CN109187770A - A kind of anchor pole AI detection method - Google Patents
A kind of anchor pole AI detection method Download PDFInfo
- Publication number
- CN109187770A CN109187770A CN201811238914.6A CN201811238914A CN109187770A CN 109187770 A CN109187770 A CN 109187770A CN 201811238914 A CN201811238914 A CN 201811238914A CN 109187770 A CN109187770 A CN 109187770A
- Authority
- CN
- China
- Prior art keywords
- model
- data
- detection
- anchor pole
- processing step
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
- G01N29/4409—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison
- G01N29/4418—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison with a model, e.g. best-fit, regression analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
- G01N29/4481—Neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/02—Indexing codes associated with the analysed material
- G01N2291/028—Material parameters
- G01N2291/02818—Density, viscosity
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/02—Indexing codes associated with the analysed material
- G01N2291/028—Material parameters
- G01N2291/02854—Length, thickness
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a kind of anchor pole AI detection methods, including data collection steps and data processing step, the data processing step is the parsing collected data of data collection steps to obtain testing result, the data processing step are as follows: will test collection in worksite to data be sent to server, the established AI model of server calls automatically parses detection data, Grouted density is determined by disaggregated model, rock-bolt length is predicted using regression model, AI parsing result is finally returned into scene in real time.The detection method can effectively ensure that the objectivity of testing result.
Description
Technical field
The present invention relates to impact elasticity wave analysis technical fields, more particularly to a kind of anchor pole AI detection method.
Background technique
The anchor pole of the prior art is widely used in railway, tunnel, in slope construction, and considers the construction quality of anchor pole,
After completing construction, generally requires and the inspection such as rock-bolt length, grouting quality is carried out to anchor pole.
Impact elasticity wave method is the important method that the prior art carries out complete evaluation for Detection of Bolt Bonding Integrity: rock-bolt length
Existing with the detection method of compactness is impact elasticity wave method at present, carries out exciting using exciting device in end, recycles and receive
Device receives elastic wave signal, calculates rock-bolt length by identification bar bottom reflection interval, is in the milk by the decay calculation of elastic wave
Compactness, to obtain corresponding testing result.
Existing anchor pole detection method is advanced optimized so that its it is more enough preferably serve construction engineering test,
It is the important technical research of those skilled in the art or seeks direction.
Summary of the invention
Existing anchor pole detection method is advanced optimized for set forth above, so that it more enough is preferably served
Construction engineering test, the problem of being the important technical research of those skilled in the art or seek direction, the present invention provides one kind
Anchor pole AI detection method, the detection method can effectively ensure that the objectivity of testing result.
The technological means of this programme is as follows, a kind of anchor pole AI detection method, including data collection steps and data processing step
Suddenly, the data processing step is the parsing collected data of data collection steps to obtain testing result, the data processing
Step are as follows: will test collection in worksite to data be sent to server, the established AI model of server calls is to detection data
It is automatically parsed, Grouted density is determined by disaggregated model, rock-bolt length is predicted using regression model,
AI parsing result is finally returned into scene in real time.
In the prior art, anchor pole it is actually detected in, pass through the energy in the reflection interval of impact elasticity wave and propagation
Length and Grouted density to determine anchor pole is lost in amount, and traditional detection process and disadvantage are as follows:
1, personnel carry out data acquisition to scene need to take back indoor parsing by data, lead since resolving is more complicated
Cause detection cycle long;
2, since detection data takes back indoor parsing, so there is the risk artificially modified in detection data;
3, detection data needs artificial parsing, relatively high to the professional ability requirement of testing staff, and testing staff is needed to have
Detection experience abundant;
4, the examined personnel of accuracy in detection parse experience and influence.
In the present solution, above data acquisition step can obtain data processing step based on impact elasticity wave method needed for number
According to, it then is sent to the server by way of wireless transmission or wire transmission, in the server, the data processing step
It suddenly include that AI automatically parses step and AI judgement calculating step, it is described to automatically parse step are as follows: the data that server will receive
It is parsed, parses the characteristic parameter needed for AI model determines;
The AI determines to calculate step are as follows: the well-established AI model of server calls carries out the result after automatically parsing
It is automatic to determine to obtain result.
As those skilled in the art, the AI is artificial intelligence, the AI model in the detection process, AI model one
Being set up permanently to use, and can automatically derive as a result, simultaneously in the detection process, also can be by persistently reconstructing training set completeness
The more AI models of mode carry out retraining, AI solution to model analysis accuracy rate, precision etc. is continuously improved.
It is mainly had the advantage that compared with the detection of existing anchor pole
1, parsing result is obtained in real time, shortens detection cycle;
2, detection process is simplified, detection difficulty is reduced;
3, the risk of human intervention detection data, the objectivity of the testing result of guarantee have been evaded;
4, detection accuracy is improved, and with the complete sustainable raising detection accuracy of training set;
5, it is directed to on-site data gathering, the requirement to related personnel is only that can get accumulation signal and transmit,
Substantially reduce requirement of the anchor pole detection to operator's professional standards and engineering experience.
Further technical solution are as follows:
As AI model concrete implementation mode, setting are as follows: the AI model is to be learnt by machine to training set
Training, obtain have automatic discrimination, calculating ability model.
The disaggregated model is the model based on Bayes and neuroid classifier.
The regression model is the model based on weka software M5P/REPTree model tree method.
The invention has the following advantages:
It is mainly had the advantage that compared with the detection of existing anchor pole
1, parsing result is obtained in real time, shortens detection cycle;
2, detection process is simplified, detection difficulty is reduced;
3, the risk of human intervention detection data, the objectivity of the testing result of guarantee have been evaded;
4, detection accuracy is improved, and with the complete sustainable raising detection accuracy of training set;
5, it is directed to on-site data gathering, the requirement to related personnel is only that can get accumulation signal and transmit,
Substantially reduce requirement of the anchor pole detection to operator's professional standards and engineering experience.
Specific embodiment
Below with reference to embodiment, the present invention is described in further detail, but structure of the invention be not limited only to it is following
Embodiment.
Embodiment 1:
A kind of anchor pole AI detection method, including data collection steps and data processing step, the data processing step are
The collected data of data collection steps are parsed to obtain testing result, which is characterized in that the data processing step are as follows: will examine
Survey collection in worksite to data be sent to server, the established AI model of server calls solves detection data automatically
Analysis, is determined Grouted density by disaggregated model, is predicted using regression model rock-bolt length, finally solved AI
Scene is returned to when analysing fructufy.
In the prior art, anchor pole it is actually detected in, pass through the energy in the reflection interval of impact elasticity wave and propagation
Length and Grouted density to determine anchor pole is lost in amount, and traditional detection process and disadvantage are as follows:
1, personnel carry out data acquisition to scene need to take back indoor parsing by data, lead since resolving is more complicated
Cause detection cycle long;
2, since detection data takes back indoor parsing, so there is the risk artificially modified in detection data;
3, detection data needs artificial parsing, relatively high to the professional ability requirement of testing staff, and testing staff is needed to have
Detection experience abundant;
4, the examined personnel of accuracy in detection parse experience and influence.
In the present solution, above data acquisition step can obtain data processing step based on impact elasticity wave method needed for number
According to, it then is sent to the server by way of wireless transmission or wire transmission, in the server, the data processing step
It suddenly include that AI automatically parses step and AI judgement calculating step, it is described to automatically parse step are as follows: the data that server will receive
It is parsed, parses the characteristic parameter needed for AI model determines;
The AI determines to calculate step are as follows: the well-established AI model of server calls carries out the result after automatically parsing
It is automatic to determine to obtain result.
As those skilled in the art, the AI is artificial intelligence, the AI model in the detection process, AI model one
Being set up permanently to use, and can automatically derive as a result, simultaneously in the detection process, also can be by persistently reconstructing training set completeness
The more AI models of mode carry out retraining, AI solution to model analysis accuracy rate, precision etc. is continuously improved.
It is mainly had the advantage that compared with the detection of existing anchor pole
1, parsing result is obtained in real time, shortens detection cycle;
2, detection process is simplified, detection difficulty is reduced;
3, the risk of human intervention detection data, the objectivity of the testing result of guarantee have been evaded;
4, detection accuracy is improved, and with the complete sustainable raising detection accuracy of training set;
5, it is directed to on-site data gathering, the requirement to related personnel is only that can get accumulation signal and transmit,
Substantially reduce requirement of the anchor pole detection to operator's professional standards and engineering experience.
Embodiment 2:
The present embodiment is further qualified on the basis of embodiment 1: as AI model concrete implementation mode, setting
Are as follows: the AI model be by machine to training set carry out learning training, obtain have automatic discrimination, calculating ability mould
Type.
The disaggregated model is the model based on Bayes and neuroid classifier.
The regression model is the model based on weka software M5P/REPTree model tree method.
Embodiment 3:
The present embodiment 1 provides a kind of specific implementation based on weka software in conjunction with the embodiments:
Technology path
1) structure main Types and content are tested
(1) solid anchor pole (Solid corresponds to common bolt), hollow bolt (Hollow);
(2) length prediction (automatic returning velocity of wave): _ L.ARFF
(3) Grouted density prediction (energy loss in communication process): _ G.ARFF
Illustrate: length prediction and the ARFF file of Grouted density prediction are needed while being generated.
2) process
(1) using integrating average waveform as integrating foundation
3) length detection ARFF basic parameter
(1) anchor pole type: Solid, Hollow
(2) bolt diameter
(3) Soil Anchor Design length
(4) rock hardness information: Hard (hard, correspond to I~IV grade of country rock), Soft (it is soft, correspond to soil property, V
Grade country rock)
(5) mortar information: Null (no mortar, unhardened), Hardened (hardening).
(6) anchor pole and the compound velocity of wave calculated value of body
(7) exciting information: sensor installs Front (pasting in front), Side (side stickup)
(8) reflection interval
(9) FFT frequency difference value:
(10) CLASS: rock-bolt length value
Illustrate: AI model analysis method uses M5P model tree method
4) Grouted density detection ARFF basic parameter (elder generation)
(1) anchor pole type: Solid, Hollow
(2) bolt diameter
(3) rock hardness information: Hard (hard, correspond to I~IV grade of country rock), Soft (it is soft, correspond to soil property, V
Grade country rock)
(4) the accumulation signal half period
(5) rock-bolt length is tested
(6) attenuation curve parameter: duration
(7) amplitude ratio of the reflection signal after distance correction, i.e. reflected amplitude ratio/calculation of bolts length
(8) CLASS: Grouted density grade, numerical value (1,2,3,4)
For railway, 1,2 grade is SOUND, and 3,4 grades are DEFECT
Illustrate: AI model analysis method uses M5P/REPTree model tree method
OAC the file information
1st row: parsing coding (anchor pole)
2nd row: anchor pole type, diameter:
3rd row: rock mass grade
After 4th row: analysis of object setup parameter.
The above content is combine specific preferred embodiment to the further description of the invention made, and it cannot be said that originally
The specific embodiment of invention is only limited to these instructions.For those of ordinary skill in the art to which the present invention belongs,
The other embodiments obtained in the case where not departing from technical solution of the present invention should be included in the protection scope of corresponding invention.
Claims (4)
1. a kind of anchor pole AI detection method, including data collection steps and data processing step, the data processing step is solution
The collected data of data collection steps are analysed to obtain testing result, which is characterized in that the data processing step are as follows: will test
Collection in worksite to data be sent to server, the established AI model of server calls automatically parses detection data,
Grouted density is determined by disaggregated model, rock-bolt length is predicted using regression model, is finally parsed AI
Back to scene when fructufy.
2. a kind of anchor pole AI detection method according to claim 1, which is characterized in that the AI model is to pass through machine pair
Training set carry out learning training, obtain have automatic discrimination, calculating ability model.
3. a kind of anchor pole AI detection method according to claim 1, which is characterized in that the disaggregated model is based on pattra leaves
The model of this and neuroid classifier.
4. a kind of anchor pole AI detection method according to claim 1, which is characterized in that the regression model is based on weka
The model of software M5P/REPTree model tree method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811238914.6A CN109187770A (en) | 2018-10-23 | 2018-10-23 | A kind of anchor pole AI detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811238914.6A CN109187770A (en) | 2018-10-23 | 2018-10-23 | A kind of anchor pole AI detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109187770A true CN109187770A (en) | 2019-01-11 |
Family
ID=64942982
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811238914.6A Pending CN109187770A (en) | 2018-10-23 | 2018-10-23 | A kind of anchor pole AI detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109187770A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN203145978U (en) * | 2013-01-14 | 2013-08-21 | 山西潞安矿业(集团)有限责任公司 | Anchor rod with safety detection device |
CN103927458A (en) * | 2014-04-30 | 2014-07-16 | 长安大学 | Determination method of sensibility of influence factors of anchoring force of soil anchors |
CN106501465A (en) * | 2016-12-23 | 2017-03-15 | 石家庄铁道大学 | A kind of detection method for detecting Detection of Bolt Bonding Integrity |
-
2018
- 2018-10-23 CN CN201811238914.6A patent/CN109187770A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN203145978U (en) * | 2013-01-14 | 2013-08-21 | 山西潞安矿业(集团)有限责任公司 | Anchor rod with safety detection device |
CN103927458A (en) * | 2014-04-30 | 2014-07-16 | 长安大学 | Determination method of sensibility of influence factors of anchoring force of soil anchors |
CN106501465A (en) * | 2016-12-23 | 2017-03-15 | 石家庄铁道大学 | A kind of detection method for detecting Detection of Bolt Bonding Integrity |
Non-Patent Citations (5)
Title |
---|
QQ66688869: "WEKA", 《WWW.DOC88.COM/P-9953454197509》 * |
于广明 等: "《土木工程科学技术研究与应用(二)》", 31 October 2007 * |
杨其新 等: "《地下工程施工与管理》", 31 August 2015, 西南交通大学出版社 * |
武继胜: "沿空留巷适应性预测及巷道支护技术研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
盛小龙 等: "锚杆锚固质量无损检测技术研究应用", 《河南理工大学学报》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100416269C (en) | Non destructive detection mothod used for anchor rod anchored system | |
CN104197873B (en) | Many division transmission line icing thickness measuring methods and device based on ultrasonic ranging | |
CN101303329A (en) | Integrated strength testing method based on neural network technology | |
CN110045016A (en) | A kind of tunnel-liner lossless detection method based on audio analysis | |
Liu et al. | Overall grouting compactness detection of bridge prestressed bellows based on RF feature selection and the GA-SVM model | |
CN110082429A (en) | A kind of auxiliary judgement method of the tunnel-liner non-destructive testing of combination machine learning | |
CN110108864A (en) | A kind of prestressed concrete beam nondestructive detection system and method | |
CN106198870B (en) | A kind of anchor rod body damage location identification method based on neural network | |
CN102183584A (en) | Method and device for detecting mudjacking saturation of post-tensioned prestressed concrete beam pipe | |
CN112364296B (en) | P-wave arrival time automatic pickup method based on deep learning | |
CN106153727A (en) | A kind of building grouting plumpness detection device and detection method | |
CN108803312A (en) | A kind of new road damage detecting system | |
CN108270499B (en) | A kind of underwater sound communication performance test apparatus and method | |
CN103278843A (en) | Rockburst real-time forecasting technique device used in process of rock tunnel construction | |
CN102608223B (en) | Digital signal analysis method based on spectral energy ratio | |
CN101581223A (en) | Method for detecting slip casting effect of tunnel | |
CN115341462A (en) | Method and equipment for detecting and evaluating dense state in steel pipe concrete arch bridge pipe | |
CN109187770A (en) | A kind of anchor pole AI detection method | |
CN106323808A (en) | System and method for detecting density in due time on basis of microwave moisture meter | |
CN110059936A (en) | A kind of new construction passes through appraisal procedure before the work of existing road equipment | |
CN108037274A (en) | A kind of concrete structure real-time monitoring device and method | |
CN112926267A (en) | TBM tunnel rock burst grade prediction method and system based on tunneling parameter inversion | |
CN110231401A (en) | Information correlation method and civil structure lossless detection method between a kind of measuring point | |
CN100485412C (en) | Method for controlling field earthquake prospecting collection process | |
CN1257265A (en) | Imitation anti-disaster training examination and judging system |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190111 |
|
RJ01 | Rejection of invention patent application after publication |