that those matters not described in detail in the present specification are not particularly limited to the specific examples described herein.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a tunnel post-disaster damage early warning method based on dynamic deep learning according to an embodiment of the present invention is shown, and includes the following steps:
s101, an acquisition module collects post-disaster information of an existing tunnel sample, wherein the information comprises tunnel structure design parameters and key indexes influencing tunnel risks after disasters, and the information comprises the following steps: the method comprises the following steps of tunnel section geometric dimension, lining thickness, lining material parameters, tunnel settlement, tunnel section deformation, temperature, carbon dioxide concentration, methane concentration, hydrogen sulfide concentration and nitrogen concentration.
S102, setting respective multi-level threshold values for each key index, and setting multi-damage levels for the damage degree of the tunnel system after the disaster, wherein the multi-damage levels comprise four levels: mild level, medium level, severe level, damage level. And selecting the post-disaster information of the tunnel samples in the S101, and classifying each post-disaster tunnel system into a corresponding damage level through expert evaluation. The method comprises the steps of training a softmax classifier which can determine damage levels based on tunnel post-disaster information by taking existing tunnel sample post-disaster information and corresponding damage levels as data sets, wherein the softmax classifier comprises an input layer, a hidden layer and an output layer, input parameters of the input layer are tunnel structural parameters before a disaster and key indexes which influence tunnel risks after the disaster, the number of units of the hidden layer is determined by a grid search method, and the number of units of the output layer is four and represents the four corresponding damage levels respectively.
S103, installing a sensor for acquiring key indexes at a specific section position of the tunnel, referring to a frame diagram of an early warning system shown in FIG. 2 after the tunnel is subjected to disaster, taking time-interval high-frequency acquired data of a key index i as a data set, taking data at n continuous moments as input parameters (n is the number of data in the data set), taking data at the next moment as output parameters to train an LSTM deep learning model of the key index i, and continuously increasing the time and continuously expanding the acquired new data set, so that the LSTM deep learning model is continuously updated in real time by the new data set, and a dynamic LSTM deep learning model is obtained. The LSTM deep learning model training process for the key index i is as follows:
at T m Time of day (subscript m is order, T) m The moment is the mth moment acquired at equal time intervals) to acquire T in the order of equal time intervals 1 、T 2 、T 3 …T m Data of a key index i of a time is used as a data set, and data (n) of n continuous times is input as a parameter<Number of data in data set), the output parameter is the data at the next moment, and T is obtained m LSTM deep learning model of key index i of time.
And S104, referring to the early warning system frame diagram shown in the step 2, predicting the future development trend of each key index after the disaster based on the data set collected in the step S103 and the dynamic LSTM deep learning model of each key index, wherein the prediction process is as follows:
at T m Time of day based on T m LSTM deep learning model of key index i of time, T m-n+1 、T m-n+2 、…、T m Data of a key index i acquired at any moment are used as input parameters to predict T m+1 The value of the key index at the moment;
by T m-n+2 、T m-n+3 、…、T m And predicted T m+1 Predicting T by using data of a key index i of time as an input parameter m+2 The numerical value of the key index i at the moment is predicted by analogy to obtain the preset numerical value of the key index i at the future moment;
and inputting the key index numerical value and the tunnel structure design parameter at the future moment into a softmax classifier, and evaluating the damage level of the tunnel system at the future moment.
And S105, referring to the early warning system frame diagram shown in FIG. 2, if any key index exceeds a threshold value or the tunnel system reaches a certain damage level, outputting corresponding warning information and residual rescue time, making a corresponding rescue scheme according to the residual rescue time, and taking emergency measures.
According to the scheme provided by the embodiment of the invention, the dynamic LSTM deep learning model is adopted to predict the development trend of the key indexes influencing the tunnel risk after the disaster, the softmax model is adopted to construct the relationship among the tunnel structure design parameters, the key indexes and the damage level of the tunnel system, the advanced prediction is carried out on the development trend of the tunnel damage state after the disaster, and the tunnel rescue and emergency rescue efficiency is improved, so that the further development of the tunnel damage is inhibited.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (6)
1. A tunnel post-disaster damage early warning method based on dynamic deep learning is characterized by comprising the following steps:
step 1: collecting post-disaster information of an existing tunnel sample, wherein the information comprises tunnel structure design parameters and key indexes influencing tunnel risks after disasters;
step 2: setting multistage threshold values for key indexes, setting multiple damage levels for the damage degree of the post-disaster tunnel system, selecting post-disaster information of the tunnel sample in the step 1, and classifying each post-disaster tunnel system into a corresponding damage level through expert evaluation;
and step 3: taking the post-disaster information of the tunnel sample and the corresponding damage level as a data set, and training a softmax classifier capable of determining the damage level based on the post-disaster information of the tunnel;
and 4, step 4: installing a sensor for acquiring key indexes at a specific section position of a tunnel, acquiring data after the tunnel is subjected to a disaster, taking the acquired data as a data set, and training a dynamic LSTM deep learning model for predicting the future development trend of each key index after the disaster;
and 5: predicting the key index value at the future time by adopting the dynamic LSTM deep learning model of each key index obtained in the step (4);
step 6: transferring the key index value and the tunnel structure design parameter at the future moment predicted in the step 5 to a softmax classifier for damage level evaluation; when any key index exceeds a certain threshold or the damage of the tunnel system reaches a certain level, sending out the alarm information of the level and the residual rescue time reaching the level, making a rescue scheme according to the residual rescue time and taking corresponding rescue measures.
2. The tunnel post-disaster damage early warning method based on dynamic deep learning of claim 1, wherein the tunnel structure design parameters and the key indexes affecting the tunnel risk after the disaster in the step 1 are as follows:
the design parameters of the tunnel structure comprise the geometric dimension of the section of the tunnel, the thickness of a lining and the parameters of lining materials;
key indicators affecting tunnel risk after a disaster include tunnel settlement, tunnel section deformation, temperature, carbon dioxide concentration, methane concentration, hydrogen sulfide concentration, and nitrogen concentration.
3. The tunnel post-disaster damage early warning method based on dynamic deep learning of claim 1, wherein the multiple damage levels in step 2 include four levels: mild level, moderate level, severe level, damage level.
4. The tunnel post-disaster damage early warning method based on dynamic deep learning as claimed in claim 1, wherein in step 3, the softmax classifier comprises an input layer, a hidden layer and an output layer, wherein input parameters of the input layer are tunnel structure parameters before disaster and key indexes affecting tunnel risk after disaster, the number of units of the hidden layer is determined by grid search method, and the number of units of the output layer is four, and represents four damage levels respectively.
5. The tunnel post-disaster damage early warning method based on dynamic deep learning of claim 1, wherein the dynamic LSTM deep learning model for predicting the future development trend of each key index after disaster is trained in the step 4, and the process is as follows:
after a tunnel is subjected to disaster, high-frequency acquired data with equal time intervals of a key index i serve as a data set, data at n continuous moments serve as input parameters (n is less than the number of data in the data set), the data at the next moment serve as output parameters to train an LSTM deep learning model of the key index i, the acquired new data set is continuously expanded along with continuous increase of time, and therefore the LSTM deep learning model is continuously updated in real time through the new data set, and the dynamic LSTM deep learning model is obtained.
6. The system for adopting any one of the tunnel post-disaster damage early warning methods based on dynamic deep learning of claim 1~5 is characterized by comprising an information acquisition module, a processing module and an early warning module;
the information acquisition module is used for acquiring post-disaster information of the existing tunnel sample and post-disaster information of the monitored tunnel;
the processing module is used for evaluating the damage level of the existing tunnel sample according to the information acquired by the information acquisition module and training a softmax model; training a dynamic LSTM deep learning model according to the post-disaster information of the monitored tunnel; loading and updating an LSTM model in real time, and predicting a key index value of the tunnel at a future moment after a disaster; loading a softmax model, and calculating the damage level of the tunnel system at a future moment;
the early warning module is used for judging whether the key indexes exceed a set threshold value and the damage level of the tunnel system, when any key index exceeds a certain threshold value or the damage of the tunnel system reaches a certain level, sending out alarm information of the level and the residual rescue time reaching the level, and making a rescue scheme and taking corresponding rescue measures according to the residual rescue time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211091997.7A CN115238365B (en) | 2022-09-07 | 2022-09-07 | Tunnel post-disaster damage early warning method and system based on dynamic deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211091997.7A CN115238365B (en) | 2022-09-07 | 2022-09-07 | Tunnel post-disaster damage early warning method and system based on dynamic deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115238365A true CN115238365A (en) | 2022-10-25 |
| CN115238365B CN115238365B (en) | 2022-12-16 |
Family
ID=83680647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211091997.7A Active CN115238365B (en) | 2022-09-07 | 2022-09-07 | Tunnel post-disaster damage early warning method and system based on dynamic deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115238365B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006318290A (en) * | 2005-05-13 | 2006-11-24 | Taisei Corp | Tunnel fire risk assessment system, method and program |
| CN108104822A (en) * | 2017-12-13 | 2018-06-01 | 中冶交通建设集团有限公司 | A kind of Tunnel Passing large size karst cavity construction geology disaster real-time predicting method |
| CN111460646A (en) * | 2020-03-27 | 2020-07-28 | 南京理工大学 | Subway station explosion intensity grading early warning method based on deep learning |
| CN113011645A (en) * | 2021-03-15 | 2021-06-22 | 国网河南省电力公司电力科学研究院 | Power grid strong wind disaster early warning method and device based on deep learning |
| CN214464318U (en) * | 2021-03-09 | 2021-10-22 | 彭烽 | Tunnel ventilation intelligence control system |
| JP2022013640A (en) * | 2020-07-01 | 2022-01-18 | 中国科学院地質與地球物理研究所 | Visualization system for simulating deformation of deep drift affected by groundwater freezing and thawing cycle |
| CN114417698A (en) * | 2021-12-10 | 2022-04-29 | 东北大学 | Rail transit external environment risk monitoring system and assessment method |
| CN114611193A (en) * | 2022-03-15 | 2022-06-10 | 浙江省交通运输科学研究院 | Tunnel mechanical property rapid inversion method based on deep learning |
| CN114741974A (en) * | 2022-05-12 | 2022-07-12 | 重庆交通大学 | Highway tunnel fire disaster growth period parameter identification and prediction method |
| CN114818363A (en) * | 2022-05-16 | 2022-07-29 | 中建三局绿色产业投资有限公司 | Urban deep drainage tunnel silting risk assessment and prediction early warning method |
-
2022
- 2022-09-07 CN CN202211091997.7A patent/CN115238365B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006318290A (en) * | 2005-05-13 | 2006-11-24 | Taisei Corp | Tunnel fire risk assessment system, method and program |
| CN108104822A (en) * | 2017-12-13 | 2018-06-01 | 中冶交通建设集团有限公司 | A kind of Tunnel Passing large size karst cavity construction geology disaster real-time predicting method |
| CN111460646A (en) * | 2020-03-27 | 2020-07-28 | 南京理工大学 | Subway station explosion intensity grading early warning method based on deep learning |
| JP2022013640A (en) * | 2020-07-01 | 2022-01-18 | 中国科学院地質與地球物理研究所 | Visualization system for simulating deformation of deep drift affected by groundwater freezing and thawing cycle |
| CN214464318U (en) * | 2021-03-09 | 2021-10-22 | 彭烽 | Tunnel ventilation intelligence control system |
| CN113011645A (en) * | 2021-03-15 | 2021-06-22 | 国网河南省电力公司电力科学研究院 | Power grid strong wind disaster early warning method and device based on deep learning |
| CN114417698A (en) * | 2021-12-10 | 2022-04-29 | 东北大学 | Rail transit external environment risk monitoring system and assessment method |
| CN114611193A (en) * | 2022-03-15 | 2022-06-10 | 浙江省交通运输科学研究院 | Tunnel mechanical property rapid inversion method based on deep learning |
| CN114741974A (en) * | 2022-05-12 | 2022-07-12 | 重庆交通大学 | Highway tunnel fire disaster growth period parameter identification and prediction method |
| CN114818363A (en) * | 2022-05-16 | 2022-07-29 | 中建三局绿色产业投资有限公司 | Urban deep drainage tunnel silting risk assessment and prediction early warning method |
Non-Patent Citations (5)
| Title |
|---|
| 乔怀玉: "隧道衬砌结构火灾损伤评定方法", 《公路隧道》 * |
| 安永林等: "可拓法评估隧道衬砌损伤与实证", 《采矿技术》 * |
| 安永林等: "基于FDS仿真火灾温度下隧道衬砌安全评估", 《灾害学》 * |
| 李清: "隧道火灾后结构物快速检测评估及处治设计分析", 《广东公路交通》 * |
| 林志等: "公路隧道衬砌混凝土火灾高温下的物理力学损伤规律", 《公路交通技术》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115238365B (en) | 2022-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11335179B1 (en) | Water environment risk prediction and early warning method | |
| Wang et al. | Real-time defect detection of high-speed train wheels by using Bayesian forecasting and dynamic model | |
| Zhu et al. | Dynamic prediction of traffic incident duration on urban expressways: A deep learning approach based on LSTM and MLP | |
| Giglioni et al. | Autoencoders for unsupervised real‐time bridge health assessment | |
| CN109035777B (en) | Traffic operation situation analysis method and system | |
| KR102564191B1 (en) | Disaster response system that detects and responds to disaster situations in real time | |
| JP5145501B2 (en) | Large building diagnostic system, large building diagnostic program, recording medium, and large building diagnostic method | |
| Liu et al. | Semantic modeling of bridge deterioration knowledge for supporting big bridge data analytics | |
| KR101685334B1 (en) | Disaster detection technique based on the key word relevance and the method for managing the disaster using the same | |
| Rosales et al. | Data driven innovations in structural health monitoring | |
| CN115238365B (en) | Tunnel post-disaster damage early warning method and system based on dynamic deep learning | |
| CN103852414B (en) | A kind of bridge corrosion monitoring and life-span prediction method | |
| Rahman et al. | Deep learning model for railroad structural health monitoring via distributed acoustic sensing | |
| CN113112115A (en) | Comprehensive current evaluation method for deep underground subway station | |
| CN111723970A (en) | Power operation hidden danger prediction method | |
| Alkam et al. | A robust method of the status monitoring of catenary poles installed along high-speed electrified train tracks | |
| CN115577309A (en) | Belt conveyor sudden fault early warning method and system | |
| CN114372522A (en) | Traffic accident duration influence factor analysis method based on RSF model | |
| CN116882302B (en) | Rock burst micro-macro precursor information identification method based on graph representation learning | |
| CN104462770B (en) | Signal reconstruction method taking sensor performance degradation into consideration | |
| Vagnoli et al. | A machine learning classifier for condition monitoring and damage detection of bridge infrastructure | |
| CN117345344B (en) | Intelligent prediction method and system for mine acoustic and electric signals | |
| Yin et al. | A neural network-based approach to predict the condition for sewer pipes | |
| Monavari | SHM-based structural deterioration assessment | |
| Monavari et al. | Deterioration sensitive feature using enhanced AR model residuals |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |