CN107067078A - A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method - Google Patents
A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method Download PDFInfo
- Publication number
- CN107067078A CN107067078A CN201710397208.5A CN201710397208A CN107067078A CN 107067078 A CN107067078 A CN 107067078A CN 201710397208 A CN201710397208 A CN 201710397208A CN 107067078 A CN107067078 A CN 107067078A
- Authority
- CN
- China
- Prior art keywords
- metal
- metal erosion
- knowledge base
- erosion
- sample
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N17/00—Investigating resistance of materials to the weather, to corrosion, or to light
- G01N17/006—Investigating resistance of materials to the weather, to corrosion, or to light of metals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Environmental Sciences (AREA)
- Immunology (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental & Geological Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
Abstract
The present invention discloses a kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method, it is characterised in that:Using the Leonardo da Vinci's processor for possessing ARM and DSP Duo-Core Architectures and H.264 algorithm is studied the color characteristic of metal erosion image, size scaling is carried out to typical metal corrosion sample image, it is steady and rotationally-varying, set up complete metal erosion Sample Storehouse, realize monitoring, analysis and condition discrimination real time implementation, integration, create the artificial intelligence system based on artificial neural network technology expert system knowledge base, neural network structure is selected and determines according to application, select learning algorithm, pair sample relevant with Solve problems learns, to adjust the connection weight of system, complete Automated Acquisition of Knowledge and distributed storage, the knowledge base of constructing system；And metal surface breakage how many to the batch metal erosion texture, metal erosion color and luster, metal erosion shape, metal erosion etc. is detected after the salt mist experiment of metal erosion is carried out, according to anticorrosion quality grade of the knowledge base to the analysis proposition of the metal erosion related data batch metal.
Description
Technical field
The present invention relates to a kind of Metal Production manufacture field, more particularly to a kind of metal corrosion test sample and test specimen it is automatic
Ranking method.
Background technology
Domestic metal erosion experiment is generally using manually being graded, on the one hand, it is not high to there is control accuracy, grades not
Enough stabilizations, the problems such as associated picture analysis of material is not provided；On the other hand, it need to be graded with reference to current anticorrosive metal related
National standard, the experience of the information combination user oneself such as metal erosion image and phase after the salt mist experiment of metal erosion is carried out
Hope, manual analysis proposes the batch anticorrosive metal quality grade.The present invention is commented automatically in metal corrosion test sample and test specimen
In level equipment development, the artificial intelligence system being combined using expert system knowledge base and SVMs is theoretical, reaches metal
Anticorrosion quality carries out accurate automatic measure grading purpose.
The content of the invention
The purpose of the present invention proposes a kind of metal corrosion test sample and test specimen automatic measure grading method, passes through point of knowledge base
Analysis is able to carry out automatic measure grading to anticorrosive metal quality in actual production is determined.
The problems such as present invention generally uses artificial graded for metal erosion experiment, the control of engineer and brainstrust
Experience processed is generalized into one group of conditional statement of generation of qualitative description, obtains controller its quantification with fuzzy and sets theory
To receive the experience of engineer, the operation strategy of expert is imitated, the core fuzzy controller based on Fuzzy Set Theory is produced, used
In design metal corrosion test sample and test specimen intelligence grading equipment.
The technical solution of foregoing invention as shown in Fig. 2 mainly include image capturing system, inference machine, knowledge base and
Database, controller, the knowledge base are more including metal erosion texture, metal erosion color and luster, metal erosion shape, metal erosion
Few and metal surface breakage;The information of the knowledge base is eventually delivered to inference machine progress the fuzzy of metal erosion image and pushed away
Reason, reasoning finishes rear inference machine access knowledge base and database draws anticorrosive metal grade.The concrete technical scheme of the present invention
For:Using the Leonardo da Vinci's processor for possessing ARM and DSP Duo-Core Architectures and H.264 color characteristic of the algorithm to metal erosion image
Studied, size scaling, steady and rotationally-varying is carried out to typical metal corrosion sample image, complete metal erosion is set up
Sample Storehouse, realizes monitoring, analysis and condition discrimination real time implementation, integrated, creates based on artificial neural network technology expert system
The artificial intelligence system of knowledge base, neural network structure is selected and determines according to application, selects learning algorithm, pair is asked with solution
The relevant sample of topic is learnt, and to adjust the connection weight of system, is completed Automated Acquisition of Knowledge and distributed storage, is built
The knowledge base of system；To batch metal erosion texture, metal erosion color and luster, the gold after the salt mist experiment of metal erosion is carried out
Category corrosion shape, metal erosion be how many and metal surface breakage etc. is detected, according to knowledge base to metal erosion correlation
The analysis of data proposes the quality grade of the batch metal.
Beneficial effects of the present invention:For the monitoring characteristics of metal erosion, propose to be based on expert system knowledge base and support
Artificial intelligence system that vector machine is combined is theoretical, it is ensured that the precision of metal corrosion test sample and test specimen automatic measure grading equipment with
And the reliability and ageing judged.
Brief description of the drawings
Fig. 1 is the SVMs structure chart of the embodiment of the present invention.
Fig. 2 is the FB(flow block) of the embodiment of the present invention.
Embodiment
The present invention is stated in detail below in conjunction with the drawings and specific embodiments.
Concrete technical scheme of the present invention is:Using the Leonardo da Vinci's processor for possessing ARM and DSP Duo-Core Architectures
H.264 algorithm is studied the color characteristic of metal erosion image, and size contracting is carried out to typical metal corrosion sample image
Put, it is steady and rotationally-varying, set up complete metal erosion Sample Storehouse, realize monitoring, analysis and condition discrimination real time implementation, one
Change, create the artificial intelligence system based on artificial neural network technology expert system knowledge base, select and determine according to application
Neural network structure, selects learning algorithm, a pair sample relevant with Solve problems learns, to adjust the connection weight of system
Value, completes Automated Acquisition of Knowledge and distributed storage, the knowledge base of constructing system；Carry out metal erosion salt mist experiment it
How many to the batch metal erosion texture, metal erosion color and luster, metal erosion shape, the metal erosion afterwards and damaged feelings in metal surface
Condition etc. is detected, proposes the quality grade of the batch metal to the analysis of metal erosion related data according to knowledge base.
The structure chart of Fig. 1 SVMs of the present embodiment.SVMs uses different kernel functionsk(x,x i ) by shape
Into different algorithms, plan uses Radial basis kernel function：
K(x i ,x j )=exp {-︱ x- x i ︱/δ 2 =Φ(x i )Φ(x j )
Because abrasive particle is strong reflection metal surface in metal erosion image, therefore one-to-one multi-class classification method is applied, tested using intersection
Card method obtains parameter C and γ, selects wherein two category features to carry out Classification and Identification successively, is described in detail below：
Step A, chooses a small number of training samples and system is trained, training process is as follows:
Step A-1, gives c training sample of two kinds of features, extracts wear debris classifying characteristic vector in metal image and is used as classification
According to
Step A-2, carries out linear inner product supporting vector using the characteristic vector of above-mentioned sample and trains, determine linear identification model
Function；
Step A-3, carries out the linear inner product SVMs of kernel function using the characteristic vector of above-mentioned sample and trains, determine non-thread
Property identification model function
Step A-4, chooses abrasive particle test sample and enters identification, cognitive phase algorithm to system：
Step B, for given test sample to be identified, extracts wherein abrasive particle characteristic vector
Step B-1, the characteristic vector M of sample to be identified is substituted into the pattern function of linear classification SVMs, if f (M)
＞ τ, then can determine that the species of sample to be identified, and the classification thresholds that τ is obtained by training go to step 3 if f (M) ＜ τ.
Step B-2, the characteristic vector M of sample to be identified is substituted into the pattern function of Nonlinear Classification SVMs, will
Fmax (M) is grouped into corresponding defect class.
Step C, set SVMs parameter transformation scope beC=[2 6 ,2 7 ,2 8 ,…,2 12 ],γ=[2 -4 ,2 -3 ,…,
2 2 ], (C, γ) is bound into calculating cross validation precision, cross validation precision one group of parameter of highest is then chosen and is trained
And test, until completing rule-based specimen discerning.
Fig. 2 is the FB(flow block) of the embodiment of the present invention.Metal corrosion test sample and test specimen automatic measure grading system are adopted including image
Collecting system, inference machine, knowledge base and database, controller, the knowledge base include metal erosion texture, metal erosion color and luster,
Metal erosion shape, metal erosion be how many and metal surface breakage;The information of the knowledge base eventually delivers to inference machine
The fuzzy reasoning of metal erosion image is carried out, reasoning finishes rear inference machine access knowledge base and database draws anticorrosive metal etc.
Level.
The knowledge base and database of the expert system of the embodiment of the present invention include metal erosion texture database, metal erosion
Color and luster knowledge base, metal erosion shape knowledge storehouse, each quality metal anticorrosion requirement and metal surface breakage database, gold
Belong to corrosion curve database.
The knowledge base of the embodiment of the present invention has three parts, respectively camera collection image, corrosion experiment environmental information with
And metal corrosion condition information.Camera collection image includes the short transverse dot matrix metal that sonde-type image collecting device is measured
The metal surface abrasion region that abrasion image and spectrum-type image collecting device are measured.Corrosion experiment environmental information includes corrosion and tried
Agent information and corrosion device information, corrosion reagent information include the corrosion reagent name of an article, corrosion reagent concentration and corrosion reagent proportioning,
Corrosion device information includes corrosion device material, corrosion device capacity and corrosion device structure.Metal corrosion condition information includes
Metal erosion texture, metal erosion color and luster and metal erosion shape for drawing etc. are detected by camera.First carry out multiple zonules
Anticorrosive metal attributional analysis, the main contents of analysis are metal erosion shape, how many, metal erosion situation of metal erosion etc.
Information, obtained multiple information are to judge the specific data such as corrosion of metal area in zonule, metal erosion depth;Its
It is secondary, the metal erosion analysis in big region is carried out, the main contents of analysis are the information such as metal erosion texture, metal erosion color and luster,
These information are used as the foundation of intelligence system metal erosion automatic measure grading to judge the overall corrosion condition of metal.
The information that the information source of three parts of data entry system above is provided, which finally collects to inference machine, carries out gold
The fuzzy reasoning of category corrosion automatic measure grading, inference machine access associated data storehouse and knowledge base draw this collection of anticorrosive metal etc.
Level.
Described above is only that the preferred embodiment for the present invention is described, and not the spirit and scope of the present invention are entered
Row is limited, on the premise of design philosophy of the present invention is not departed from, skill of the ordinary skill technical staff such as to the present invention in this area
Metal corrosion test sample and test specimen automatic measure grading equipment and ranking method in art scheme are changed, and all should belong to the present invention
Protection domain.
Claims (3)
1. by the analysis to knowledge base, a kind of metal corrosion test sample and test specimen automatic measure grading method are proposed, and in reality
Production carries out automatic measure grading in determining to anticorrosive metal quality, it is characterised in that：It is described generally to be adopted for metal erosion experiment
The problems such as manually being graded, is generalized into the control experience of engineer and brainstrust one group of condition language of generation of qualitative description
Sentence, enables controller to receive the experience of engineer its quantification, imitates the operation plan of expert with fuzzy and sets theory
Slightly, the core fuzzy controller based on Fuzzy Set Theory is produced, is intelligently commented for designing metal corrosion test sample and test specimen
Level equipment.
2. technical solution according to claim 1, mainly including image capturing system, inference machine, knowledge base and data
Storehouse, controller, the knowledge base include metal erosion texture, metal erosion color and luster, metal erosion shape, metal erosion it is how many and
Metal surface breakage;The information of the knowledge base eventually delivers to the fuzzy reasoning that inference machine carries out metal erosion image,
Reasoning finishes rear inference machine access knowledge base and database draws anticorrosive metal grade.
3. concrete technical scheme is according to claim 1:Using the Leonardo da Vinci's processor for possessing ARM and DSP Duo-Core Architectures
H.264 algorithm is studied the color characteristic of metal erosion image, and size contracting is carried out to typical metal corrosion sample image
Put, it is steady and rotationally-varying, set up complete metal erosion Sample Storehouse, realize monitoring, analysis and condition discrimination real time implementation, one
Change, create the artificial intelligence system based on artificial neural network technology expert system knowledge base, select and determine according to application
Neural network structure, selects learning algorithm, a pair sample relevant with Solve problems learns, to adjust the connection weight of system
Value, completes Automated Acquisition of Knowledge and distributed storage, the knowledge base of constructing system；Carry out metal erosion salt mist experiment it
How many to the batch metal erosion texture, metal erosion color and luster, metal erosion shape, the metal erosion afterwards and damaged feelings in metal surface
Condition etc. is detected, proposes the quality grade of the batch metal to the analysis of metal erosion related data according to knowledge base.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710397208.5A CN107067078A (en) | 2017-05-31 | 2017-05-31 | A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710397208.5A CN107067078A (en) | 2017-05-31 | 2017-05-31 | A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107067078A true CN107067078A (en) | 2017-08-18 |
Family
ID=59617530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710397208.5A Pending CN107067078A (en) | 2017-05-31 | 2017-05-31 | A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107067078A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109300119A (en) * | 2018-09-11 | 2019-02-01 | 石家庄铁道大学 | Detection method, detection device and the terminal device in steel structure surface corrosion region |
| CN111094956A (en) * | 2017-09-22 | 2020-05-01 | 沙特阿拉伯石油公司 | Processing the thermographic image with a neural network to identify Corrosion Under Insulation (CUI) |
| CN112836717A (en) * | 2020-12-02 | 2021-05-25 | 国网重庆市电力公司电力科学研究院 | Identification method for metal corrosion aging degree of power equipment |
| CN113160909A (en) * | 2021-04-30 | 2021-07-23 | 苏州华碧微科检测技术有限公司 | Method for rapidly identifying metal corrosion failure |
| CN114046456A (en) * | 2021-11-23 | 2022-02-15 | 重庆大学 | Corrosion assessment method and system integrating fuzzy inference and neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063550A (en) * | 2011-01-07 | 2011-05-18 | 浙江工业大学 | Intelligent design system of cold extrusion piece with machine intelligence involved design decision |
| CN103674968A (en) * | 2013-12-20 | 2014-03-26 | 纪钢 | Method and device for evaluating machine vision original-value detection of exterior corrosion appearance characteristics of material |
| CN105675802A (en) * | 2014-11-19 | 2016-06-15 | 国网河南省电力公司南阳供电公司 | Transformer fault diagnosis method |
-
2017
- 2017-05-31 CN CN201710397208.5A patent/CN107067078A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063550A (en) * | 2011-01-07 | 2011-05-18 | 浙江工业大学 | Intelligent design system of cold extrusion piece with machine intelligence involved design decision |
| CN103674968A (en) * | 2013-12-20 | 2014-03-26 | 纪钢 | Method and device for evaluating machine vision original-value detection of exterior corrosion appearance characteristics of material |
| CN105675802A (en) * | 2014-11-19 | 2016-06-15 | 国网河南省电力公司南阳供电公司 | Transformer fault diagnosis method |
Non-Patent Citations (1)
| Title |
|---|
| 纪钢 等: "对镀层材料外观腐蚀特征的识别及分析处理", 《表面技术》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111094956A (en) * | 2017-09-22 | 2020-05-01 | 沙特阿拉伯石油公司 | Processing the thermographic image with a neural network to identify Corrosion Under Insulation (CUI) |
| CN109300119A (en) * | 2018-09-11 | 2019-02-01 | 石家庄铁道大学 | Detection method, detection device and the terminal device in steel structure surface corrosion region |
| CN109300119B (en) * | 2018-09-11 | 2020-10-30 | 石家庄铁道大学 | Detection method and detection device for corrosion area on surface of steel structure and terminal equipment |
| CN112836717A (en) * | 2020-12-02 | 2021-05-25 | 国网重庆市电力公司电力科学研究院 | Identification method for metal corrosion aging degree of power equipment |
| CN113160909A (en) * | 2021-04-30 | 2021-07-23 | 苏州华碧微科检测技术有限公司 | Method for rapidly identifying metal corrosion failure |
| CN114046456A (en) * | 2021-11-23 | 2022-02-15 | 重庆大学 | Corrosion assessment method and system integrating fuzzy inference and neural network |
| CN114046456B (en) * | 2021-11-23 | 2024-02-02 | 重庆大学 | Corrosion evaluation method and system for fusing fuzzy reasoning and neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107067078A (en) | A kind of metal corrosion test sample and test specimen automatic measure grading equipment and its ranking method | |
| CN108173708A (en) | Anomalous traffic detection method, device and storage medium based on incremental learning | |
| CN111104981A (en) | Hydrological prediction precision evaluation method and system based on machine learning | |
| CN101510262B (en) | Automatic measurement method for separated-out particles in steel and morphology classification method thereof | |
| CN109344905A (en) | A kind of transmission facility automatic fault recognition methods based on integrated study | |
| CN108257114A (en) | A kind of transmission facility defect inspection method based on deep learning | |
| CN110288032B (en) | Vehicle driving track type detection method and device | |
| CN107657603A (en) | A kind of industrial appearance detecting method based on intelligent vision | |
| CN109408389A (en) | A kind of aacode defect detection method and device based on deep learning | |
| CN108520273A (en) | A kind of quick detection recognition method of dense small item based on target detection | |
| CN108615046A (en) | A kind of stored-grain pests detection recognition methods and device | |
| CN105069470A (en) | Classification model training method and device | |
| CN110827260B (en) | Cloth defect classification method based on LBP characteristics and convolutional neural network | |
| CN107808358A (en) | Image watermark automatic testing method | |
| CN110263934B (en) | Artificial intelligence data labeling method and device | |
| CN106651574A (en) | Personal credit assessment method and apparatus | |
| CN113901977A (en) | Deep learning-based power consumer electricity stealing identification method and system | |
| CN108647722B (en) | Zinc ore grade soft measurement method based on process size characteristics | |
| CN109948522A (en) | A kind of X-ray hand osseous maturation degree interpretation method based on deep neural network | |
| CN107622276B (en) | Deep learning training method based on combination of robot simulation and physical sampling | |
| CN108932712A (en) | A kind of rotor windings quality detecting system and method | |
| CN115673596B (en) | Welding abnormity real-time diagnosis method based on Actor-Critic reinforcement learning model | |
| CN109345050A (en) | A kind of quantization transaction prediction technique, device and equipment | |
| CN108763096A (en) | Software Defects Predict Methods based on depth belief network algorithm support vector machines | |
| Zhang et al. | Research on surface defect detection algorithm of strip steel based on improved YOLOV3 |
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 |
Application publication date: 20170818 |
|
| WD01 | Invention patent application deemed withdrawn after publication |