CN108363896A - A kind of hydraulic cylinder method for diagnosing faults - Google Patents
A kind of hydraulic cylinder method for diagnosing faults Download PDFInfo
- Publication number
- CN108363896A CN108363896A CN201810441856.0A CN201810441856A CN108363896A CN 108363896 A CN108363896 A CN 108363896A CN 201810441856 A CN201810441856 A CN 201810441856A CN 108363896 A CN108363896 A CN 108363896A
- Authority
- CN
- China
- Prior art keywords
- hydraulic cylinder
- under
- model
- data
- simulation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a kind of hydraulic cylinder method for diagnosing faults, carry out physical modeling emulation to hydraulic cylinder first in AMESim softwares and complete direct fault location process, obtain fault data;Then wavelet-packet energy extraction, construction feature vector are carried out to data;Then the fault diagnosis model based on high dimensional nonlinear grader is established;Fault diagnosis model is trained using training sample, and calculates the optimal solution of parameter needed for model using genetic algorithm, constructs Multi-class Classifier;Hydraulic cylinder test sample is input to model and carries out fault diagnosis.The present invention efficiently solves the problem of fault data deficiency during proof of algorithm, and the application that the high dimensional nonlinear classifier algorithm of ginseng is sought in conjunction with genetic algorithm optimization effectively solves the problems, such as non-linear, small sample, improves the fault identification ability of fault diagnosis model.
Description
Technical field
The present invention relates to hydraulic cylinder fault diagnosis fields, being based on WAVELET PACKET DECOMPOSITION more particularly to one kind and heredity is calculated
The hydraulic cylinder method for diagnosing faults of the support vector machines (WPT-GASVM) of method optimization.
Background technology
Hydraulic system is widely used in industrial production, is the important component of complication system.And hydraulic cylinder is as liquid
The execution part in servo-drive system is pressed, very extensive utilization is obtained on realizing linear motion or rotary reciprocating motion, is good for
Health state directly affects entire production activity.Hydraulic cylinder once breaks down, and gently then causes the damage of equipment or load, heavy then prestige
Personal safety as well as the property safety is coerced, is caused a serious accident.As equipment is intended to enlargement, complication, traditional planned maintenance is difficult
To meet industrial demand.
Often there is fault data deficiency in the previous method for diagnosing faults for being directed to hydraulic cylinder, and equipment was run
Lack fault data in the data generated in journey and there are a large amount of hashes.On the other hand, in the selection of method for diagnosing faults
On, traditional to be difficult to be applied because modeling accuracy the problem of based on the method for model, the feelings that regard based on data-driven are tieed up
It repaiies and receives more and more attention and develop.
Chinese Patent Application No.:201610497436.5 patent name:《Hydraulic cylinder interior leakage leaks fault diagnosis assessment side
Method》, which carries out feature extraction using wavelet analysis method to pressure signal, in conjunction with BP neural network, realization hydraulic cylinder
The diagnostic assessment of internal leakage grade.
Document《The research of method for diagnosing faults is revealed based on LS-SVM hydraulic cylinders》(journal title:《Lathe and hydraulic pressure》, reel number
Etc. information:2017,45(15):184-187) using hydraulic cylinder leakage as research object, classifies to hydraulic cylinder leak degree, adopt
Diagnosis target is realized with least square method supporting vector machine method.
Although the method that the prior art has successfully used data-driven carries out fault diagnosis to hydraulic cylinder, there are still failures
The problem of data acquisition difficulty and model parameter select.
Invention content
The technical problem to be solved by the present invention is to for defect involved in background technology, provide a kind of based on small
Wave packet decomposes and the hydraulic cylinder method for diagnosing faults of the support vector machines (WPT-GASVM) of genetic algorithm optimization.
The present invention uses following technical scheme to solve above-mentioned technical problem:
A kind of hydraulic cylinder method for diagnosing faults, including following specific steps:
Step 1) establishes simulation of hydraulic cylinder model by AMESim softwares;
Step 2) carries out direct fault location, obtains N group simulation of hydraulic cylinder models under each state of state set respectively
Data on flows is imported and exported, the state set includes normal operating conditions, hydraulic cylinder leakage malfunction, load sudden change failure shape
State and piston rod line skew malfunction;
Step 3), using WAVELET PACKET DECOMPOSITION method to every group of inlet and outlet flow of simulation of hydraulic cylinder model under each state
Data carry out four layers of decomposition, obtain the decomposed signal under 16 frequency bands of every group of inlet and outlet data on flows, and calculate every group of disengaging
It energy value under each frequency range of the inlet flow rate data of mouthful data on flows, the decomposed signal of rate of discharge data and then obtains
The group imports and exports the feature vector of data on flows, in conjunction with the shape of the corresponding simulation of hydraulic cylinder model of each group inlet and outlet data on flows
State ultimately forms characteristic data set;
Step 4) joins the penalty factor of supporting vector machine model and preset kernel function using genetic algorithm
Number carries out optimizing, and establishes multivalue support vector machine classifier on this basis, then characteristic set pair multivalue is utilized to support
Vector machine classifier is trained, the multivalue support vector machine classifier after being trained;
Step 5) carries out feature extraction to hydraulic cylinder inlet and outlet data on flows to be tested, obtains its feature vector, and will
It inputs the multivalue support vector machine classifier after training, obtains the state of hydraulic cylinder to be tested, completes fault diagnosis.
As a kind of further prioritization scheme of hydraulic cylinder method for diagnosing faults of the present invention, the detailed step of the step 1)
It is as follows:
Step 1.1) places servo amplifier, servo valve, hydraulic cylinder and position under the draft mode of AMESim softwares
Sensor model forms the sketch of simulation of hydraulic cylinder model;
Step 1.2), under the submodel pattern of AMESim softwares be servo amplifier, servo valve, hydraulic cylinder and position pass
Sensor model chooses mathematical model;
Step 1.3) is respectively servo amplifier, servo valve, hydraulic cylinder and position under the parameter model of AMESim softwares
Sensor model arrange parameter completes the setting of simulation of hydraulic cylinder model;
Step 1.4) runs the simulation of hydraulic cylinder model under the operational mode of AMESim softwares, until run time is big
In preset duration threshold value, to confirm that the simulation of hydraulic cylinder model running is normal.
As a kind of further prioritization scheme of hydraulic cylinder method for diagnosing faults of the present invention, the detailed step of the step 2)
It is as follows:
Step 2.1) runs simulation of hydraulic cylinder model under the operational mode of AMESim softwares, obtains N groups and works normally item
The inlet and outlet data on flows of simulation of hydraulic cylinder model under part, N are the natural number more than 0;
Step 2.2) selects simulation of hydraulic cylinder model under the parameter mode of AMESim softwares, changes parameter
Leakagecoefficient reveals failure to form hydraulic cylinder, and runs hydraulic cylinder under the operational mode of AMESim softwares and imitate
True mode obtains the inlet and outlet data on flows of simulation of hydraulic cylinder model under N group hydraulic cylinders leakage failure;
Step 2.3) selects simulation of hydraulic cylinder model, change parameter total under the parameter mode of AMESim softwares
Mass being moved run simulation of hydraulic cylinder mould to form load sudden change failure under the operational mode of AMESim softwares
Type obtains the inlet and outlet data on flows of simulation of hydraulic cylinder model under N group load sudden change failures;
Step 2.4) selects simulation of hydraulic cylinder model, change parameter angle under the parameter mode of AMESim softwares
Rod makes with horizontal are to form piston rod line skew failure, and under the operational mode of AMESim softwares
Simulation of hydraulic cylinder model is run, the inlet and outlet data on flows of simulation of hydraulic cylinder model under N group piston rod line skew failures is obtained.
As a kind of further prioritization scheme of hydraulic cylinder method for diagnosing faults of the present invention, the detailed step of the step 3)
It is as follows:
Step 3.1), choose wavelet basis function system under normal running conditions, hydraulic cylinder leakage failure under, load sudden change therefore
Every group of inlet and outlet data on flows of simulation of hydraulic cylinder model carries out four layers points of discrete wavelet packet under barrier, piston rod line skew failure
Solution, obtains decomposed signal of the every group of inlet and outlet data on flows under 16 frequency bands;
Step 3.2) calculates separately the decomposition of inlet flow rate data, rate of discharge data in every group of inlet and outlet data on flows
Energy value of the signal under each frequency band, and be ranked up according to sequence from big to small;
Step 3.3) chooses its inlet flow rate data energy value under each frequency band for every group of inlet and outlet data on flows
Preceding k1A, its rate of discharge data energy value under each frequency band preceding k2The vector of a composition is as its feature vector;
Step 3.4) imports and exports the feature vector and its corresponding simulation of hydraulic cylinder model of data on flows according to each group
State forms characteristic data set.
As a kind of further prioritization scheme of hydraulic cylinder method for diagnosing faults of the present invention, the detailed step of the step 4)
It is as follows:
Step 4.1), is arranged the parameter initialization value of genetic algorithm, and the parameter initialization value includes evolutionary generation, population
Scale, select probability, crossover probability, mutation probability, fitness function and individual optimizing space;
Step 4.2), using the penalty factor of supporting vector machine model and preset kernel functional parameter as population at individual,
Initial population is randomly generated in optimizing space;
Step 4.3) iteratively solves optimum model parameter, by selection operation, crossover operation and mutation operation, in conjunction with suitable
Response function obtains parameter optimal value;
Step 4.4) builds multivalue support vector machine classifier according to parameter optimal value is obtained, and from the characteristic
Concentrate random selection 90% data as training set, residue 10% conduct test set to the multivalue support vector machine classifier into
Row training;
Step 4.5) repeats step 4.1) to step 4.4), until the fault diagnosis accuracy rate of test set is more than preset
Accuracy rate threshold value.
It is described pre- in the step 4) as a kind of further prioritization scheme of hydraulic cylinder method for diagnosing faults of the present invention
The kernel functional parameter first set is gaussian kernel function parameter, and penalty factor=16 of supporting vector machine model, gaussian kernel function
Parameter γ=3.0314.
The present invention has the following technical effects using above technical scheme is compared with the prior art:
1) it is arranged by physical modeling and simulation of hydraulic cylinder model parameter, various faults injection may be implemented, and can criticize
Amount operation, simplifies direct fault location process, solves the problems, such as fault data deficiency;
2) support vector machine method of genetic algorithm optimizing is used to solve supporting vector machine model parameter selection difficult and fast
Slow problem is spent, and compared to neural network method, there can be better effect in Small Sample Database.
Description of the drawings
Fig. 1 is hydraulic cylinder troubleshooting step flow chart;
Fig. 2 is the algorithm of support vector machine flow chart of genetic algorithm optimization.
Specific implementation mode
Technical scheme of the present invention is described in further detail below in conjunction with the accompanying drawings:
The present invention can be embodied in many different forms, and should not be assumed that be limited to the embodiments described herein.On the contrary,
These embodiments are provided so that the disclosure is thorough and complete, and the model for the present invention being given full expression to those skilled in the art
It encloses.In the accompanying drawings, for the sake of clarity it is exaggerated component.
Present example object is a kind of typical valve control cylinder mode position control system, include mainly servo amplifier,
Four part of servo valve, hydraulic cylinder and position sensor.The system mainly controls hydraulic pressure by hydraulic pump and three position four-way directional control valve
Cylinder moves, and then again by sensor feedback location information, further adjustment control reversing valve signal, realizes elaborate position control.
As shown in Figure 1, the present invention proposes a kind of hydraulic cylinder method for diagnosing faults for such system, implement according to the following steps:
Step 1:System simulation model is established using AMESim Rev13 softwares
Step 1.1:Determine valve control cylinder mode position control system model and parameter setting
AMESim software phantoms are broadly divided into draft mode, submodel pattern, parameter mode and operational mode.First
Each element is placed under draft mode completes model sketch;Each element mathematical model is chosen under submodel pattern again, generally
Select Primal Mode;Then it is each emulation element arrange parameter under parameter model, is specifically shown in Table 1;Finally transporting
Run time is set under row pattern 12 seconds.By observing hydraulic cylinder inlet flow rate curve, rate of discharge curve, piston rod displacement song
Line etc., it can be determined that the correctness of institute's established model.
1 pressurized strut position control system component parameters of table
Step 1.2:Fault filling method
Under parameter mode, select Hydraulic Cylinder Model, change parameter leakage coefficient that can obtain hydraulic cylinder
Reveal failure;Change total mass being moved can obtain load sudden change failure;Change angle rod makes
With horizontal can obtain piston rod line skew failure.Multi-group data can be obtained simultaneously by batch processing, specifically
Operation is:Under parameter mode, the Batch parameters tabss in the Settings in menu bar are selected, are then double-clicked
Parameter such as leakage coefficient are dragged to Batch parameters tabs left areas by Hydraulic Cylinder Model, are connect
It and inputs multigroup setting value on the right side of tabs.
Step 1.3:Operation and data preserve
In the operating mode, run time Final time and sampling step length Print interval are set, Run is selected
Batch operational modes in type.After end of run, Hydraulic Cylinder Model is double-clicked, by flow in Variable List dialog boxes
1 parameters of rate at port are dragged to model interface blank space pop-up AMEPlot windows and show inlet flow rate curve, similarly
2 parameters of flow rate at port are dragged to model interface blank space display outlet flow curve.In AMEPlot windows,
Save data ... the orders in File menus are selected, data are saved as into .data files, this article is then opened in notepad
Part, in all data copies to Excel tables.By above step, respectively in normal running conditions and three kinds of failures
50 groups of inlet and outlet datas on flows are respectively obtained under pattern, 200 groups of inlet and outlet datas on flows are obtained, and every group of data include inlet flow rate
Data and rate of discharge data.
Step 2:Wavelet pack energy feature extracts
Step 2.1:4 layers of WAVELET PACKET DECOMPOSITION are carried out to every group of inlet flow rate signal using db5 wavelet functions system, 16 after decomposition
The WAVELET PACKET DECOMPOSITION coefficient of a frequency range isWhereinIndicate inlet flow rate signal 4
The coefficient under n-th of frequency range after layer WAVELET PACKET DECOMPOSITION,It is specific coefficient value;
Step 2.2:Energy feature is further extracted on the basis of WAVELET PACKET DECOMPOSITION and draws energy histogram.Specific meter
Calculation method is:Each band energy value isWherein i indicates that i-th of frequency range, j indicate the jth under i-th of frequency range
A coefficient;Then it is normalized,Obtain energy vectors
Step 2.3:The value of energy Relatively centralized is chosen from 16 energy values as feature vector, observes energy histogram
Select preceding 4 feature vectors as inlet flow rate.
Step 2.4:According to above step, same treatment is done to rate of discharge, still selects preceding 4 energy values.Therefore, often
Every data on flows under kind operating mode can obtain the feature vector T=[E of one 8 dimension1 E2 E3 E4 E5 E6 E7
E8], wherein E1 E2 E3 E4It is hydraulic cylinder inlet flow rate energy eigenvalue, E5 E6 E7 E8It is that hydraulic cylinder rate of discharge energy is special
Value indicative.Therefore, the characteristic of 200 group of 8 dimension is obtained, and nominal situation, leakage failure, load are uprushed failure and piston rod
Axle center shift fault is respectively labeled as 1,2,3 and 4.
Step 3:Supporting vector machine model based on genetic algorithm optimization is built, trains and is tested, as shown in Figure 2.
Step 3.1:Multi-category support vector machines (SVM) are built.Using " one-to-one " method, by combining multiple two-values point
Class device constructs multi-categorizer, by k class data combination of two, builds k* (k-1)/2 two-value grader;Each two-value grader is only
It is trained for two class data, the final classification of every group of characteristic is differentiated in conjunction with ballot method;Further, since data characteristics number
Amount less and sample size is normal, gaussian kernel function k (x may be used1, x2)=exp (- γ | | x1-x2||2), wherein γ is Gauss
The parameter of kernel function.
Step 3.2:Determine the SVM model basic parameter penalty factors for needing to optimize and gaussian kernel function parameter γ.
Step 3.3:Optimal parameter is selected using genetic algorithm (GA).GA methods randomly generate one comprising multiple first
The population of solution selects certain amount individual as next-generation kind then according to intersection, variation and selection operation out of current population
Group, after evolving several times, converges on optimal individual.GA parameters are set:Evolutionary generation 200, population quantity 20, selection are general
Rate 0.9, crossover probability 0.7, mutation probability 0.7;Parameter C optimizing space be (0,100];Parameter γ optimizing space [0,100];Choosing
It is fitness function value to select accuracy rate of the training set under 5 folding cross validations.After evolution, selection makes fitness function most
High individual is optimum individual.Finally obtained optimized parameter C=16, γ=3.0314.
Step 3.4:SVM model trainings and test.The present embodiment selects the tool boxes LibSVM to be instructed in Matlab platform buildings
Practice environment, carry out 3 Experiment Trainings and test in total, randomly chooses characteristic is concentrated 90% every time and be used as training set, remain
10% is remaininged as test set.Test result can indicate the effect of fault diagnosis by confusion matrix and whole accuracy rate three times,
Diagnostic result is shown in Table 2.
2 fault diagnosis result of table
Those skilled in the art of the present technique are it is understood that unless otherwise defined, all terms used herein (including skill
Art term and scientific terminology) there is meaning identical with the general understanding of the those of ordinary skill in fields of the present invention.Also
It should be understood that those terms such as defined in the general dictionary should be understood that with in the context of the prior art
The consistent meaning of meaning, and unless defined as here, will not be explained with the meaning of idealization or too formal.
Above-described specific implementation mode has carried out further the purpose of the present invention, technical solution and advantageous effect
It is described in detail, it should be understood that the foregoing is merely the specific implementation mode of the present invention, is not limited to this hair
Bright, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in the present invention
Protection domain within.
Claims (6)
1. a kind of hydraulic cylinder method for diagnosing faults, which is characterized in that include following specific steps:
Step 1) establishes simulation of hydraulic cylinder model by AMESim softwares;
Step 2) carries out direct fault location, obtains the disengaging of N group simulation of hydraulic cylinder models under each state of state set respectively
Mouthful data on flows, the state set include normal operating conditions, hydraulic cylinder leakage malfunction, load sudden change malfunction and
Piston rod line skew malfunction;
Step 3), using WAVELET PACKET DECOMPOSITION method to every group of inlet and outlet data on flows of simulation of hydraulic cylinder model under each state
Four layers of decomposition are carried out, obtain the decomposed signal under 16 frequency bands of every group of inlet and outlet data on flows, and calculate every group of inlet and outlet stream
It measures energy value of the decomposed signal of the inlet flow rate data of data, rate of discharge data under each frequency range and then obtains the group
The feature vector for importing and exporting data on flows, in conjunction with the state of the corresponding simulation of hydraulic cylinder model of each group inlet and outlet data on flows, most
End form is at characteristic data set;
Step 4), using genetic algorithm to the penalty factor of supporting vector machine model and preset kernel functional parameter into
Row optimizing, and multivalue support vector machine classifier is established on this basis, then utilize characteristic set pair multivalue supporting vector
Machine grader is trained, the multivalue support vector machine classifier after being trained;
Step 5) carries out feature extraction to hydraulic cylinder inlet and outlet data on flows to be tested, obtains its feature vector, and its is defeated
Enter the multivalue support vector machine classifier after training, obtain the state of hydraulic cylinder to be tested, completes fault diagnosis.
2. hydraulic cylinder method for diagnosing faults according to claim 1, which is characterized in that the detailed step of the step 1) is such as
Under:
Step 1.1) places servo amplifier, servo valve, hydraulic cylinder and position sensing under the draft mode of AMESim softwares
Device model forms the sketch of simulation of hydraulic cylinder model;
Step 1.2) is servo amplifier, servo valve, hydraulic cylinder and position sensor under the submodel pattern of AMESim softwares
Model chooses mathematical model;
Step 1.3) is respectively servo amplifier, servo valve, hydraulic cylinder and position sensing under the parameter model of AMESim softwares
Device model arrange parameter completes the setting of simulation of hydraulic cylinder model;
Step 1.4) runs the simulation of hydraulic cylinder model under the operational mode of AMESim softwares, until run time is more than in advance
If duration threshold value, to confirm that the simulation of hydraulic cylinder model running is normal.
3. hydraulic cylinder method for diagnosing faults according to claim 1, which is characterized in that the detailed step of the step 2) is such as
Under:
Step 2.1) runs simulation of hydraulic cylinder model under the operational mode of AMESim softwares, obtains under N group normal running conditions
The inlet and outlet data on flows of simulation of hydraulic cylinder model, N are the natural number more than 0;
Step 2.2) selects simulation of hydraulic cylinder model, change parameter leakage under the parameter mode of AMESim softwares
Coefficient reveals failure to form hydraulic cylinder, and simulation of hydraulic cylinder model is run under the operational mode of AMESim softwares,
Obtain the inlet and outlet data on flows of simulation of hydraulic cylinder model under N group hydraulic cylinders leakage failure;
Step 2.3) selects simulation of hydraulic cylinder model, change parameter total mass under the parameter mode of AMESim softwares
Being moved run simulation of hydraulic cylinder model to form load sudden change failure under the operational mode of AMESim softwares, obtain
Obtain the inlet and outlet data on flows of simulation of hydraulic cylinder model under N group load sudden change failures;
Step 2.4) selects simulation of hydraulic cylinder model, change parameter angle rod under the parameter mode of AMESim softwares
Makes with horizontal are run with forming piston rod line skew failure under the operational mode of AMESim softwares
Simulation of hydraulic cylinder model obtains the inlet and outlet data on flows of simulation of hydraulic cylinder model under N group piston rod line skew failures.
4. hydraulic cylinder method for diagnosing faults according to claim 1, which is characterized in that the detailed step of the step 3) is such as
Under:
Step 3.1), choose wavelet basis function system under normal running conditions, hydraulic cylinder leakage failure under, load sudden change failure,
Every group of inlet and outlet data on flows of simulation of hydraulic cylinder model carries out four layers of decomposition of discrete wavelet packet under piston rod line skew failure,
Obtain decomposed signal of the every group of inlet and outlet data on flows under 16 frequency bands;
Step 3.2) calculates separately the decomposed signal of inlet flow rate data, rate of discharge data in every group of inlet and outlet data on flows
Energy value under each frequency band, and be ranked up according to sequence from big to small;
Step 3.3) chooses the preceding k of its inlet flow rate data energy value under each frequency band for every group of inlet and outlet data on flows1
A, its rate of discharge data energy value under each frequency band preceding k2The vector of a composition is as its feature vector;
Step 3.4) imports and exports the feature vector of data on flows and the shape of its corresponding simulation of hydraulic cylinder model according to each group
State forms characteristic data set.
5. a kind of hydraulic cylinder method for diagnosing faults according to claim 1, which is characterized in that the detailed step of the step 4)
It is rapid as follows:
Step 4.1), is arranged the parameter initialization value of genetic algorithm, and the parameter initialization value includes evolutionary generation, population rule
Mould, select probability, crossover probability, mutation probability, fitness function and individual optimizing space;
Step 4.2) is being sought using the penalty factor of supporting vector machine model and preset kernel functional parameter as population at individual
Initial population is randomly generated in excellent space;
Step 4.3) iteratively solves optimum model parameter, by selection operation, crossover operation and mutation operation, in conjunction with fitness
Function obtains parameter optimal value;
Step 4.4) builds multivalue support vector machine classifier according to parameter optimal value is obtained, and is concentrated from the characteristic
90% data are randomly choosed to instruct the multivalue support vector machine classifier as test set as training set, residue 10%
Practice;
Step 4.5) repeats step 4.1) to step 4.4), until the fault diagnosis accuracy rate of test set is accurate more than preset
Rate threshold value.
6. hydraulic cylinder method for diagnosing faults according to claim 1, which is characterized in that described advance in the step 4)
The kernel functional parameter set is gaussian kernel function parameter, and penalty factor=16 of supporting vector machine model, gaussian kernel function ginseng
Number γ=3.0314.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810441856.0A CN108363896B (en) | 2018-05-10 | 2018-05-10 | Fault diagnosis method for hydraulic cylinder |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810441856.0A CN108363896B (en) | 2018-05-10 | 2018-05-10 | Fault diagnosis method for hydraulic cylinder |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108363896A true CN108363896A (en) | 2018-08-03 |
CN108363896B CN108363896B (en) | 2021-12-21 |
Family
ID=63012215
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810441856.0A Active CN108363896B (en) | 2018-05-10 | 2018-05-10 | Fault diagnosis method for hydraulic cylinder |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108363896B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109783959A (en) * | 2019-01-23 | 2019-05-21 | 保定市博宏高科控制技术有限公司 | Joint multi-model emulation platform and method based on valve |
CN110135066A (en) * | 2019-05-15 | 2019-08-16 | 北京交通大学 | A kind of method for diagnosing faults of dynamic power shift gear box pilot operated compound relief valve |
CN111753465A (en) * | 2020-06-05 | 2020-10-09 | 长江三峡通航管理局 | Method for diagnosing leakage fault in buffering oil cylinder of anti-collision buffering hydraulic system of ship lift |
CN112036456A (en) * | 2020-08-19 | 2020-12-04 | 阳光电源股份有限公司 | Photovoltaic fault data generation method and device and computer readable storage medium |
CN112650146A (en) * | 2020-12-18 | 2021-04-13 | 国家机床质量监督检验中心 | Fault diagnosis optimization method, system and equipment of numerical control machine tool under multiple working conditions |
CN112861437A (en) * | 2021-02-22 | 2021-05-28 | 四川大学 | Fault diagnosis method and device for aileron actuator based on AMESim model |
CN112926400A (en) * | 2021-01-29 | 2021-06-08 | 华南理工大学 | Intelligent diagnosis method and system for leakage fault in hydraulic cylinder based on data driving |
CN113656985A (en) * | 2021-09-01 | 2021-11-16 | 上海仁童电子科技有限公司 | Modeling method, fault diagnosis method and device for valve elements in brake system |
CN116150676A (en) * | 2023-04-19 | 2023-05-23 | 山东能源数智云科技有限公司 | Equipment fault diagnosis and identification method and device based on artificial intelligence |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7106903B2 (en) * | 2001-09-24 | 2006-09-12 | Vima Technologies, Inc. | Dynamic partial function in measurement of similarity of objects |
CN101122585A (en) * | 2007-09-12 | 2008-02-13 | 天津大学 | Automatic identification method for supersonic phased array for detecting oil gas pipeline girth weld defect type |
CN104061208A (en) * | 2014-07-02 | 2014-09-24 | 北京机械设备研究所 | Online fault diagnosis method for hydraulic system |
CN106053067A (en) * | 2016-05-24 | 2016-10-26 | 广东石油化工学院 | Bearing fault diagnosis method based on quantum genetic algorithm optimized support vector machine |
CN106682298A (en) * | 2016-12-22 | 2017-05-17 | 西北工业大学 | Aviation hydraulic steering engine system fault simulation model base building method |
CN107255785A (en) * | 2017-04-28 | 2017-10-17 | 南京邮电大学 | Based on the analog-circuit fault diagnosis method for improving mRMR |
CN107291997A (en) * | 2017-05-31 | 2017-10-24 | 南京航空航天大学 | A kind of cold rolling hydraulic AGC system design of Fault Diagnosis Strategy method |
-
2018
- 2018-05-10 CN CN201810441856.0A patent/CN108363896B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7106903B2 (en) * | 2001-09-24 | 2006-09-12 | Vima Technologies, Inc. | Dynamic partial function in measurement of similarity of objects |
CN101122585A (en) * | 2007-09-12 | 2008-02-13 | 天津大学 | Automatic identification method for supersonic phased array for detecting oil gas pipeline girth weld defect type |
CN104061208A (en) * | 2014-07-02 | 2014-09-24 | 北京机械设备研究所 | Online fault diagnosis method for hydraulic system |
CN106053067A (en) * | 2016-05-24 | 2016-10-26 | 广东石油化工学院 | Bearing fault diagnosis method based on quantum genetic algorithm optimized support vector machine |
CN106682298A (en) * | 2016-12-22 | 2017-05-17 | 西北工业大学 | Aviation hydraulic steering engine system fault simulation model base building method |
CN107255785A (en) * | 2017-04-28 | 2017-10-17 | 南京邮电大学 | Based on the analog-circuit fault diagnosis method for improving mRMR |
CN107291997A (en) * | 2017-05-31 | 2017-10-24 | 南京航空航天大学 | A kind of cold rolling hydraulic AGC system design of Fault Diagnosis Strategy method |
Non-Patent Citations (2)
Title |
---|
曹斌 等: "基于小波包与遗传算法和支持向量机的液压泵故障诊断", 《组合机床与自动化加工技术》 * |
李刚 等: "基于小波分析的电力***故障时空检测与诊断", 《***仿真学报》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109783959B (en) * | 2019-01-23 | 2023-10-10 | 保定市博宏高科控制技术有限公司 | Valve-based combined multi-model simulation platform and method |
CN109783959A (en) * | 2019-01-23 | 2019-05-21 | 保定市博宏高科控制技术有限公司 | Joint multi-model emulation platform and method based on valve |
CN110135066A (en) * | 2019-05-15 | 2019-08-16 | 北京交通大学 | A kind of method for diagnosing faults of dynamic power shift gear box pilot operated compound relief valve |
CN110135066B (en) * | 2019-05-15 | 2021-04-13 | 北京交通大学 | Fault diagnosis method for pilot type overflow valve of power shifting gearbox |
CN111753465A (en) * | 2020-06-05 | 2020-10-09 | 长江三峡通航管理局 | Method for diagnosing leakage fault in buffering oil cylinder of anti-collision buffering hydraulic system of ship lift |
CN111753465B (en) * | 2020-06-05 | 2022-08-16 | 长江三峡通航管理局 | Method for diagnosing leakage fault in buffering oil cylinder of anti-collision buffering hydraulic system of ship lift |
CN112036456A (en) * | 2020-08-19 | 2020-12-04 | 阳光电源股份有限公司 | Photovoltaic fault data generation method and device and computer readable storage medium |
CN112650146A (en) * | 2020-12-18 | 2021-04-13 | 国家机床质量监督检验中心 | Fault diagnosis optimization method, system and equipment of numerical control machine tool under multiple working conditions |
CN112650146B (en) * | 2020-12-18 | 2021-11-19 | 国家机床质量监督检验中心 | Fault diagnosis optimization method, system and equipment of numerical control machine tool under multiple working conditions |
CN112926400A (en) * | 2021-01-29 | 2021-06-08 | 华南理工大学 | Intelligent diagnosis method and system for leakage fault in hydraulic cylinder based on data driving |
CN112861437A (en) * | 2021-02-22 | 2021-05-28 | 四川大学 | Fault diagnosis method and device for aileron actuator based on AMESim model |
CN112861437B (en) * | 2021-02-22 | 2022-05-03 | 四川大学 | Fault diagnosis method and device for aileron actuator based on AMESim model |
CN113656985A (en) * | 2021-09-01 | 2021-11-16 | 上海仁童电子科技有限公司 | Modeling method, fault diagnosis method and device for valve elements in brake system |
CN113656985B (en) * | 2021-09-01 | 2024-03-15 | 上海仁童电子科技有限公司 | Modeling method, fault diagnosis method and device for valve element in braking system |
CN116150676B (en) * | 2023-04-19 | 2023-09-26 | 山东能源数智云科技有限公司 | Equipment fault diagnosis and identification method and device based on artificial intelligence |
CN116150676A (en) * | 2023-04-19 | 2023-05-23 | 山东能源数智云科技有限公司 | Equipment fault diagnosis and identification method and device based on artificial intelligence |
Also Published As
Publication number | Publication date |
---|---|
CN108363896B (en) | 2021-12-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108363896A (en) | A kind of hydraulic cylinder method for diagnosing faults | |
CN104537415B (en) | A kind of non-linear process industry failure predication and recognition methodss based on compressed sensing and DROS ELM | |
CN102707256B (en) | Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter | |
CN101587155B (en) | Oil soaked transformer fault diagnosis method | |
CN103810328B (en) | Transformer maintenance decision method based on hybrid model | |
CN108197648A (en) | A kind of Fault Diagnosis Method of Hydro-generating Unit and system based on LSTM deep learning models | |
CN103886405B (en) | Boiler combustion condition identification method based on information entropy characteristics and probability nerve network | |
CN103106535B (en) | Method for solving collaborative filtering recommendation data sparsity based on neural network | |
CN105354587A (en) | Fault diagnosis method for gearbox of wind generation unit | |
CN103914064A (en) | Industrial process fault diagnosis method based on multiple classifiers and D-S evidence fusion | |
CN107688825A (en) | A kind of follow-on integrated weighting extreme learning machine sewage disposal failure examines method | |
CN103927412A (en) | Real-time learning debutanizer soft measurement modeling method on basis of Gaussian mixture models | |
CN102606557A (en) | Health evaluation method for hydraulic system based on fault observer and SOM (self-organized mapping) | |
CN107392304A (en) | A kind of Wind turbines disorder data recognition method and device | |
CN105574589B (en) | Transformer oil chromatographic method for diagnosing faults based on niche genetic algorithm | |
CN107704962A (en) | A kind of smelter steam flow interval prediction method based on imperfect time series data collection | |
CN107291830A (en) | A kind of creation method of equipment knowledge base | |
CN107886193A (en) | A kind of time sequence forecasting method based on grey wolf optimization echo state network | |
CN108563875A (en) | Analog circuit measuring point and frequency based on multiple-objection optimization combine preferred method | |
CN109634108A (en) | The different factor full format non-model control method of the MIMO of parameter self-tuning | |
CN104218571B (en) | A kind of running status appraisal procedure of wind power plant | |
CN106570562A (en) | Adaptive-DE-algorithm-based fuzzy modeling method for bridge crane | |
CN107656152A (en) | One kind is based on GA SVM BP Diagnosis Method of Transformer Faults | |
CN104634265B (en) | A kind of mineral floating froth bed soft measurement method of thickness based on multiplex images Fusion Features | |
CN107578100A (en) | The result feedback method and device of a kind of Kernel-based methods parameter |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |