CN110378461A - Error self-adaptation control method and system based on space-time error separate - Google Patents
Error self-adaptation control method and system based on space-time error separate Download PDFInfo
- Publication number
- CN110378461A CN110378461A CN201910516535.7A CN201910516535A CN110378461A CN 110378461 A CN110378461 A CN 110378461A CN 201910516535 A CN201910516535 A CN 201910516535A CN 110378461 A CN110378461 A CN 110378461A
- Authority
- CN
- China
- Prior art keywords
- error
- space
- model
- fuzzy
- vector machine
- 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
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- 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
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Automation & Control Theory (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Numerical Control (AREA)
Abstract
The present invention provides a kind of error self-adaptation control methods and system based on space-time error separate, comprising: acquires the mismachining tolerance of workpiece in previous process, Working position data;Collected data are standardized and be blurred, the training set for being used for model of fit is formed;Construct the quadratic programming problem of fuzzy support vector machine model;The parameter and fitness function of particle swarm optimization algorithm are set;The unknown parameter in fuzzy support vector machine model is recognized using subgroup optimization algorithm, the regression model of optimal estimation fuzzy support vector machine model, the output quantity for being fitted obtained regression model is the error of current processing grid, characterizes the spatially-correlated errors at different Working positions;Carry out compensation space correlated error using the method that offline machining path is modified, is compensated by the way of real-time compensation.The present invention compensates the spatially-correlated errors at Working position by way of modifying cutter track offline, can effectively reduce real-time compensation amount, guarantees the validity of compensation.
Description
Technical field
The present invention relates to intelligence manufacture fields, and in particular, to the error self adaptive control side based on space-time error separate
Method and system can be applied in the deformation-compensated processing of aerospace large thin-wall element mirror image milling, guarantee the final of part
Machining accuracy.
Background technique
Large thin-wall element machining accuracy would generally be influenced by various factors, such as form error, the dress of workpiece blank
Press from both sides deformation caused by error and cutting force.Due to " extreme " weak rigidity of large thin-wall element, these factors are usually that can not keep away
Exempt from and at random occurs.
Weak hard parts compensation working research is broadly divided into three classes at present:
First kind method is to establish the prediction mould of cutting force and deflection based on cutting Force Model and finite element method
Type is before processing adjusted knife rail, to achieve the purpose that error compensation.Such method has been successfully applied to small-sized thin-walled
The processing of part, processing compensation precision depend on the levels of precision of finite element model.However, the boundary condition of large thin-wall element is opposite
Increasingly complex for small-sized thin-wall part, the precision of finite element model is difficult to ensure.In addition to this, large thin-wall element is processed
There are enchancement factor in journey, the Accurate Prediction of part deformation is still problem, and such method mainly for be only to cut
Power deformation.Such as a kind of " prediction of Turning Force with Artificial and temperature for end mill cutting disclosed in the patent of publication number CN104268343B
Spend the method for prediction ".
Second class method introduces process by the way of " off-line measurement+cutter track adjustment ", by measurement, to data into
Cutter path is adjusted offline on the basis of row modeling, error evaluation analysis.Cutting output and machining deformation when due to thin-wall part processing
There are coupling effect, the adjustment of cutter path can change cutting data and then influence deflection, need repeatedly to shut down measurement to protect
It demonstrate,proves revised cutter track and meets required precision, but is time-consuming often relatively more long, processing efficiency is difficult to ensure.
Third class method carries out real-time measurement for cutting force etc. and processing quality correlative, and is mended in real time accordingly
It repays, can theoretically compensate all types of errors.On the one hand efficiency can be improved in such method, on the other hand can significantly reduce
Influence of the uncertain factors such as the machining deformation in the position error and cutting process of process unit to machining accuracy has more
High compensation precision.The big cutting force as caused by " extreme " weak rigidity and highly-efficient processing of large thin-wall element, large thin-wall element
It easily deforms in mirror image milling process, and deflection is larger.In processing on real-time, the time of each cutter location compensation processing has
Limit, is limited to the delay of the response frequency and data transmission, compensation value calculation etc. of compensation device (kinematic axis of lathe), depends merely on reality
When compensation method its machining deformation can not be fully compensated.
Summary of the invention
For the defects in the prior art, the object of the present invention is to provide a kind of error based on space-time error separate is adaptive
Answer control method and system.
A kind of error self-adaptation control method based on space-time error separate provided according to the present invention, comprising:
Data collection steps: the mismachining tolerance of workpiece in previous process, Working position data are acquired;
Training set forming step: standardizing and is blurred collected data, forms the training set for being used for model of fit;
Quadratic programming step: the quadratic programming problem of construction fuzzy support vector machine model;
Optimization algorithm setting steps: the parameter and fitness function of particle swarm optimization algorithm are set;
Optimization Steps: the unknown parameter in fuzzy support vector machine model is recognized using subgroup optimization algorithm, most
The regression model of excellent ambiguous estimation supporting vector machine model, the output quantity for the regression model being fitted are current processing grid
Error characterizes the spatially-correlated errors at different Working positions;
Error compensation step: carrying out compensation space correlated error using the method that offline machining path is modified, in process,
It is compensated by the way of real-time compensation.
Preferably, the data collection steps further include: denoised using data of the wavelet package transforms to acquisition.
Preferably, the fuzzy support vector machine Construction of A Model step includes:
The ambiguity of sample data is characterized using symmetric extension Triangular Fuzzy Number;For one group of training setx
It is Working position coordinate, y Representative errors, subscript i indicates i-th group of training set, and the total n group of training set is retouched in triangle fuzzy space
It states are as follows:WithWhereinθ is fuzzy coefficient;
Regression model f (x)=w φ (x)+b is indicated in triangle fuzzy space are as follows:
F (x)=(w φ (rx)+b, ρ (Δ rx))
Wherein, w, rx,Δrx∈Rn, b ∈ R, ρ (Δ rx)=| w | Δ rx;
R is real number field;
The quadratic programming problem of fuzzy support vector machine model can construct as follows:
s.t.
Wherein,
By solution quadratic programming problem, Lagrange multiplier α is obtained, by meeting KKT condition, solution obtains parameter b:
Wherein,
Obtain the regression equation based on fuzzy support vector machine model:
Preferably, the fitness function is made of fitting precision and spatial autocorrelation indicators Local Geary ' s C, is used
In guaranteeing that error that model is estimated has spatial auto-correlation.
Preferably, Local Geary ' s C is indicated at the i of position are as follows:
Wherein, n is all positional numbers, fiIt is the error amount in position i, wijIt is the distance between position i and position j power,
It represents the degree that influences each other between position i and position j;Coefficient wijCalculation formula are as follows:
Wherein, dijThe space length between position i and position j is represented,If k=3;According to definition,
The absolute value of Local Geary ' s C is smaller to show that data positive space autocorrelation is stronger.
A kind of error adaptive control system based on space-time error separate provided according to the present invention, comprising:
Data acquisition module: the mismachining tolerance of workpiece in previous process, Working position data are acquired;
Training set forms module: standardizing and is blurred collected data, form the training set for being used for model of fit;
Quadratic programming module: the quadratic programming problem of construction fuzzy support vector machine model;
Optimization algorithm setup module: the parameter and fitness function of particle swarm optimization algorithm are set;
Optimization module: the unknown parameter in fuzzy support vector machine model is recognized using subgroup optimization algorithm, most
The regression model of excellent ambiguous estimation supporting vector machine model, the output quantity for the regression model being fitted are current processing grid
Error characterizes the spatially-correlated errors at different Working positions;
Error compensation module: carrying out compensation space correlated error using the method that offline machining path is modified, in process,
It is compensated by the way of real-time compensation.
Preferably, the data acquisition module further include: denoised using data of the wavelet package transforms to acquisition.
Preferably, the fuzzy support vector machine model construction module includes:
The ambiguity of sample data is characterized using symmetric extension Triangular Fuzzy Number;For one group of training setx
It is Working position coordinate, y Representative errors, subscript i indicates i-th group of training set, and the total n group of training set is retouched in triangle fuzzy space
It states are as follows:WithWhereinθ is fuzzy coefficient;
Regression model f (x)=w φ (x)+b is indicated in triangle fuzzy space are as follows:
F (x)=(w φ (rx)+b, ρ (Δ rx))
Wherein, w, rx,Δrx∈Rn, b ∈ R, ρ (Δ rx)=| w | Δ rx;
R is real number field;
The quadratic programming problem of fuzzy support vector machine model can construct as follows:
s.t.
Wherein,
By solution quadratic programming problem, Lagrange multiplier α is obtained, by meeting KKT condition, solution obtains parameter b:
Wherein,
Obtain the regression equation based on fuzzy support vector machine model:
Preferably, the fitness function is made of fitting precision and spatial autocorrelation indicators Local Geary ' s C, is used
In guaranteeing that error that model is estimated has spatial auto-correlation.
Preferably, Local Geary ' s C is indicated at the i of position are as follows:
Wherein, n is all positional numbers, fiIt is the error amount in position i, wijIt is the distance between position i and position j power,
It represents the degree that influences each other between position i and position j;Coefficient wijCalculation formula are as follows:
Wherein, dijThe space length between position i and position j is represented,If k=3;According to definition,
The absolute value of Local Geary ' s C is smaller to show that data positive space autocorrelation is stronger.
Compared with prior art, the present invention have it is following the utility model has the advantages that
The present invention compensates the spatially-correlated errors at Working position by way of modifying cutter track offline, can effectively reduce
Real-time compensation amount in finishing guarantees the validity of compensation.
Detailed description of the invention
Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention,
Objects and advantages will become more apparent upon:
Fig. 1 is flow diagram of the invention;
Fig. 2 is the Real-time Error tracking schematic diagram based on laser displacement sensor;
Fig. 3 is large thin-wall slab geomitry scale diagrams;
Fig. 4 is the autocorrelation exponent schematic diagram of error information in square net rough milling;
Fig. 5 is error separate and compensation flow diagram;
Fig. 6 is error compensation effect diagram.
Specific embodiment
The present invention is described in detail combined with specific embodiments below.Following embodiment will be helpful to the technology of this field
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill of this field
For personnel, without departing from the inventive concept of the premise, several changes and improvements can also be made.These belong to the present invention
Protection scope.
As shown in Figure 1, a kind of error self-adaptation control method based on space-time error separate provided by the invention, comprising:
Data collection steps: the mismachining tolerance of workpiece in previous process, Working position data are acquired;
Training set forming step: standardizing and is blurred collected data, forms the training set for being used for model of fit;
Quadratic programming step: the quadratic programming problem of construction fuzzy support vector machine model;
Optimization algorithm setting steps: the parameter and fitness function of particle swarm optimization algorithm are set;
Optimization Steps: the unknown parameter in fuzzy support vector machine model is recognized using subgroup optimization algorithm, most
The regression model of excellent ambiguous estimation supporting vector machine model, the output quantity for the regression model being fitted are current processing grid
Error characterizes the spatially-correlated errors at different Working positions;
Error compensation step: carrying out compensation space correlated error using the method that offline machining path is modified, in process,
It is compensated by the way of real-time compensation.
Embodiment
Step 1, before input process acquire mismachining tolerance, Working position data, pass through the side of wavelet package transforms
Formula denoises the data of acquisition, and process data can be measured by laser sensor, as shown in Fig. 2, large size to be processed
Thin-wall part size is as shown in Figure 3;
Step 2, the data for standardizing and being blurred acquisition, form the training set for being used for model of fitAltogether
There are n group data.Wherein, x is Working position coordinate, y Representative errors.
Step 3, the quadratic programming problem of fuzzy support vector machine model is constructed;
Step 4, the parameter and fitness function of particle swarm optimization algorithm are set.Fitness function is by fitting precision and space
Autocorrelation exponent Local Geary ' s C is constituted, for guaranteeing that the error that model is estimated has spatial auto-correlation.It adapts to
Degree function is made of fitting precision and spatial autocorrelation indicators Local Geary ' s C, the mistake estimated for guaranteeing model
Difference has spatial auto-correlation.Processed process space autocorrelation exponent is as shown in Figure 4.
Step 5, customized penalty factor, regularization parameter ν and gaussian radial basis function letter are obtained by particle swarm optimization algorithm
Number standard deviation sigma, regression model of the optimal estimation based on fuzzy support vector machine.The input quantity of regression model is current processing grid
Error information and corresponding Working position.The output quantity for being fitted obtained regression model is the error of current processing grid,
It characterizes the spatially-correlated errors at different Working positions;Fuzzy support vector machine unknown-model parameter mainly includes customized
Penalty factor, regularization parameter ν, the standard deviation sigma and mould of the kernel function Gaussian radial basis function of fuzzy support vector machine model
Paste coefficient θ.Parameter C, ν and σ are recognized using particle swarm optimization algorithm, particle swarm optimization algorithm parameter includes population
Scale, particle size, inertial factor, maximum number of iterations and two positive acceleration constants c1 and c2.Parameter θ is according to processed
What the noise level in journey determined.
Parameter | Setting method |
Customized penalty factor | Particle swarm optimization algorithm identification |
Regularization parameter ν | Particle swarm optimization algorithm identification |
The standard deviation sigma of kernel function Gaussian radial basis function | Particle swarm optimization algorithm identification |
Fuzzy coefficient θ | It is arranged according to processing environment, range is 0 to 1 |
Step 6, Regression Model Simulator obtains spatially-correlated errors, the difference quilt of the spatially-correlated errors of initial error and extraction
It is defined as error with relation to time.Carry out compensation space correlated error using the method that offline machining path is modified, in process, adopts
It is compensated with the mode of real-time compensation.Error separate that this method is proposed and compensation process are as shown in figure 5, error after compensation
Uniformly ideal, final precision is less than 0.05mm, as shown in Figure 6.
The quadratic programming problem of step 3 fuzzy support vector machine model includes the following steps:
Step 3.1, the ambiguity of sample data is characterized using symmetric extension Triangular Fuzzy Number.For one group of training setIt can be described as in triangle fuzzy space:With yi=(ryi,Δryi), wherein θ is fuzzy coefficient.Based on this, regression model f (x)=w φ (x)+b is in triangle fuzzy space
It indicates are as follows:
F (x)=(w φ (rx)+b,ρ(Δrx)) (1)
Wherein, w, rx,Δrx∈Rn, b ∈ R, ρ (Δ rx)=| w | Δ rx。
Step 3.2, theoretical based on fuzzy support vector machine, the parameter of above-mentioned regression model can pass through following constrained optimization
The solution of problem is estimated to obtain:
s.t.
Wherein, C is a customized penalty factor,It is slack variable, and v ∈ (0,1] then
It is an adjustable regularization parameter.It is a quadratic programming problem on above-mentioned constrained optimization question essence, by introducing Lagrange
Its Lagrange's equation can be obtained in multiplier method, next, willProblem is converted into dual problem
To solve dual problem, L (w, b, ε, ξ are allowed(*),α(*),β,η(*)) minimized about w and b, i.e. L (w, b, ε, ξ(*),α(*),β,
η(*)) to w, b, ε and ξki *Local derviation be 0, the dual problem of available equation (2):
s.t.
Wherein,
By solving the quadratic programming problem of above formula, available Lagrange multiplier α.Notice that dual problem exists not
Equation, it is therefore desirable to meet following KKT (Karush-Kuhn-Tucker) condition.By meeting KKT condition, parameter b can be in the hope of
Solution obtains:
Wherein,
Therefore, the regression equation based on fuzzy support vector machine model can obtain:
Preferably, in the step 4, the calculation method of Local Geary ' s C spatial autocorrelation coefficient specifically:
Local spatial statistics analysis is by the spatial autocorrelation between evaluation current data and all space interval data
Property compared with global analysis's method, can more accurately extract spatially-correlated errors come what is carried out.In all local spatial statistics
In analysis method, Local Geary ' s C is widely applied because it is sensitive to spatial autocorrelation result.According to definition,
Local Geary ' s C can be indicated at the i of position are as follows:
Wherein, n is all positional numbers, fiIt is the error amount in position i, wijIt is the distance between position i and position j power,
It represents the degree that influences each other between position i and position j.Coefficient wijCalculation formula are as follows:
Wherein, dijThe space length between position i and position j is represented,K=3 is set in the method.Root
According to definition, the absolute value of Local Geary ' s C is smaller to show that data positive space autocorrelation is stronger.
Fitness function is made of fitting precision and spatial autocorrelation indicators Local Geary ' s C, for guaranteeing estimation
Obtained error has spatial auto-correlation.
On the basis of a kind of above-mentioned error self-adaptation control method based on space-time error separate, the present invention also provides one
Error adaptive control system of the kind based on space-time error separate, comprising:
Data acquisition module: the mismachining tolerance of workpiece in previous process, Working position data are acquired;
Training set forms module: standardizing and is blurred collected data, form the training set for being used for model of fit;
Quadratic programming module: the quadratic programming problem of construction fuzzy support vector machine model;
Optimization algorithm setup module: the parameter and fitness function of particle swarm optimization algorithm are set;
Optimization module: the unknown parameter in fuzzy support vector machine model is recognized using subgroup optimization algorithm, most
The regression model of excellent ambiguous estimation supporting vector machine model, the output quantity for the regression model being fitted are current processing grid
Error characterizes the spatially-correlated errors at different Working positions;
Error compensation module: carrying out compensation space correlated error using the method that offline machining path is modified, in process,
It is compensated by the way of real-time compensation.
One skilled in the art will appreciate that in addition to realizing system provided by the invention in a manner of pure computer readable program code
It, completely can be by the way that method and step be carried out programming in logic come so that the present invention provides and its other than each device, module, unit
System and its each device, module, unit with logic gate, switch, specific integrated circuit, programmable logic controller (PLC) and embedding
Enter the form of the controller that declines etc. to realize identical function.So system provided by the invention and its every device, module, list
Member is considered a kind of hardware component, and to include in it can also for realizing the device of various functions, module, unit
To be considered as the structure in hardware component;It can also will be considered as realizing the device of various functions, module, unit either real
The software module of existing method can be the structure in hardware component again.
In the description of the present application, it is to be understood that term " on ", "front", "rear", "left", "right", " is erected at "lower"
Directly ", the orientation or positional relationship of the instructions such as "horizontal", "top", "bottom", "inner", "outside" is orientation based on the figure or position
Relationship is set, description the application is merely for convenience of and simplifies description, rather than the device or element of indication or suggestion meaning are necessary
It with specific orientation, is constructed and operated in a specific orientation, therefore should not be understood as the limitation to the application.
Specific embodiments of the present invention are described above.It is to be appreciated that the invention is not limited to above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or modify within the scope of the claims, this not shadow
Ring substantive content of the invention.In the absence of conflict, the feature in embodiments herein and embodiment can any phase
Mutually combination.
Claims (10)
1. a kind of error self-adaptation control method based on space-time error separate characterized by comprising
Data collection steps: the mismachining tolerance of workpiece in previous process, Working position data are acquired;
Training set forming step: standardizing and is blurred collected data, forms the training set for being used for model of fit;
Quadratic programming step: the quadratic programming problem of construction fuzzy support vector machine model;
Optimization algorithm setting steps: the parameter and fitness function of particle swarm optimization algorithm are set;
Optimization Steps: recognizing the unknown parameter in fuzzy support vector machine model using subgroup optimization algorithm, optimal to estimate
The regression model of fuzzy support vector machine model is counted, the output quantity for the regression model being fitted is the mistake of current processing grid
Difference characterizes the spatially-correlated errors at different Working positions;
Error compensation step: carrying out compensation space correlated error using the method that offline machining path is modified, and in process, uses
The mode of real-time compensation compensates.
2. the error self-adaptation control method according to claim 1 based on space-time error separate, which is characterized in that described
Data collection steps further include: denoised using data of the wavelet package transforms to acquisition.
3. the error self-adaptation control method according to claim 1 based on space-time error separate, which is characterized in that described
Fuzzy support vector machine Construction of A Model step includes:
The ambiguity of sample data is characterized using symmetric extension Triangular Fuzzy Number;For one group of training setX is to add
Work position coordinates, y Representative errors, subscript i indicate i-th group of training set, and the total n group of training set describes in triangle fuzzy space are as follows:WithWhereinθ is fuzzy coefficient;
Regression model f (x)=w φ (x)+b is indicated in triangle fuzzy space are as follows:
F (x)=(w φ (rx)+b,ρ(Δrx))
Wherein, w, rx,Δrx∈Rn, b ∈ R, ρ (Δ rx)=| w | Δ rx;
R is real number field;
The quadratic programming problem of fuzzy support vector machine model can construct as follows:
s.t.
Wherein,
By solution quadratic programming problem, Lagrange multiplier α is obtained, by meeting KKT condition, solution obtains parameter b:
Wherein, αki,
Obtain the regression equation based on fuzzy support vector machine model:
4. the error self-adaptation control method according to claim 1 based on space-time error separate, which is characterized in that described
Fitness function is made of fitting precision and spatial autocorrelation indicators Local Geary ' s C, for guaranteeing that model is estimated to obtain
Error have spatial auto-correlation.
5. the error self-adaptation control method according to claim 4 based on space-time error separate, which is characterized in that
Local Geary ' s C is indicated at the i of position are as follows:
Wherein, n is all positional numbers, fiIt is the error amount in position i, wijIt is the distance between position i and position j power, generation
Epitope sets the degree that influences each other between i and position j;Coefficient wijCalculation formula are as follows:
Wherein, dijThe space length between position i and position j is represented,If k=3;According to definition, Local
The absolute value of Geary ' s C is smaller to show that data positive space autocorrelation is stronger.
6. a kind of error adaptive control system based on space-time error separate characterized by comprising
Data acquisition module: the mismachining tolerance of workpiece in previous process, Working position data are acquired;
Training set forms module: standardizing and is blurred collected data, form the training set for being used for model of fit;
Quadratic programming module: the quadratic programming problem of construction fuzzy support vector machine model;
Optimization algorithm setup module: the parameter and fitness function of particle swarm optimization algorithm are set;
Optimization module: recognizing the unknown parameter in fuzzy support vector machine model using subgroup optimization algorithm, optimal to estimate
The regression model of fuzzy support vector machine model is counted, the output quantity for the regression model being fitted is the mistake of current processing grid
Difference characterizes the spatially-correlated errors at different Working positions;
Error compensation module: carrying out compensation space correlated error using the method that offline machining path is modified, and in process, uses
The mode of real-time compensation compensates.
7. the error adaptive control system according to claim 6 based on space-time error separate, which is characterized in that described
Data acquisition module further include: denoised using data of the wavelet package transforms to acquisition.
8. the error adaptive control system according to claim 6 based on space-time error separate, which is characterized in that described
Fuzzy support vector machine model construction module includes:
The ambiguity of sample data is characterized using symmetric extension Triangular Fuzzy Number;For one group of training setX is to add
Work position coordinates, y Representative errors, subscript i indicate i-th group of training set, and the total n group of training set describes in triangle fuzzy space are as follows:WithWhereinθ is fuzzy coefficient;
Regression model f (x)=w φ (x)+b is indicated in triangle fuzzy space are as follows:
F (x)=(w φ (rx)+b,ρ(Δrx))
Wherein, w, rx,Δrx∈Rn, b ∈ R, ρ (Δ rx)=| w | Δ rx;
R is real number field;
The quadratic programming problem of fuzzy support vector machine model can construct as follows:
s.t.
Wherein,
By solution quadratic programming problem, Lagrange multiplier α is obtained, by meeting KKT condition, solution obtains parameter b:
Wherein, αki,
Obtain the regression equation based on fuzzy support vector machine model:
9. the error adaptive control system according to claim 6 based on space-time error separate, which is characterized in that described
Fitness function is made of fitting precision and spatial autocorrelation indicators Local Geary ' s C, for guaranteeing that model is estimated to obtain
Error have spatial auto-correlation.
10. the error adaptive control system according to claim 9 based on space-time error separate, which is characterized in that
Local Geary ' s C is indicated at the i of position are as follows:
Wherein, n is all positional numbers, fiIt is the error amount in position i, wijIt is the distance between position i and position j power, generation
Epitope sets the degree that influences each other between i and position j;Coefficient wijCalculation formula are as follows:
Wherein, dijThe space length between position i and position j is represented,If k=3;According to definition, Local
The absolute value of Geary ' s C is smaller to show that data positive space autocorrelation is stronger.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910516535.7A CN110378461A (en) | 2019-06-14 | 2019-06-14 | Error self-adaptation control method and system based on space-time error separate |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910516535.7A CN110378461A (en) | 2019-06-14 | 2019-06-14 | Error self-adaptation control method and system based on space-time error separate |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110378461A true CN110378461A (en) | 2019-10-25 |
Family
ID=68250363
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910516535.7A Pending CN110378461A (en) | 2019-06-14 | 2019-06-14 | Error self-adaptation control method and system based on space-time error separate |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110378461A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103308292A (en) * | 2013-05-16 | 2013-09-18 | 国家电网公司 | Vacuum breaker mechanical state detecting method based on vibration signal analysis |
CN103675011A (en) * | 2013-09-22 | 2014-03-26 | 浙江大学 | Soft industrial melt index measurement instrument and method of optimal support vector machine |
CN103675006A (en) * | 2013-09-22 | 2014-03-26 | 浙江大学 | Least-squares-based industrial melt index soft measuring meter and method |
CN105046346A (en) * | 2015-06-19 | 2015-11-11 | 华侨大学 | Method of using maximum relative error minimum multi-input multi-output support vector regression machine to carry out concrete multicomponent anti-prediction |
CN105574231A (en) * | 2015-11-27 | 2016-05-11 | 上海交通大学 | Storage battery surplus capacity detection method |
CN109470477A (en) * | 2018-09-27 | 2019-03-15 | 昆明理工大学 | A kind of Fault Diagnosis of Roller Bearings based on improved PSO algorithm optimization FSVM |
-
2019
- 2019-06-14 CN CN201910516535.7A patent/CN110378461A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103308292A (en) * | 2013-05-16 | 2013-09-18 | 国家电网公司 | Vacuum breaker mechanical state detecting method based on vibration signal analysis |
CN103675011A (en) * | 2013-09-22 | 2014-03-26 | 浙江大学 | Soft industrial melt index measurement instrument and method of optimal support vector machine |
CN103675006A (en) * | 2013-09-22 | 2014-03-26 | 浙江大学 | Least-squares-based industrial melt index soft measuring meter and method |
CN105046346A (en) * | 2015-06-19 | 2015-11-11 | 华侨大学 | Method of using maximum relative error minimum multi-input multi-output support vector regression machine to carry out concrete multicomponent anti-prediction |
CN105574231A (en) * | 2015-11-27 | 2016-05-11 | 上海交通大学 | Storage battery surplus capacity detection method |
CN109470477A (en) * | 2018-09-27 | 2019-03-15 | 昆明理工大学 | A kind of Fault Diagnosis of Roller Bearings based on improved PSO algorithm optimization FSVM |
Non-Patent Citations (4)
Title |
---|
QI WU: "Regression application based on fuzzy m-support vector machine in symmetric triangular fuzzy space", 《EXPERT SYSTEMS WITH APPLICATIONS》 * |
QINGZHEN BI等: "Fv-SVM-Based Wall-Thickness Error Decomposition for Adaptive Machining of Large Skin Parts", 《IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS》 * |
李忠国等: "模糊支持向量机在路面识别中的应用", 《数据采集与处理》 * |
石磊等: "基于改进PSO算法参数优化的模糊支持向量分类机", 《信阳师范学院学报:自然科学版》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021004154A1 (en) | Method for predicting remaining life of numerical control machine tool | |
Nojedeh et al. | Tool path accuracy enhancement through geometrical error compensation | |
CN109145471B (en) | Virtual assembly system and method based on CAD and measured data co-fusion model | |
CN107748539B (en) | Five-axis machine tool multi-axis linkage error tracing method based on RTCP error feature recognition | |
CN105046684A (en) | Image matching method based on polygon generalized Hough transform | |
US11982996B2 (en) | Method and apparatus for configuring processing parameters of production equipment, and computer-readable medium | |
Salem et al. | Assessment of methods for industrial indoor object recognition | |
Liu et al. | Thermal error modeling for machine tools: Mechanistic analysis and solution for the pseudocorrelation of temperature-sensitive points | |
Tashlinskii et al. | Analysis of methods of estimating objective function gradient during recurrent measurements of image parameters | |
CN110378461A (en) | Error self-adaptation control method and system based on space-time error separate | |
D'Errico | A la Kalman filtering for metrology tool with application to coordinate measuring machines | |
CN112990190A (en) | Method and device for identifying reading of pointer instrument | |
CN115752243A (en) | Measurement data fusion method | |
CN116740665A (en) | Point cloud target detection method and device based on three-dimensional cross-correlation ratio | |
CN116243329A (en) | High-precision multi-target non-contact ranging method based on laser radar and camera fusion | |
Semere et al. | Skin model based tolerance and variations analysis | |
CN112946697B (en) | Satellite signal cycle slip detection and repair method based on deep learning | |
Xu et al. | A generic algorithm for position-orientation estimation with microscopic vision | |
CN108090396A (en) | A kind of finger print matching method and device | |
Su et al. | Fast detection method of checkerboard corners based on the combination of template matching and Harris Operator | |
Mostofi et al. | Using vision metrology system for quality control in automotive industries | |
Elhaouari et al. | Robust IMU-Monocular-SLAM For Micro Aerial Vehicle Navigation Using Smooth Variable Structure Filter | |
Dong et al. | Optimization algorithm of RGB-D SLAM visual odometry based on triangulation | |
Dong et al. | Bayesian support vector regression for tool condition monitoring and feature selection | |
Zhu et al. | Full compensation method of thermal error of NC machine tool based on sequence depth learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191025 |
|
RJ01 | Rejection of invention patent application after publication |