CN106442291A - Corrosion fatigue life prediction method based on BP neural network and application - Google Patents
Corrosion fatigue life prediction method based on BP neural network and application Download PDFInfo
- Publication number
- CN106442291A CN106442291A CN201610870731.0A CN201610870731A CN106442291A CN 106442291 A CN106442291 A CN 106442291A CN 201610870731 A CN201610870731 A CN 201610870731A CN 106442291 A CN106442291 A CN 106442291A
- Authority
- CN
- China
- Prior art keywords
- sample
- fatigue life
- corrosion fatigue
- neural network
- corrosion
- 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
- G01N17/00—Investigating resistance of materials to the weather, to corrosion, or to light
-
- 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
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biodiversity & Conservation Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- Ecology (AREA)
- Environmental & Geological Engineering (AREA)
- Environmental Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
Abstract
The invention relates to a corrosion fatigue life prediction method based on a BP neural network and application. The prediction method comprises the following steps: selecting maximum stress, stress ratio, loading frequency and pH value of a solution as main factors influencing corrosion fatigue life; designing and processing a corrosion solution circulating device matched with a corrosion fatigue test, and carrying out a corrosion fatigue circulation failure series experiments on a high-strength sucker rod sample in a specific production environment, collecting and neatening experiment data and dividing the experiment data into training samples and prediction samples; setting artificial neuron network parameters, and establishing nonlinear mapping between the influencing factors and the corrosion fatigue life; training and testing a nervous system; and predicting the corrosion fatigue life of a new sample. The corrosion fatigue life prediction method based on the BP neural network has the beneficial effects that the corrosion fatigue life of a high-strength sucker rod is predicted by high non-linear approximation capability of the BP neural network model, and operation is simple; and the prediction method is high in generalization performance, and engineering application is facilitated.
Description
Technical field
The present invention relates to engineering mechanics field, particularly to a kind of corrosion fatigue life prediction side based on BP neural network
Method and application, for the prediction of high-strength pumping rod material corrosion fatigue life, for there being bar oil recovery scheme optimization to provide foundation.
Background technology
Increase with deep-well, ultradeep well and viscous crude well development quantity, sucker rod pump setting depth, roofbolt load increase, often
Rule E, D grade pumping rod has not adapted to the exploitation needs of unconventional oil well, and each elephant is gradually researched and developed, adopted high-strength pumping-oil
Bar is replacing conventional sucker rod.Simultaneously because the development of tertiary oil recovery, the application of especially produced-water reinjection technology causes sucker rod
Working environment is increasingly severe, and the pH value of well liquid is higher, and sucker rod corrosion phenomenon is very universal.In tension and compression alternate load and corrosion ring
Under the synergy in border, sucker rod is susceptible to fatigue corrosion fracture.In this case, the service life of high-strength pumping rod
Far below projected life, greatly limit oil field development efficiency.
Because corrosion fatigue has very strong material-ambient interdependence, material type, load factor（Stress amplitude, stress
When load frequency, waveform etc.）And environmental factor（Solution composition, concentration and pH value）Etc. the corruption that can largely affect material
Erosion fatigue life, because influence factor is more, material corrosion fatigue life is difficult to explicit expression, and these influence factors are random simultaneously
Property big, load and ambient parameter vary slightly, and are just difficult to, according to the experimental data corresponding corrosion fatigue life of prediction, significantly limit
The popularization of corrosion fatigue experimental result, for meeting high-strength pumping rod life forecast under corrosive environment, optimizes unconventional
Oil field have bar oil recovery scheme.
Content of the invention
The purpose of the present invention is aiming at the drawbacks described above of prior art presence, provides a kind of corruption based on BP neural network
Erosion Prediction method for fatigue life and application.
A kind of corrosion fatigue life Forecasting Methodology based on BP neural network that the present invention mentions, comprises the following steps：
Step one, selection maximum stress, stress ratio, loading frequency, solution ph are the principal element of impact corrosion fatigue life；
Step 2, carries out the corrosion fatigue circulation inefficacy experiment under specific production environment, collects for high-strength pumping rod sample
Arrange experimental data, sample experiment data is divided into 2 groups, 1 group of training sample as BP neural network, another group as test
Sample；
Step 3, as network inputs, corrosion fatigue life is as net for the principal element selecting above-mentioned impact corrosion fatigue life
Network exports, and arranges artificial neural network parameter, training sample is input to network model, sets up influence factor and corrosion fatigue longevity
Nonlinear Mapping between life；
Step 4, after neural metwork training finishes, test sample is input to the neutral net training, such as consensus forecast knot
Fruit is less than specification error, then effectively, such as error is higher than specification error to network model, then need BP network parameter to be reset, directly
Reaching requirement to test result；
Step 5, the fatigue load parameter of input sample to be tested and etchant solution pH value parameter, predict corrosion fatigue life.
Preferably, the concrete grammar of step 2 is as follows：
Corrosion fatigue circulation failure test is to carry out in rotary bending tester, and high-strength pumping rod materials processing is become rod
Shape sample, sample two ends are cylinder, are easy to clamp；It is part of detecting in the middle of sample, cylindrical in variable cross-section, using becoming molding sand
Wheel plunge grinding forms, and transition arc radius R is not less than 5 times of diameter d at smallest cross-sectional；
During on-test, assemble experimental rig：First by left for testing machine chuck（12）Fastening bolt be screwed to appropriate location, by sample
Lower bare terminal end is put in lower chuck guide groove, tightens fastening bolt；Right clip position is adjusted according to specimen size, clamps the right folder of sample
Hold section to be fixed, after sample installs, successively etchant solution EGR is installed, and in etchant solution circulating box（3）
The appropriate etchant solution of middle addition, test is ready；
Fatigue load loaded frequency range is 30Hz-80Hz, Sine-wave loading, loads maximum stress and takes four groups respectively, starts examination
The machine of testing is tested, record sample to be tested load parameter, solution ph and sample fracture cycle-index, each group of load in experiment
Parameter carries out three efficiency tests altogether, averages；
Fatigue life experimental result data under the different maximum stress of record, stress ratio, loading frequency and solution ph, brush choosing,
Be organized into artificial nerve network model data sample, data sample be divided into 2 groups, 1 group be training sample, another group be test specimens
This.
Preferably, above-mentioned etchant solution EGR, including square container（1）, corrosion-resistant flexible pipe（2）, etchant solution
Circulating box（3）With corrosion-resistant electric pump（5）；Square container（1）Be divided into upper and lower two parts, side open circular hole be easy to place sample；Rotten
Erosion solution circulating box（3）Inside fill etchant solution（4）, etchant solution（4）By corrosion-resistant electric pump（5）, through flexible pipe（2）It is transported to
Square container（1）Top, through shower nozzle（6）Uniformly it is sprayed onto experimental test section surface, etchant solution（4）Again through square container（1）
Bottom connects flexible pipe（7）Return to etchant solution circulating box（3）；Due to square container（1）Solution pass through connect flexible pipe（7）By weight
Power effect is back to etchant solution circulating box（3）.
Preferably, for preventing from overflowing etchant solution circular hole in the middle of square container, in flexible pipe（2）Upper installation choke valve control
Capacity of sprinkler processed.
Preferably, wrapped up super-strong moisture absorbing functional fibre in sample testing section surface（9）It is ensured that etchant solution can be with
Sample testing section surface fully acts on；Simultaneously in sample and square container（1）Junction is mounted with sealing ring, prevents experimentation
Middle etchant solution spills etching apparatus.
Preferably, the concrete grammar of step 3 is as follows：
Using three layers of BP artificial nerve network model, including：Input layer, hidden layer and output layer, pass through adjustable between layers
The weight function of section connects, and in addition to input layer processing unit, the processing unit of other layers is non-linear input/output relation, that is,
Transforming function transformation function between unit is set up using S type function, selects maximum stress, stress ratio, loading frequency, solution ph to be that network is defeated
Enter, fatigue life exports for network, sets up the Nonlinear Mapping relation between influence factor and corrosion fatigue life.
Preferably, in step 2, load maximum stress and take 0.6 respectively、0.7、0.8、0.9Four groups, corrosion is molten
Liquid pH value is adjusted by concentration 0.1mol/L hydrochloric acid solution, arranges 5.0,5.5,6.0,6.5 and 7.0 5 pH value models altogether
Enclose.
A kind of application of corrosion fatigue life Forecasting Methodology based on BP neural network that the present invention mentions, adopts
ActiveX automatic technology realizes MATLAB and VB hybrid programming method, and application VB language organized data sample exchanges interface, adjusts
With MATLAB Neural Network Toolbox Neural Network Toolbox, realize the visualization model operation of above-mentioned steps.
The invention has the beneficial effects as follows：It is impact that the present invention selects maximum stress, stress ratio, loading frequency, solution ph
The principal element of corrosion fatigue life；The supporting etchant solution EGR of a set of corrosion fatigue test is processed in design, for height
Intensity sucker rod sample carries out the corrosion fatigue circulation inefficacy serial experiment under specific production environment, compiles experimental data simultaneously
It is divided into training sample and forecast sample；Setting artificial neural network parameter, sets up between influence factor and corrosion fatigue life
Nonlinear Mapping；Nervous system training and test；New samples corrosion fatigue life is predicted, simple to operate；Forecasting Methodology generalization
By force, it is easy to engineer applied.
Brief description
Fig. 1 is the schematic flow sheet of the corrosion fatigue life Forecasting Methodology based on BP neural network of the present invention；
Fig. 2 is the corrosion fatigue sample design figure of the present invention；
Fig. 3 is the corrosion fatigue test apparatus of the present invention；
Fig. 4 is the three-layer neural network structure chart of the present invention；
Upper in figure：Square container 1, corrosion-resistant flexible pipe 2, etchant solution circulating box 3, etchant solution 4, corrosion-resistant electric pump 5, shower nozzle 6,
Flexible pipe 7, choke valve 8, super-strong moisture absorbing functional fibre 9, sealing ring 10, right chuck 11, left chuck 12, transition arc radius R, minimum
Diameter d at section.
Specific embodiment
As shown in figure 1, the corrosion fatigue life Forecasting Methodology based on BP neural network, comprise the following steps：
Step one, material corrosion fatigue and cyclic failure test data preparation, concrete grammar is as follows：
High-strength pumping rod materials processing is become rod sample, as shown in Fig. 2 sample two ends are cylinder, is easy to clamp；Sample
Middle is part of detecting, cylindrical in variable cross-section, is formed using forming grinding wheel plunge grinding, and transition arc radius R is not less than minimum cutting
5 times of diameter d at face.
Acted on by etchant solution for guarantee specimen surface, design etchant solution acts on and EGR, as shown in figure 3, bag
Include：Square container 1, corrosion-resistant flexible pipe 2, etchant solution circulating box 3 and corrosion-resistant electric pump 5；Square container 1 is divided into upper and lower two
Point, side open circular hole be easy to place sample；Etchant solution 4 is filled, etchant solution 4 passes through corrosion-resistant in etchant solution circulating box 3
Electric pump 5, is transported to square container 1 top through flexible pipe 2, is uniformly sprayed onto experimental test section surface through shower nozzle 6, etchant solution 4 is again
Connect flexible pipe 7 through square container 1 bottom and return to etchant solution circulating box 3.Because the solution of square container 1 passes through to connect flexible pipe 7
Etchant solution circulating box 3 is back to by Action of Gravity Field, for preventing from overflowing etchant solution circular hole in the middle of square container, in flexible pipe
Choke valve 8 is installed on 2 and controls capacity of sprinkler.In experimentation, sample is at the uniform velocity rotation status, for ensureing that etchant solution can be with
Sample testing section surface fully acts on, and has wrapped up super-strong moisture absorbing functional fibre 9 in sample testing section surface；Simultaneously in sample and side
Describe that device 1 junction is mounted with sealing ring 10, prevent etchant solution in experimentation from spilling etching apparatus.
During on-test, assemble experimental rig as shown in Figure 3：First the fastening bolt of left for testing machine chuck 12 is screwed to suitably
Position, bare terminal end under sample is put in lower chuck guide groove, tightens fastening bolt；According to the specimen size right chuck of adjustment 11
Put, the clamping right gripping section of sample is fixed.After sample installs, successively etchant solution EGR is installed, and in corruption
Appropriate etchant solution is added, test is ready in erosion solution circulating box 3.
Fatigue load loaded frequency range is 30Hz-80Hz, Sine-wave loading, loads maximum stress and takes 0.6 respectively、
0.7、0.8、0.9Four groups, etchant solution pH value is adjusted by concentration 0.1mol/L hydrochloric acid solution, arranges altogether
5.0th, 5.5,6.0,6.5 and 7.0 5 pH value range.Firing test machine is tested, record sample to be tested load ginseng in experiment
Number, solution ph and sample fracture cycle-index.Each group of load parameter carries out three efficiency tests altogether, averages.Test
During it is noted that check container sealing.According to experimental period, change a corrosive solution within every 24 hours.
Fatigue life experimental result data under the different maximum stress of record, stress ratio, loading frequency and solution ph, brush
Select, be organized into artificial nerve network model data sample, according to formula [1], experimental data is normalized.By data
Sample is divided into 2 groups, 1 group be training sample, another group be test sample.
Using three layers of BP artificial nerve network model, including：Input layer, hidden layer and output layer, pass through between layers
Adjustable weight function connects, as shown in Figure 4.In addition to input layer processing unit, the processing unit of other layers is non-linear defeated
Enter/output relation, that is, adopt S type function（Sigmoid function）Set up the transforming function transformation function between unit.Select maximum stress, stress
Than, loading frequency, solution ph be network inputs, fatigue life be network output, set up influence factor and corrosion fatigue life
Between Nonlinear Mapping relation.
A kind of application of corrosion fatigue life Forecasting Methodology based on BP neural network that the present invention mentions, implements
It is：
Application interface is write using VB language, is divided into four modules：Pre-treatment, training module, survey digital-to-analogue block and prediction module.
The Grid control of pretreatment part application VB, realizes reading training sample, automatic decision sample data number from text.Should
Realize the normalization of sample data with formula [1].First according to training sample dimension in training module, set BP neural network ginseng
Number, including hidden layer neuron number, target error, learning rate and maximum cycle of training；Using ActiveX automatic technology,
Calling MATLAB BP neural network tool box, realizes the training of data sample, and call statement is as follows：
Dim matlab As Object
Set matlab = CreateObject("matlab.application")
Dim objmatlab As Object
When BP network training error is less than target error, network training finishes.
For testing the confidence level of the neural network forecast result training, network is tested.Extract test sample data, will
Network inputs layer parameter is input to the neutral net training, and network, according to the connection weight training, threshold value, automatically calculates and surveys
This bimetry of sample, test result and test sample actual result is analyzed contrasting, such as test error is less than and allows
Error, then neutral net is available, otherwise then needs to adjust network parameter or increase number of training further, re -training god
Through network till test result can use.
When network test results meet needs, neutral net can be applied to carry out the pre- of new operating mode sample corrosion fatigue life
Survey.This process is no longer necessary to carry out corrosion fatigue test, can reduce experiment and put into.
With reference to embodiment, the inventive method is described specifically.The present invention is not limited only to following instance, and all utilizations are originally
The mentality of designing of invention all enters within protection scope of the present invention.
Example sample data is FG20 super-strength sucker rod（16Mn2SiCrMoVTiA）Sample is in corrosivity well liquid
Corrosion fatigue circulate inefficacy experimental result, fatigue load maximum stress choose 540,600,660, tetra- series of 700MPa, solution
PH value is adjusted by concentration 0.1mol/L hydrochloric acid solution.
Fatigue load parameter is synthesized 24 groups with pH value random groups, each composite sample prepares 4, the circulation burn-out life takes
Its mean value, front 20 groups of sample experiment results as training sample, as shown in table 1.4 groups of experimental results are as test result afterwards,
With the validity of prediction neural network model, as shown in table 2, relative error is less than 8% to test result, disclosure satisfy that requirement of engineering.
Table 1 FG20 super-strength sucker rod steel training sample data
Table 2 BP neural network predicts the outcome
The above, be only the part preferred embodiment of the present invention, any those of ordinary skill in the art are all possibly also with upper
The technical scheme stating elaboration is changed or is revised as equivalent technical scheme.Therefore, according to technical scheme
Any simple modification being carried out or substitute equivalents, belong to the greatest extent the scope of protection of present invention.
Claims (8)
1. a kind of corrosion fatigue life Forecasting Methodology based on BP neural network, is characterized in that comprising the following steps：
Step one, selection maximum stress, stress ratio, loading frequency, solution ph are the principal element of impact corrosion fatigue life；
Step 2, carries out the corrosion fatigue circulation inefficacy experiment under specific production environment, collects for high-strength pumping rod sample
Arrange experimental data, sample experiment data is divided into 2 groups, 1 group of training sample as BP neural network, another group as test
Sample；
Step 3, as network inputs, corrosion fatigue life is as net for the principal element selecting above-mentioned impact corrosion fatigue life
Network exports, and arranges artificial neural network parameter, training sample is input to network model, sets up influence factor and corrosion fatigue longevity
Nonlinear Mapping between life；
Step 4, after neural metwork training finishes, test sample is input to the neutral net training, such as consensus forecast knot
Fruit is less than specification error, then effectively, such as error is higher than specification error to network model, then need BP network parameter to be reset, directly
Reaching requirement to test result；
Step 5, the fatigue load parameter of input sample to be tested and etchant solution pH value parameter, predict corrosion fatigue life.
2. the corrosion fatigue life Forecasting Methodology based on BP neural network according to claim 1, is characterized in that：Step 2
Concrete grammar as follows：
Corrosion fatigue circulation failure test is to carry out in rotary bending tester, and high-strength pumping rod materials processing is become rod
Shape sample, sample two ends are cylinder, are easy to clamp；It is part of detecting in the middle of sample, cylindrical in variable cross-section, using becoming molding sand
Wheel plunge grinding forms, and transition arc radius R is not less than 5 times of diameter d at smallest cross-sectional；
During on-test, assemble experimental rig：First by left for testing machine chuck（12）Fastening bolt be screwed to appropriate location, by sample
Lower bare terminal end is put in lower chuck guide groove, tightens fastening bolt；Right clip position is adjusted according to specimen size, clamps the right folder of sample
Hold section to be fixed, after sample installs, successively etchant solution EGR is installed, and in etchant solution circulating box（3）
The appropriate etchant solution of middle addition, test is ready；
Fatigue load loaded frequency range is 30Hz-80Hz, Sine-wave loading, loads maximum stress and takes four groups respectively, starts examination
The machine of testing is tested, record sample to be tested load parameter, solution ph and sample fracture cycle-index, each group of load in experiment
Parameter carries out three efficiency tests altogether, averages；
Fatigue life experimental result data under the different maximum stress of record, stress ratio, loading frequency and solution ph, brush choosing,
Be organized into artificial nerve network model data sample, data sample be divided into 2 groups, 1 group be training sample, another group be test specimens
This.
3. the corrosion fatigue life Forecasting Methodology based on BP neural network according to claim 2, is characterized in that：Described
Etchant solution EGR, including square container（1）, corrosion-resistant flexible pipe（2）, etchant solution circulating box（3）With corrosion-resistant electric pump
（5）；Square container（1）Be divided into upper and lower two parts, side open circular hole be easy to place sample；Etchant solution circulating box（3）Inside fill
Etchant solution（4）, etchant solution（4）By corrosion-resistant electric pump（5）, through flexible pipe（2）It is transported to square container（1）Top, through spray
Head（6）Uniformly it is sprayed onto experimental test section surface, etchant solution（4）Again through square container（1）Bottom connects flexible pipe（7）Return to corruption
Erosion solution circulating box（3）；Due to square container（1）Solution pass through connect flexible pipe（7）Etchant solution is back to by Action of Gravity Field
Circulating box（3）.
4. the corrosion fatigue life Forecasting Methodology based on BP neural network according to claim 3, is characterized in that：In flexible pipe
（2）Upper installation choke valve（8）Control capacity of sprinkler, be also prevented from overflowing etchant solution circular hole in the middle of square container.
5. the corrosion fatigue life Forecasting Methodology based on BP neural network according to claim 3, is characterized in that：
Wrap up super-strong moisture absorbing functional fibre in sample testing section surface（9）It is ensured that etchant solution can be with sample testing section
Surface fully acts on；Simultaneously in sample and square container（1）Junction is mounted with sealing ring, prevents etchant solution in experimentation
Spill etching apparatus.
6. the corrosion fatigue life Forecasting Methodology based on BP neural network according to claim 1, is characterized in that：Step 3
Concrete grammar as follows：
Using three layers of BP artificial nerve network model, including：Input layer, hidden layer and output layer, pass through adjustable between layers
The weight function of section connects, and in addition to input layer processing unit, the processing unit of other layers is non-linear input/output relation, that is,
Transforming function transformation function between unit is set up using S type function, selects maximum stress, stress ratio, loading frequency, solution ph to be that network is defeated
Enter, fatigue life exports for network, sets up the Nonlinear Mapping relation between influence factor and corrosion fatigue life.
7. the corrosion fatigue life Forecasting Methodology based on BP neural network according to claim 1, is characterized in that：Step 2
In, load maximum stress and take 0.6 respectively、0.7、0.8、0.9Four groups, etchant solution pH value passes through concentration
0.1mol/L hydrochloric acid solution is adjusted, and arranges 5.0,5.5,6.0,6.5 and 7.0 5 pH value range altogether.
8. a kind of corrosion fatigue life Forecasting Methodology based on BP neural network as any one of claim 1-7
Application, is characterized in that：MATLAB and VB hybrid programming method, application VB language establishment are realized using ActiveX automatic technology
Data sample exchanges interface, Calling MATLAB Neural Network Toolbox Neural Network Toolbox, realizes above-mentioned steps
Visualization model operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610870731.0A CN106442291A (en) | 2016-09-30 | 2016-09-30 | Corrosion fatigue life prediction method based on BP neural network and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610870731.0A CN106442291A (en) | 2016-09-30 | 2016-09-30 | Corrosion fatigue life prediction method based on BP neural network and application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106442291A true CN106442291A (en) | 2017-02-22 |
Family
ID=58171802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610870731.0A Pending CN106442291A (en) | 2016-09-30 | 2016-09-30 | Corrosion fatigue life prediction method based on BP neural network and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106442291A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109190749A (en) * | 2018-06-25 | 2019-01-11 | 中国电力科学研究院有限公司 | A kind of prediction technique and device for the intelligent electric meter service life |
| CN109444029A (en) * | 2018-09-30 | 2019-03-08 | 山东省分析测试中心 | A kind of fatigue test environment simulator |
| CN109855991A (en) * | 2019-01-16 | 2019-06-07 | 广州大学 | A method of prediction composite structures fatigue life |
| CN111965061A (en) * | 2020-07-30 | 2020-11-20 | 北京科技大学 | Test method and device for simulating and evaluating corrosion fatigue performance of welding heat affected zone |
| CN112287302A (en) * | 2020-12-18 | 2021-01-29 | 震坤行网络技术(南京)有限公司 | Method for detecting pH value of oil, computing equipment and computer storage medium |
| CN112380768A (en) * | 2020-11-11 | 2021-02-19 | 长沙理工大学 | BP neural network-based LED chip life prediction method |
| CN113742948A (en) * | 2021-08-23 | 2021-12-03 | 西安石油大学 | Novel model and method for P-S-N curve fitting of ultrahigh-strength sucker rod |
| CN114060007A (en) * | 2021-12-15 | 2022-02-18 | 中海石油（中国）有限公司天津分公司 | XGboost-based oil well electric pump service life prediction method and detection device |
| CN115081321A (en) * | 2022-06-15 | 2022-09-20 | 天津大学 | Corrosion fatigue life prediction method, system and equipment for marine welding structure |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101221108A (en) * | 2008-01-30 | 2008-07-16 | 中国科学院力学研究所 | Rotating and bending corrosion fatigue testing device |
| WO2014060001A1 (en) * | 2012-09-13 | 2014-04-24 | FRENKEL, Christina | Multitransmitter model of the neural network with an internal feedback |
| CN203745426U (en) * | 2014-04-02 | 2014-07-30 | 四川大学 | Ultra-long life corrosion fatigue experiment device adopting ultrasonically accelerated vibration |
| CN104062196A (en) * | 2014-01-08 | 2014-09-24 | 中国石油大学(华东) | Corrosion fatigue life prediction method based on damage evolution |
| CN105653791A (en) * | 2015-12-29 | 2016-06-08 | 中国石油天然气集团公司 | Data mining based corrosion failure prediction system for on-service oil pipe column |
| CN105866017A (en) * | 2016-06-30 | 2016-08-17 | 云南公投建设集团有限公司 | Intelligent electrochemical accelerated rusting device and control method thereof |
-
2016
- 2016-09-30 CN CN201610870731.0A patent/CN106442291A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101221108A (en) * | 2008-01-30 | 2008-07-16 | 中国科学院力学研究所 | Rotating and bending corrosion fatigue testing device |
| WO2014060001A1 (en) * | 2012-09-13 | 2014-04-24 | FRENKEL, Christina | Multitransmitter model of the neural network with an internal feedback |
| CN104062196A (en) * | 2014-01-08 | 2014-09-24 | 中国石油大学(华东) | Corrosion fatigue life prediction method based on damage evolution |
| CN203745426U (en) * | 2014-04-02 | 2014-07-30 | 四川大学 | Ultra-long life corrosion fatigue experiment device adopting ultrasonically accelerated vibration |
| CN105653791A (en) * | 2015-12-29 | 2016-06-08 | 中国石油天然气集团公司 | Data mining based corrosion failure prediction system for on-service oil pipe column |
| CN105866017A (en) * | 2016-06-30 | 2016-08-17 | 云南公投建设集团有限公司 | Intelligent electrochemical accelerated rusting device and control method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 黄小光 等: "基于ANN的材料力学行为试验分析方法及实践", 《实验室研究与探索》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109190749A (en) * | 2018-06-25 | 2019-01-11 | 中国电力科学研究院有限公司 | A kind of prediction technique and device for the intelligent electric meter service life |
| CN109444029A (en) * | 2018-09-30 | 2019-03-08 | 山东省分析测试中心 | A kind of fatigue test environment simulator |
| CN109855991A (en) * | 2019-01-16 | 2019-06-07 | 广州大学 | A method of prediction composite structures fatigue life |
| CN111965061A (en) * | 2020-07-30 | 2020-11-20 | 北京科技大学 | Test method and device for simulating and evaluating corrosion fatigue performance of welding heat affected zone |
| CN111965061B (en) * | 2020-07-30 | 2023-03-21 | 北京科技大学 | Test method and device for simulating and evaluating corrosion fatigue performance of welding heat affected zone |
| CN112380768A (en) * | 2020-11-11 | 2021-02-19 | 长沙理工大学 | BP neural network-based LED chip life prediction method |
| CN112287302A (en) * | 2020-12-18 | 2021-01-29 | 震坤行网络技术(南京)有限公司 | Method for detecting pH value of oil, computing equipment and computer storage medium |
| CN113742948A (en) * | 2021-08-23 | 2021-12-03 | 西安石油大学 | Novel model and method for P-S-N curve fitting of ultrahigh-strength sucker rod |
| CN114060007A (en) * | 2021-12-15 | 2022-02-18 | 中海石油（中国）有限公司天津分公司 | XGboost-based oil well electric pump service life prediction method and detection device |
| CN114060007B (en) * | 2021-12-15 | 2023-11-28 | 中海石油（中国）有限公司天津分公司 | XGBoost-based oil well electric pump life prediction method and detection device |
| CN115081321A (en) * | 2022-06-15 | 2022-09-20 | 天津大学 | Corrosion fatigue life prediction method, system and equipment for marine welding structure |
| CN115081321B (en) * | 2022-06-15 | 2023-04-07 | 天津大学 | Corrosion fatigue life prediction method, system and equipment for marine welding structure |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106442291A (en) | Corrosion fatigue life prediction method based on BP neural network and application | |
| Navaid et al. | A comprehensive literature review on the challenges associated with underground hydrogen storage | |
| CN101706482B (en) | Full-automatic vacuum degasser and degassing method thereof | |
| CN101226134A (en) | Dynamic simulation tester in erosion chamber | |
| CN108509694A (en) | A kind of proppant laying form prediction method based on BP neural network | |
| CN111305796A (en) | Experimental device and method for stability of tubular column in hydrate pilot production process | |
| CN108229062B (en) | Method based on sensibility micro-parameter prediction cemented fill mechanical response characteristic | |
| CN105675384B (en) | Stability evaluation and layout optimization method for deep anhydrite cavern group in oil storage environment | |
| CN109138943B (en) | Fracture-cavity carbonate reservoir targeted gas injection method and system | |
| CN108827856B (en) | Rock plate adding device and method for flow conductivity evaluation experiment | |
| CN107063941A (en) | A kind of experimental rig for simulating pyrolysis pollutant diffusion And Movement in underground water in situ | |
| CN201477086U (en) | Full-automatic vacuum degasser | |
| CN102032443A (en) | Gasholder load prediction equipment and method and gas scheduling system and method | |
| CN110134988A (en) | A kind of multi-stress reinforced test profile of device level product based on the physics of failure determines method | |
| CN1318314C (en) | Integrated dynamic balance treating method for oily sewage and apparatus thereof | |
| CN106990010A (en) | Liquid erosion test device in a kind of many reference amounts regulation and control pipeline | |
| CN205117320U (en) | Coal reservoir fracture pattern changes simulating measurement setup under different fracturing technologies | |
| CN107167376A (en) | Ocean engineering corrosive pipeline experimental facilities | |
| CN108007853B (en) | Corrosion test device and method for grounding material in soil simulation solution | |
| CN105182440A (en) | Medium-deep natural gas reservoir filling path tracing method and equipment thereof | |
| CN107314949A (en) | A kind of heavy oil wells viscosity reducing test device and method | |
| CN114578030A (en) | Multi-medium coupling variable-temperature pressure corrosion test device and test method | |
| CN209603964U (en) | A kind of offshore oil and gas field carbon dioxide drive oil production device | |
| CN201309917Y (en) | Coal abnormal thermal change simulating device | |
| CN111413228A (en) | Test device and method for determining fatigue life of steel wire under different corrosion solvents |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170222 |