CN106845136A - A kind of needle-valve body crush and grind accuracy prediction method based on SVMs - Google Patents
A kind of needle-valve body crush and grind accuracy prediction method based on SVMs Download PDFInfo
- Publication number
- CN106845136A CN106845136A CN201710092895.XA CN201710092895A CN106845136A CN 106845136 A CN106845136 A CN 106845136A CN 201710092895 A CN201710092895 A CN 201710092895A CN 106845136 A CN106845136 A CN 106845136A
- Authority
- CN
- China
- Prior art keywords
- needle
- valve body
- grind
- crush
- value
- 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
- 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
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Grinding And Polishing Of Tertiary Curved Surfaces And Surfaces With Complex Shapes (AREA)
Abstract
The invention discloses a kind of needle-valve body crush and grind accuracy prediction method based on SVMs, it is comprised the following steps:Determine the evaluation index and factor of influence of needle-valve body grinding precision;By testing collection needle-valve body crush and grind sample data, data are divided into training set and test set;Needle-valve body crush and grind forecast model is set up, is model kernel function from Radial basis kernel function, penalty factor and nuclear parameter g are determined using cross-validation method;Using particle cluster algorithm optimal prediction model, assessed by carrying out various Optimum Experiments using different parameters mse and, predicting the outcome under different parameters value is probed into, then contrasted to obtain optimal SVM prioritization schemes.The method predicts the outcome accurately, meets processing request, can provide scientific guidance to the selection of following process parameter.
Description
Technical field
The invention belongs to machining precision prediction field, it is related to needle-valve body grinding precision Forecasting Methodology, it is more particularly to a kind of
Needle-valve body crush and grind accuracy prediction method based on SVMs.
Background technology
Needle-valve body couple is one of part of most critical in diesel fuel engine injection system, is responsible for final dynamic of fuel system
Work-oil spout.The precision of needle-valve body couple and the performance of the whole spraying system of performance impact, especially the spray orifice aperture of needle-valve body
With the atomizing effect and combustion rate that spray orifice flow directly affects fuel oil, so as to affect economy, dynamic property, the startup of diesel engine
Property and emission performance., there is substantial amounts of burr and wedge angle in the needle-valve body after being processed by early stage technique, cause flow system in spray orifice
Number is relatively low, and flow scattered error is too big, it is impossible to meet diesel engine need of work, it is necessary to carry out finished machined.Crush and grind technology conduct
The important process of diesel engine jet orifice of needle valve finished machined, by using the larger semisolid ground slurry of viscosity within the workpiece
Flowing, so as to effectively remove spray orifice inner burr and wedge angle, increases needle-valve body discharge coefficient, reduces the flow error rate of needle-valve body.
The prediction of needle-valve body grinding precision is final according to grinding precision factor of influence and each machined parameters prediction of input
Grinding precision, so that it is guaranteed that the value of machined parameters meets the requirement of grinding precision.The accuracy prediction of conventional needle-valve body grinding is more
Using empirical formula method, such as V.K.Jain and S.G.Adsul et al. by a series of experiments draw technological parameter such as abrasive concentration,
The relation between material removing rate and surface roughness such as abrasive grain and processing number of times;The Tang Wei of Wuxi Oil Pump Nozzle Inst
Equality people is drawn needle-valve body flow increment rate and tonnage, process time and tonnage and is added using orthogonal experimental method
The regression equation between reciprocation between man-hour;Yellow grain husk of Tsing-Hua University et al. utilizes homemade crush and grind equipment, passes through
A series of quantitative tests, it is determined that discharge coefficient increment rate and process time tonnage, abrasive concentration and abrasive grain
Relation and carrier between.But due to influenceing the parameter of needle-valve body crush and grind precision a lot, and between machining accuracy and parameter
Relationship Comparison is complicated, is difficult to predict both exactly with the non-linear of height, therefore the empirical equation that conventional method is obtained
Between Process Law.
SVMs is that the principle of the structural risk minimization in Statistical Learning Theory by Vapnik et al. is proposed.
Possess that regulation parameter is few, learning rate fast, classification is high with predictablity rate due to SVMs, strong robustness and good
The advantages of generalization ability, on the premise of the background information data without substantial amounts of needle-valve body attrition process, it can also be obtained
Accuracy rate higher.Forefathers attempt applying SVMs in production and processing intelligent predicting field, and such as Der-Chiang Li are
The efficiency of colour filter production technology is improved, various linear fit methods are tested in its research, it is relative using 6 in experiment
Independent variable.He finally has found that Support vector regression model of fit is the best approach for predicting manufacture efficiency.Pao-Hua
Chou et al. develops a wafer quality forecast model, it was demonstrated that support vector machine method is obtained and compares radial basis function neural network
(RBFN) and reverse transmittance nerve network (BPNN) preferably precision of prediction.But to SVMs of the optimum choice of parameter
The quality for practising precision and Generalization Ability plays decisive role, general at present all by the method for cross validation tentative calculation, Huo Zheti
Degree descent method determines that not only efficiency is low but also stability is not high.
The content of the invention
It is an object of the present invention to provide a kind of needle-valve body crush and grind accuracy prediction method based on SVMs,
SVMs (SVM) model is set up using cross-validation method, the parameter of SVM models is carried out using particle cluster algorithm (PSO)
Optimization, for the selection of following process parameter provides guidance, improves the accuracy and pin of needle-valve body crush and grind accuracy prediction
The quality of valve body crush and grind processing, so as to increased economy, dynamic property, startability and the emission performance of diesel engine.
To reach above-mentioned purpose, the technical solution adopted by the present invention is:
A kind of needle-valve body crush and grind accuracy prediction method based on SVMs, comprises the following steps:
(1) evaluation index and factor of influence of needle-valve body grinding precision, are determined,
(2), collection and treatment sample data,
(3) needle-valve body crush and grind forecast model, is set up,
(4), using particle cluster algorithm Support Vector Machines Optimized,
(5) model, is verified, optimum optimization scheme is obtained.
The step (1) is specially:Choose evaluation index of the flow error rate as grinding precision, flow error rate formula
For:
Wherein, QdIt is the target flow value of needle-valve body, QPIt is the actual measured value after processing;
Choosing four maximum factors of influence of influence needle-valve body crush and grind machining accuracy is:System temperature, system pressure,
Ground slurry use time, allowance.
Step (2) collection is specially with treatment sample data:With reference to orthogonal test, in homemade high-precision intelligent pin
Attrition process experiment, collecting sample data are carried out on valve crush and grind lathe;Data to being obtained are arranged, analyzed and excellent
Change, and data are normalized.
The step (3) sets up needle-valve body crush and grind forecast model, specially:
(3-1) is model kernel function from Radial basis kernel function, and penalty factor and nuclear parameter are determined using cross-validation method
g;
(3-2) is by sample data { x1,x2,x3,...xnIt is divided into 2 parts, wherein preceding m data is carried out as training sample
The foundation of forecast model, rear N-m data are used for forecast test, and training sample and test sample ratio follow formula:
The step (4) uses particle cluster algorithm Support Vector Machines Optimized, specially:
(4-1) is initialized:The normalized of [- 1,1] is carried out to sample data and sample data is read;Setup parameter is transported
Dynamic scope, setting Studying factors C1And C2, evolutionary generation E, penalty factor and kernel function g;
(4-2) fitness evaluation:Calculate ideal adaptation angle value, the individual optimal and global optimum of initialization;
(4-3) compares optimizing:The speed of more new particle and position produce new population, calculate the individual adaptation degree of new population
Value, is respectively compared the adaptive value and itself history optimal value and population optimal value, Population Regeneration optimized parameter C of parameter current C and g
With the global optimum of g;
(4-4) checks termination condition, and optimizing reaches maximum evolutionary generation, terminates optimizing, output optimal parameter C and g.
The step (5) verifies model, specially:
Using PSO optimizing in global scope to SVM parameters, various optimizations are carried out using different parameters using sample data
Test to assess mean square error (E/mse) and coefficient of determination R2, predicting the outcome under different parameters value is probed into, then carry out
Contrast to obtain optimal SVM prioritization schemes:
Wherein, n is test set number of samples;yi(i=1,2 ..., n) it is i-th actual value of sample;y’i(i=1,
2 ..., n) it is i-th predicted value of sample;The smaller degrees of accuracy for representing prediction of E are bigger, coefficient of determination R2Size determine
Related level of intimate, R2Represent that the goodness of fit is bigger closer to 1.
Compared with prior art, the present invention has the advantage that:
The present invention combines particle cluster algorithm (PSO) and sets up comprehensively and accurately mathematics using algorithm of support vector machine (SVM)
Model predicts the rule of the precision of needle-valve body crush and grind.Wherein, PSO optimized algorithms can join to SVM models quickly
Number optimizing, its superior overall situationization can ensure the accuracy of model.It is accurate that the present invention had both met the prediction of needle-valve body grinding precision
The requirement high of true property, realizes the requirement high of its forecasting efficiency again.
Brief description of the drawings
Fig. 1 is the flow chart of the needle-valve body crush and grind accuracy prediction method based on SVMs of the present invention.
Fig. 2 is particle group optimizing SVMs flow chart of the present invention.
Fig. 3 is the figure that predicts the outcome of SVM regressive prediction models training set of the present invention.
Fig. 4 is the figure that predicts the outcome of SVM regressive prediction models test set of the present invention.
Fig. 5 is the PSO parameter optimization fitness curve synoptic diagrams of SVM models of the present invention.
Fig. 6 is the figure that predicts the outcome of the forecast model after use PSO algorithm optimizations of the present invention.
Specific embodiment
The present invention is described in detail with reference to the accompanying drawings and detailed description.
As shown in figure 1, a kind of needle-valve body crush and grind accuracy prediction method based on SVMs, including following step
Suddenly:
Step one:Determine the evaluation index and factor of influence of needle-valve body grinding precision.
Evaluation index of the flow error rate as grinding precision is chosen, flow error rate formula is:
Wherein, QdIt is the target flow value of needle-valve body, QPIt is the actual measured value after processing;
Choosing four maximum factors of influence of influence needle-valve body crush and grind machining accuracy is:System temperature, system pressure,
Ground slurry use time, allowance.
Step 2:Sample data is gathered and treatment.
Attrition process experiment is carried out on homemade high-precision intelligent needle-valve crush and grind lathe, with reference to orthogonal experiment plan
Meter, is obtained setting the grinding effect corresponding to technological parameter, and data to being obtained are arranged, analyzed and optimized, logarithm
According to being normalized.
Step 3:Set up needle-valve body crush and grind forecast model.
(1) it is model kernel function from Radial basis kernel function, penalty factor and nuclear parameter g is determined using cross-validation method.
(2) by sample data { x1,x2,x3,...xnBe divided into 2 parts, wherein preceding m data carried out as training sample it is pre-
The foundation of model is surveyed, rear N-m data are used for forecast test.
Step 4:Particle group optimizing SVMs, as shown in Figure 2.
(1) initialize.The normalized of [- 1,1] is carried out to sample data and sample data is read.Setup parameter is moved
Scope, setting Studying factors (C1And C2), evolutionary generation (E), penalty factor and kernel function.
(2) fitness evaluation.Calculate ideal adaptation angle value, the individual optimal and global optimum of initialization.
(3) optimizing is compared.The speed of more new particle and position produce new population, calculate the ideal adaptation angle value of new population.
It is respectively compared the adaptive value and itself history optimal value and population optimal value, Population Regeneration optimized parameter C and g of parameter current C and g
Global optimum.
(4) termination condition is checked.Optimizing reaches maximum evolutionary generation, terminates optimizing, output optimal parameter C and g.
Step 5:Model is verified.
Using PSO optimizing in global scope to SVM parameters.Using sample data various optimizations are carried out using different parameters
Test to assess mse and R2, predicting the outcome under different parameters value is probed into, then contrasted to obtain optimal SVM
Prioritization scheme.
Attrition process experiment is carried out on homemade high-precision intelligent needle-valve crush and grind lathe, choosing abrasive media is
ASF-IS-A013, is processed as a example by the needle-valve body for thinking model ZCK154S427, and the needle-valve body is elongated porous formula oil spout
Mouth, its angle between spray orifices are 154 °, big to hold external diameter for 17mm, 4 spray orifices, and injection diameter is 0.27mm.Selecting system pressure, abrasive material
Use time, allowance, system temperature as grinding precision factor of influence, flow error rate grinding precision weigh factor.
Tonnage takes 3MPa, 3.5MPa, 4MPa, 4.5MPa, 5MPa, 5.5MPa, 6Mpa respectively during processing;Abrasive material is used
Time be taken as 1-35 days between numerical value;Needle-valve body allowance to be processed is taken to be typically between 7%-10%;By temperature
Control system makes system temperature control at 15-35 DEG C.
Experiment is processed by packet mode, and every ten is one group, controls the identical i.e. selection constant current of target flow of processing
Amount processing mode, the needle-valve body inputoutput data of every group of processing is present in processing history database, facilitates observation to call.It is real
Test gathered data and be such as shown in Table 1;
The sample data table of the experiment collection of table 1
Operation is normalized to sample, data normalization is interregional to (- 1,1), it is shown in Table 2;
Sample data table after the normalization of table 2
After completing preparation, start the training of sample data and the foundation of model.First with the method for cross validation
Seek optimal parameter C and g.Then the sample data after normalization is learnt and is predicted using SVM regressive prediction models.
Predicting the outcome for SVM regressive prediction model training sets is as shown in figure 3, predicting the outcome such as Fig. 4 for SVM regressive prediction model test sets
It is shown.
Using PSO optimizing in global scope to SVM parameters.Various Optimum Experiments are carried out using different parameters to assess
Mse and R2, predicting the outcome under different parameters value is probed into, PSO optimized algorithms are in parameter Maxgen=500, pop=100
When obtain optimal classification result mse=0.97656, R2=0.0065651.PSO parameter optimization fitness curves as shown in figure 5,
Predicting the outcome as shown in Figure 6 after optimization.
Above example may certify that the present invention realizes the prediction to needle-valve body crush and grind precision, and predict the outcome essence
Really, processing request is met.
Claims (6)
1. a kind of needle-valve body crush and grind accuracy prediction method based on SVMs, it is characterised in that comprise the following steps:
(1) evaluation index and factor of influence of needle-valve body grinding precision, are determined,
(2), collection and treatment sample data,
(3) needle-valve body crush and grind forecast model, is set up,
(4), using particle cluster algorithm Support Vector Machines Optimized,
(5) model, is verified, optimum optimization scheme is obtained.
2. the needle-valve body crush and grind accuracy prediction method based on SVMs according to claim 1, its feature exists
In the step (1) is specially:Evaluation index of the flow error rate as grinding precision is chosen, flow error rate formula is:
Wherein, QdIt is the target flow value of needle-valve body, QPIt is the actual measured value after processing;
Choosing four maximum factors of influence of influence needle-valve body crush and grind machining accuracy is:System temperature, system pressure, grinding
Slurry use time, allowance.
3. the needle-valve body crush and grind accuracy prediction method based on SVMs according to claim 1, its feature exists
In step (2) collection is specially with treatment sample data:With reference to orthogonal test, squeezed in homemade high-precision intelligent needle-valve
Attrition process experiment, collecting sample data are carried out on pressure lapping machine;Data to being obtained are arranged, analyzed and optimized,
And data are normalized.
4. the needle-valve body crush and grind accuracy prediction method based on SVMs according to claim 1, its feature exists
In the step (3) sets up needle-valve body crush and grind forecast model, specially:
(3-1) is model kernel function from Radial basis kernel function, and penalty factor and nuclear parameter g are determined using cross-validation method;
(3-2) is by sample data { x1,x2,x3,...xnIt is divided into 2 parts, wherein preceding m data is predicted as training sample
The foundation of model, rear N-m data are used for forecast test, and training sample and test sample ratio follow formula:
5. the needle-valve body crush and grind accuracy prediction method based on SVMs according to claim 1, its feature exists
In the step (4) uses particle cluster algorithm Support Vector Machines Optimized, specially:
(4-1) is initialized:The normalized of [- 1,1] is carried out to sample data and sample data is read;Setup parameter moves model
Enclose, setting Studying factors C1And C2, evolutionary generation E, penalty factor and kernel function g;
(4-2) fitness evaluation:Calculate ideal adaptation angle value, the individual optimal and global optimum of initialization;
(4-3) compares optimizing:The speed of more new particle and position produce new population, calculate the ideal adaptation angle value of new population, point
Do not compare the adaptive value and itself history optimal value and population optimal value of parameter current C and g, Population Regeneration optimized parameter C's and g
Global optimum;
(4-4) checks termination condition, and optimizing reaches maximum evolutionary generation, terminates optimizing, output optimal parameter C and g.
6. the needle-valve body crush and grind accuracy prediction method based on SVMs according to claim 1, its feature exists
In the step (5) verifies model, specially:
Using PSO optimizing in global scope to SVM parameters, various Optimum Experiments are carried out using different parameters using sample data
To assess mean square error (E/mse) and coefficient of determination R2, predicting the outcome under different parameters value is probed into, then contrasted
To obtain optimal SVM prioritization schemes:
Wherein, n is test set number of samples;yi(i=1,2 ..., n) it is i-th actual value of sample;y’i(i=1,2 ...,
N) it is i-th predicted value of sample;The smaller degrees of accuracy for representing prediction of E are bigger, coefficient of determination R2Size determine correlation
Level of intimate, R2Represent that the goodness of fit is bigger closer to 1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710092895.XA CN106845136A (en) | 2017-02-21 | 2017-02-21 | A kind of needle-valve body crush and grind accuracy prediction method based on SVMs |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710092895.XA CN106845136A (en) | 2017-02-21 | 2017-02-21 | A kind of needle-valve body crush and grind accuracy prediction method based on SVMs |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106845136A true CN106845136A (en) | 2017-06-13 |
Family
ID=59133208
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710092895.XA Pending CN106845136A (en) | 2017-02-21 | 2017-02-21 | A kind of needle-valve body crush and grind accuracy prediction method based on SVMs |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106845136A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563029A (en) * | 2017-08-22 | 2018-01-09 | 贵州大学 | A kind of SVMs approximate model optimization method based on K folding cross-validation methods |
CN108875156A (en) * | 2018-05-29 | 2018-11-23 | 广东工业大学 | A kind of extrusion die process parameter optimizing method based on data-driven |
CN108960486A (en) * | 2018-06-12 | 2018-12-07 | 郑州航空工业管理学院 | Interactive set evolvement method based on grey support vector regression prediction adaptive value |
CN111008791A (en) * | 2019-12-24 | 2020-04-14 | 重庆科技学院 | Bread production modeling and decision parameter optimization method based on support vector machine |
CN111805958A (en) * | 2020-07-13 | 2020-10-23 | 武汉轻工大学 | Parameter optimization method and system of spiral oil press |
CN112201822A (en) * | 2020-09-16 | 2021-01-08 | 武汉海亿新能源科技有限公司 | Temperature self-learning cooling method, device and system for hydrogen fuel cell |
CN112345382A (en) * | 2020-11-03 | 2021-02-09 | 西北农林科技大学 | Method for detecting mechanical strength of heat-treated wood |
TWI767368B (en) * | 2020-10-20 | 2022-06-11 | 國立勤益科技大學 | Intelligent ultrasonic grinding and polishing aided system and method thereof |
CN117900927B (en) * | 2024-03-05 | 2024-06-25 | 苏州力华米泰克斯胶辊制造有限公司 | Efficiency monitoring method and system for full-automatic rubber roll polishing |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101338721A (en) * | 2008-08-07 | 2009-01-07 | 余姚市舒春机械有限公司 | Marine diesel guiding valve type pintle nozzle matching parts |
CN103252706A (en) * | 2013-01-15 | 2013-08-21 | 上海大学 | Control system and method for high-precision needle valve orifice grinding |
CN103679263A (en) * | 2012-08-30 | 2014-03-26 | 重庆邮电大学 | Thunder and lightning approach forecasting method based on particle swarm support vector machine |
CN104476379A (en) * | 2014-12-03 | 2015-04-01 | 中国第一汽车股份有限公司无锡油泵油嘴研究所 | Injection nozzle extruding and grinding system |
-
2017
- 2017-02-21 CN CN201710092895.XA patent/CN106845136A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101338721A (en) * | 2008-08-07 | 2009-01-07 | 余姚市舒春机械有限公司 | Marine diesel guiding valve type pintle nozzle matching parts |
CN103679263A (en) * | 2012-08-30 | 2014-03-26 | 重庆邮电大学 | Thunder and lightning approach forecasting method based on particle swarm support vector machine |
CN103252706A (en) * | 2013-01-15 | 2013-08-21 | 上海大学 | Control system and method for high-precision needle valve orifice grinding |
CN104476379A (en) * | 2014-12-03 | 2015-04-01 | 中国第一汽车股份有限公司无锡油泵油嘴研究所 | Injection nozzle extruding and grinding system |
Non-Patent Citations (4)
Title |
---|
周苗等: "《基于神经网络的针阀体挤压研磨工艺模型》", 《机械设计与制造》 * |
柏余杰: "《微小孔精密挤压研磨工艺与控制方法研究》", 《中国博士学位论文全文数据库 工程科技I辑》 * |
赵志刚等: "《基于粒子群优化支持向量机的太阳电池温度预测》", 《物理学报》 * |
郁磊: "《MATLAB智能算法30个案例分析》", 30 September 2015, 北京航空航天大学出版社 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107563029B (en) * | 2017-08-22 | 2020-07-28 | 贵州大学 | Support vector machine approximate model optimization method based on K-fold cross verification method |
CN107563029A (en) * | 2017-08-22 | 2018-01-09 | 贵州大学 | A kind of SVMs approximate model optimization method based on K folding cross-validation methods |
CN108875156A (en) * | 2018-05-29 | 2018-11-23 | 广东工业大学 | A kind of extrusion die process parameter optimizing method based on data-driven |
CN108875156B (en) * | 2018-05-29 | 2023-04-18 | 广东工业大学 | Extrusion die technological parameter optimization method based on data driving |
CN108960486A (en) * | 2018-06-12 | 2018-12-07 | 郑州航空工业管理学院 | Interactive set evolvement method based on grey support vector regression prediction adaptive value |
CN108960486B (en) * | 2018-06-12 | 2021-10-29 | 郑州航空工业管理学院 | Interactive set evolution method for predicting adaptive value based on gray support vector regression |
CN111008791A (en) * | 2019-12-24 | 2020-04-14 | 重庆科技学院 | Bread production modeling and decision parameter optimization method based on support vector machine |
CN111805958B (en) * | 2020-07-13 | 2022-06-14 | 武汉轻工大学 | Parameter optimization method and system of spiral oil press |
CN111805958A (en) * | 2020-07-13 | 2020-10-23 | 武汉轻工大学 | Parameter optimization method and system of spiral oil press |
CN112201822A (en) * | 2020-09-16 | 2021-01-08 | 武汉海亿新能源科技有限公司 | Temperature self-learning cooling method, device and system for hydrogen fuel cell |
TWI767368B (en) * | 2020-10-20 | 2022-06-11 | 國立勤益科技大學 | Intelligent ultrasonic grinding and polishing aided system and method thereof |
CN112345382A (en) * | 2020-11-03 | 2021-02-09 | 西北农林科技大学 | Method for detecting mechanical strength of heat-treated wood |
CN117900927B (en) * | 2024-03-05 | 2024-06-25 | 苏州力华米泰克斯胶辊制造有限公司 | Efficiency monitoring method and system for full-automatic rubber roll polishing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106845136A (en) | A kind of needle-valve body crush and grind accuracy prediction method based on SVMs | |
CN110033141B (en) | Method for establishing desulfurization system operation condition database | |
CN111369070B (en) | Multimode fusion photovoltaic power prediction method based on envelope clustering | |
CN101630376B (en) | Soft-sensing modeling method and soft meter of multi-model neural network in biological fermentation process | |
Morala-Argüello et al. | A evaluation of surface roughness classes by computer vision using wavelet transform in the frequency domain | |
CN106677763B (en) | Dynamic integrated modeling-based oil well working fluid level prediction method | |
CN107169633A (en) | A kind of gas line network, gas storage peak regulating plan integrated evaluating method | |
CN104732003B (en) | A kind of grinding process appraisal procedure based on reliability | |
CN109781411A (en) | A kind of combination improves the Method for Bearing Fault Diagnosis of sparse filter and KELM | |
CN108803486A (en) | Numerical control machining tool heat error prediction based on deep learning network in parallel and compensation method | |
Li et al. | An effective PSO-LSSVM-based approach for surface roughness prediction in high-speed precision milling | |
CN102267095A (en) | Method for monitoring and dressing grinding wheel on line | |
CN102693452A (en) | Multiple-model soft-measuring method based on semi-supervised regression learning | |
CN111160457B (en) | Scroll engine fault detection method based on soft-class extreme learning machine | |
CN101984340A (en) | Method for identifying flow type of soft grain two-phase turbulence based on artificial intelligence | |
CN107238500A (en) | Vehicle handling stability tests RES(rapid evaluation system) method for building up | |
CN1307415C (en) | Soft investigating method for overflow grain index of ore grinding system based on case inference | |
CN108764346A (en) | A kind of mixing sampling integrated classifier based on entropy | |
CN107292338A (en) | A kind of feature selection approach based on sample characteristics Distribution value degree of aliasing | |
CN109255182A (en) | A kind of hard brittle material technology-parameter predictive model and its Multipurpose Optimal Method | |
CN105138824B (en) | A kind of crude data bearing calibration towards atmospheric and vacuum distillation unit optimization | |
CN112231973A (en) | Desulfurization system pH value prediction method and device | |
CN109521001A (en) | A kind of flying marking measuring method based on PSO and ε-SVR | |
CN104978484B (en) | Fuzzy forecasting model based method for detecting pulp concentration in ore grinding process of dressing plant | |
CN112372371B (en) | Method for evaluating abrasion state of numerical control machine tool cutter |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170613 |