CN110262393B - Gray theory sectional weighted thermal error modeling method with lag data processing - Google Patents
Gray theory sectional weighted thermal error modeling method with lag data processing Download PDFInfo
- Publication number
- CN110262393B CN110262393B CN201910592635.8A CN201910592635A CN110262393B CN 110262393 B CN110262393 B CN 110262393B CN 201910592635 A CN201910592635 A CN 201910592635A CN 110262393 B CN110262393 B CN 110262393B
- Authority
- CN
- China
- Prior art keywords
- model
- thermal error
- data
- gray
- temperature
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 title claims abstract description 36
- 238000012360 testing method Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 238000012805 post-processing Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 13
- 238000009825 accumulation Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 7
- 238000009826 distribution Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000011161 development Methods 0.000 claims description 3
- 238000003754 machining Methods 0.000 abstract description 6
- 239000000284 extract Substances 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Images
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
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/404—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by control arrangements for compensation, e.g. for backlash, overshoot, tool offset, tool wear, temperature, machine construction errors, load, inertia
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/35—Nc in input of data, input till input file format
- G05B2219/35408—Calculate new position data from actual data to compensate for contour error
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Automatic Control Of Machine Tools (AREA)
Abstract
The invention discloses a gray theory sectional weighted thermal error modeling method with lag data processing, which comprises the steps of establishing a plurality of gray models with different sequence lengths through the processing capacity of a gray theory model on temperature and thermal error data to form a gray model preprocessing module, taking the output of the preprocessing module as the input of a distributed lag model post-processing module, establishing a lag model after determining the lag order of the model as the post-processing module, and forming a hierarchical model combining the gray model and the distributed lag model; establishing GM-DL compensation models at different initial temperatures based on the hierarchical model, selecting two adjacent thermal error compensation models according to the initial environment temperature of the test sample, and calculating to obtain final thermal error data in a sectional weighting mode to complete modeling. The method extracts useful information in the data to the maximum extent, so that the model has strong trend and adaptability, the adaptability of the model to all seasons and periods is improved, and the machining precision of the machine tool is greatly improved finally.
Description
Technical Field
The invention belongs to the technical field of thermal error compensation of machine tools, and particularly relates to a gray theory sectional type weighted thermal error modeling method with hysteresis data processing.
Background
A large number of experimental researches show that the thermal error is a main factor influencing the machining error of the numerical control machine tool, the proportion of the thermal error in all error sources can reach 40% -70%, and the thermal error proportion is higher for more precise machine tools. The thermal error is caused by a heat source, and the heat source is inevitably existed in the operation of the machine tool, so that the problem of the thermal error of the machine tool is inevitable, and the key for improving the processing precision of the precise numerical control machine tool is to seek a method for reducing the thermal error. Aiming at the problem of thermal error of a machine tool, a thermal error prevention method and a thermal error compensation method are two basic methods for reducing the thermal error and improving the machining precision of the machine tool. The thermal error prevention method is a hard technology, mainly improves the manufacturing precision of a machine tool in the design and manufacturing stages to meet the machining precision of a workpiece, and compared with a thermal error compensation method, the thermal error compensation method has the advantages that the cost for improving the same machining precision is very high in economy, and the precision improvement is limited, so that the thermal error compensation method is the main method for reducing the thermal error at present. The thermal error compensation method mainly establishes a mathematical model of the corresponding relation between the thermal sensitive point and the thermal deformation, and deduces the size of the thermal deformation through the temperature value of the thermal sensitive point in actual processing, thereby realizing the compensation of the thermal error through the origin offset function of the numerical control system of the machine tool and improving the processing precision of the machine tool.
There are many kinds of thermal error compensation models, and a Gray Model (GM) established according to a gray theory and a distributed hysteresis model (DL) established according to a time series principle are commonly used methods. The grey system theory regards random variables in the system as the amount of grey that changes and the random process as the grey process that generates changes within a certain range. Before the model is established, the original data which is irregular or has a weak rule is preprocessed to be changed into new data with a certain rule, and the model has good modeling effect and strong trend on a system with known information and unknown information. However, the temperature sensor measures the temperature of a certain position on the surface of the material, the model does not consider the modeling influence caused by the hysteresis of the deformation of the material relative to the temperature, particularly, when the working condition is complex, the heat source of the numerical control machine tool is frequently switched, the temperature measured by the sensor and the deformation of the interior of the material caused by the temperature change have larger hysteresis, and the model prediction precision is influenced to a certain extent by directly modeling neglecting the hysteresis. The distributed hysteresis model established by the time series principle is used for relating the value of the dependent variable with the current independent variable and a plurality of hysteresis independent variables, and the value is used for representing that the current thermal error value of the machine tool is not only related with a plurality of current temperature values but also related with a plurality of hysteresis temperature values in the thermal error modeling, so that the established model greatly weakens the influence caused by hysteresis and improves the prediction precision. However, due to the inaccuracy of the measurements and the complexity of the machine tool, the temperature data and thermal error data used for thermal error modeling inevitably contain some "grey information", and the distributed hysteresis model has no advantage in processing the "grey information".
In addition, the machine tool is generally placed in a factory, and the constant temperature environment of the factory needs to be maintained, so that the ambient temperature of the general machine tool changes along with the change of four seasons, a rule is summarized through a large amount of documents for thermal error modeling compensation, and the ambient temperature has important influence in the thermal error modeling of the machine tool. When the difference between the sample data of the modeling and the initial environment temperature of the test data is not large, the model compensation effect of the sample test is good; when the initial ambient temperature of the modeled training data and the test data for thermal error are very different, the compensation effect tends to be poor. For example, the thermal error model is built by using winter data to compensate the thermal error in summer or the thermal error model is built by using midday data to compensate the thermal error measured in the early morning. For the problem, firstly, modeling is performed by using winter and summer data as sample data through a method of increasing the number of samples, and secondly, a thermal error model with strong generalization capability, such as a support vector regression model, is selected, but through the compensation effect of predecessors, the compensation effect is improved to some extent but is limited.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a gray theory sectional weighted thermal error modeling method with hysteresis data processing aiming at the defects in the prior art, which can improve the thermal error prediction capability and finally improve the machine tool machining precision.
The invention adopts the following technical scheme:
a gray theory sectional weighted thermal error modeling method with lag data processing is used for establishing n gray models GM with different sequence lengths according to the processing capacity of a gray theory model on temperature and thermal error data1(1,N),GM2(1,N),...,GMn(1, N) forming a gray model preprocessing module, taking the output of the preprocessing module as the input of a distributed hysteresis model post-processing module, establishing a hysteresis model as the post-processing module after determining the hysteresis order of the model, and forming a layer combining the gray model and the distributed hysteresis modelA level model; establishing GM-DL compensation models at different initial temperatures based on the hierarchical model, selecting two adjacent thermal error compensation models according to the initial environment temperature of the test sample, and calculating to obtain final thermal error data in a sectional weighting mode to complete modeling.
Specifically, the data sequence has a length of n1,n2,...,nnRespectively training the gray models by using data sequences, simultaneously adjusting the sequence length according to the comparison of the training precision and the expected precision, preprocessing a new sample by n gray models to obtain n x m groups of thermal error values, wherein m is a thermal sensitive point number, and the preprocessed value and the measured value are used as a new sample training distribution hysteresis model to establish a GM-DL model.
Further, it is provided with
Is a sequence of thermal errors that is,
the method comprises the steps of selecting N-1 temperature measuring points for a model, wherein N is the length of the selected sequence, performing primary accumulation sequence on temperature and thermal error data, establishing a GM (1, N) model, calculating the thermal error data and the temperature data subjected to primary accumulation processing to form a matrix, finally calculating the predicted value of the GM (1, N) model on the thermal error through primary accumulation reduction and reduction, and repeating the method to obtain a preprocessing module of the thermal error formed by N GM (1, N) models.
Further, the once-accumulated sequence of temperature and thermal error data is:
the thermal error data and the temperature data after one-time accumulation processing form a matrix as follows:
the predicted value of the GM (1, N) model on the thermal error is obtained by one time of accumulation reduction as follows:
wherein k is 2, 3.
Further, take
Establishing GM (1, N) model by using mean value sequences generated by two adjacent terms in the sequences as follows:
wherein a is a model development coefficient; biIs the drive factor.
Further, a mathematical expression of a hysteresis model is established:
wherein,t～IID(0,σ2) N is the maximum lag phase α0Is a constant term; u is the number of thermally sensitive spots, ytAnd ultimately thermal error.
Specifically, the environment temperature of the machine tool is set to be 0-40 ℃, training samples are selected every 5 ℃, GM-DL thermal error models with initial temperatures of 0 ℃, 5 ℃, 10 ℃, 15 ℃, 20 ℃, 25 ℃, 30 ℃, 35 ℃ and 40 ℃ are respectively established, and GM-DL thermal error model libraries with different initial environment temperatures are formed.
Further, assuming that the initial temperature of the test data is t, the system respectively calculates the absolute value of the difference between the node temperatures and t, sorts the sequences according to the ascending order of the absolute values, and selects two groups of data with the minimum absolute value, wherein the two groups of data respectively correspond to the corresponding temperature nodes, and the two GM-DL models established by the two corresponding temperature nodes are the selected models.
Furthermore, the model is located between temperature nodes T1 ℃ and T2 ℃ after being selected, wherein T1 is more than T and less than T2, the test sample is brought into a GM-DL model with initial environment temperatures of T1 ℃ and T2 ℃ to obtain thermal errors of y1 and y2 respectively, and thermal error calculation under different initial environment temperatures is completed through two steps of model selection and weighted calculation on the basis of establishing a segmented model base in the previous period.
Further, the actual thermal error y is:
compared with the prior art, the invention has at least the following beneficial effects:
the method utilizes the excellent processing capability of the gray model to the gray information and the excellent modeling capability of the gray model to the less data to extract useful information in the data to the maximum extent, so that the model has strong tendency and adaptability, then utilizes the distributed hysteresis model to solve the hysteresis problem of temperature and deformation on the machine tool, so that the model has strong predictability to the serious hysteresis influence, utilizes the initial environment temperature to carry out the segmented weighted modeling on the thermal error, improves the adaptability of the model to each season and time period, and finally greatly improves the processing precision of the machine tool.
Furthermore, the problem that a single model cannot solve is solved by jointly modeling by utilizing the data processing capacity of the gray model and the capacity of solving the hysteresis of the distributed hysteresis model, and the predictability of the thermal error is improved.
Furthermore, a gray model preprocessing module is used for preprocessing the gray information to extract useful information in the model, so that the model has strong tendency and adaptability.
Furthermore, the measured temperature of the temperature sensor in data acquisition and the deformation of the machine tool caused by the temperature have hysteresis, so that the model can solve the influence of the hysteresis on modeling, and the model has strong predictability for the serious influence of the hysteresis.
Furthermore, different models for thermal error compensation are provided in the model library, and a proper model can be preferentially selected for modeling according to different initial temperatures, so that a foundation is laid for weighted modeling.
Furthermore, the initial environment temperature is used for carrying out segmented weighted modeling on the thermal error, so that the information in the model can be fully utilized, and the adaptability of the model to all seasons and periods is improved.
In conclusion, the invention utilizes the excellent processing capability of the gray model on the gray information and the excellent modeling capability of the gray model on the less data to extract the useful information in the data to the maximum extent, so that the model has strong tendency and adaptability, then utilizes the distributed hysteresis model to solve the problem of hysteresis of temperature and deformation on the machine tool, so that the model has strong predictability on the condition with serious hysteresis influence, utilizes the initial environment temperature to carry out the segmented weighted modeling on the thermal error, improves the adaptability of the model to each season and time period, and finally greatly improves the processing precision of the machine tool.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
FIG. 1 is a diagram of a GM-DL model architecture;
FIG. 2 is a piecewise weighted thermal error compensation model.
Detailed Description
The invention provides a gray theory sectional weighted thermal error modeling method with hysteresis data processing, which utilizes excellent processing capability of a gray theory model on gray information and excellent modeling capability of less data to preprocess thermal errors, and then utilizes a distributed hysteresis model to analyze and model hysteresis problems of deformation in a machine tool to form a GM-DL (GM-DL) level model. Selecting sample data with different initial environment temperatures, establishing a GM-DL thermal error model library under different initial environment temperatures, respectively calculating thermal errors according to GM-DL thermal error models with the most similar positive and negative in the initial environment temperature selection library of the test sample, and finally obtaining the final thermal error according to the weight of the initial temperature piece.
Referring to fig. 1 and fig. 2, a gray theory segmented weighted thermal error modeling method with hysteresis data processing according to the present invention includes the following steps:
s1, establishing n gray models GM with different sequence lengths1(1,N),GM2(1,N),...,GMn(1, N), the data sequence length is N respectively1,n2,...,nnRespectively training the model by using the data sequences, and simultaneously adjusting the length of the sequence according to the comparison between the training precision and the expected precision;
s2, preprocessing a new sample by n gray models to obtain n x m groups (m is a heat sensitive point number) of heat error values, taking the preprocessed values and measured values as a new sample training distribution hysteresis model, and finally establishing a GM-DL model;
s3, setting the environment temperature of the machine tool to be 0-40 ℃, selecting training samples every 5 ℃, respectively establishing GM-DL thermal error models with initial temperatures of 0 ℃, 5 ℃, 10 ℃, 15 ℃, 20 ℃, 25 ℃, 30 ℃, 35 ℃ and 40 ℃, and forming GM-DL thermal error model libraries with different initial environment temperatures;
s4, selecting two groups of GM-DL thermal error models with the initial temperature being most similar to the initial temperature in the GM-DL thermal error model base according to the initial environment temperature of the test sample, and respectively calculating thermal errors;
and S5, calculating the absolute value of the difference between the initial temperature of the test sample and the initial temperatures of the two groups of models according to the initial temperature of the test sample in the step S4, and setting weight according to inverse correlation to obtain a final thermal error value.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention relates to a gray theory sectional weighted thermal error modeling method with hysteresis data processing, which comprises the following steps:
establishing GM-DL model
1) Grey model preprocessing module
Is provided with
Is a sequence of thermal errors that is,
n, namely N-1 temperature measuring points are selected by the model, and N is the length of the selected sequence.
The once-accumulated sequence of temperature and thermal error data is:
wherein k is 1, 2.
Get
The mean sequences generated by two adjacent terms in the sequence are:
wherein k is 2,3, the. and N, and establishing a GM (1, N) model based on the change of the data sequence
Wherein k is 2,3, n, a is a model development coefficient; biIs called the driving coefficient (or the amount of gray effect).
Coefficients a and b in the formula (1)iConstituting a coefficient vector
PN＝(a,b2,b3,...,bN)T
A column vector consisting of n-1 thermal error data items of
The thermal error data and the temperature data after one-time accumulation processing form the following matrix:
equation (1) can be expressed as a system of matrix equations
YN＝BPN
From the least square law, the coefficient vector of the model can be found
PN＝(BTB)-1BTYN(2)
From the grey theory, the approximate time response of GM (1, N) can be expressed as
Obtaining a predicted value of the GM (1, N) model to the thermal error through one accumulation reduction:
setting different sequence lengths n1,n2,...,nnThe above method is repeated to obtain N GM (1, N) models to form a pre-processing module of thermal error, as shown in fig. 1.
2) Distributed hysteresis model post-processing module
In the post-processing module, a mathematical expression of a hysteresis model is established:
in the formula,t～IID(0,σ2) N is the maximum lag phase α0Is a constant term; u is the number of thermally sensitive spots, ytAnd ultimately thermal error.
N x m groups (m is the number of heat sensitive points) of heat error values obtained by preprocessing n gray models are used as the input of the distribution lag model, the error measured value in a new sample is used as the output of the distribution lag model, the coefficient of the distribution lag model is estimated according to the least square method, and the modeling of the post-processing module is completed
4.2 segmented weighted modeling
The GM-DL hierarchical model is the basis for the segmented modeling. According to the annual environmental temperature condition of the machine tool, the temperature is assumed to be 0-40 ℃, models are established at fixed temperature intervals, 5 ℃ is assumed to be selected at intervals, and therefore the initial temperature nodes of modeling data are 0 ℃, 5 ℃, 10 ℃, 15 ℃, 20 ℃, 25 ℃, 30 ℃, 35 ℃ and 40 ℃. Through measured experimental data, GM-DL models at different temperature nodes are established, so that a segmented model base is obtained, as shown in FIG. 2.
1) GM-DL model selection
Assuming that the initial temperature of the test data is t, the system calculates the absolute value of the difference between the node temperature and t, i.e. the absolute value
|-t|、|5-t|、|10-t|、|15-t|、|20-t|、|25-t|、|30-t|、|35-t|、|40-t|
And sequencing the sequences according to the ascending order of absolute values, selecting two groups of data with the minimum absolute value, wherein the two groups of data respectively correspond to corresponding temperature nodes, and then two GM-DL models established by the two corresponding temperature nodes are the models to be selected.
2) And (3) weighting calculation:
assuming that the initial environment temperature of the test sample is T, the test sample is located between temperature nodes T1 and T2 ℃ after model selection, wherein T1 is more than T and less than T2, the test sample is brought into a GM-DL model with the initial environment temperature of T1 and T2 ℃ to obtain thermal errors of y1 and y2 respectively, and then the actual thermal error y is calculated according to the following weight calculation rules:
on the basis of establishing a segmented model base in the previous period, the thermal error calculation aiming at different initial environment temperatures is completed through two steps of model selection and weighted calculation.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.
Claims (9)
1. The gray theory sectional weighted thermal error modeling method with lag data processing is characterized in that n gray models GM with different sequence lengths are established through the processing capacity of the gray theory model on temperature and thermal error data1(1,N),GM2(1,N),...,GMn(1, N) forming a gray model preprocessing module, taking the output of the preprocessing module as the input of a distributed hysteresis model post-processing module, establishing a distributed hysteresis model as the post-processing module after determining the hysteresis order of the model, and forming a hierarchical model combining a gray model and the distributed hysteresis model; establishing GM-DL thermal error compensation models at different initial temperatures based on the hierarchical model, selecting two adjacent GM-DL thermal error compensation models according to the initial environment temperature of the test sample, and calculating to obtain final thermal error data in a sectional weighting mode to complete modeling.
2. The gray theory segmented weighted thermal error modeling method with lag data processing according to claim 1, wherein the data sequence length is n1,n2,...,nnRespectively training the gray model by using the data sequence,and simultaneously, adjusting the sequence length according to the comparison of the training precision and the expected precision, preprocessing a new sample by n grey models to obtain n x m groups of thermal error values, wherein m is the number of thermal sensitive points, and the preprocessed values and the measured values are used as a new sample training distribution hysteresis model to establish a GM-DL thermal error compensation model.
3. The gray theory segmented weighted thermal error modeling method with hysteresis data processing according to claim 1 or 2, characterized in that
Is a sequence of thermal errors that is,
the method comprises the steps of selecting N-1 temperature measuring points for a model, performing primary accumulation sequence on temperature and thermal error data, establishing a GM (1, N) model, calculating a matrix formed by the thermal error data and the temperature data subjected to the primary accumulation processing, calculating a predicted value of the GM (1, N) model on the thermal error through primary accumulation reduction and reduction, and repeating the method to obtain a preprocessing module of the thermal error formed by N GM (1, N) models.
4. The gray theory segmented weighted thermal error modeling method with hysteresis data processing as claimed in claim 3, wherein the once accumulated sequence of temperature and thermal error data is:
the thermal error data and the temperature data after one-time accumulation processing form a matrix as follows:
get
The mean sequences generated by two adjacent terms in the sequence are:
wherein k is 2, 3.., n,
is composed of
And
the average value of the two is taken as the inverse,
is composed of
And
the average value of the two is taken as the inverse,
is composed of
And
taking the inverse of the mean value;
the predicted value of the GM (1, N) model on the thermal error is obtained by one time of accumulation reduction as follows:
wherein k is 2, 3.., n,
the accumulated predicted value of the thermal error is calculated sequentially according to the above formula, i.e.
Is composed of
The last accumulated prediction value.
5. The gray theory segmented weighted thermal error modeling method with hysteresis data processing as claimed in claim 3, characterized in that, taking
Establishing GM (1, N) model by using mean value sequences generated by two adjacent terms in the sequences as follows:
wherein a is a model development coefficient; biIs the drive factor.
6. The gray theory segmented weighted thermal error modeling method with the hysteresis data processing as claimed in claim 1, wherein the environment temperature of the machine tool is set to be 0-40 ℃, training samples are selected every 5 ℃ to respectively establish GM-DL thermal error compensation models with initial temperatures of 0 ℃, 5 ℃, 10 ℃, 15 ℃, 20 ℃, 25 ℃, 30 ℃, 35 ℃ and 40 ℃, and GM-DL thermal error compensation model libraries with different initial environment temperatures are formed.
7. The gray theory segmented weighted thermal error modeling method with the hysteresis data processing as claimed in claim 6, wherein assuming that the initial temperature of the test data is t, the system respectively calculates absolute values of differences between the node temperatures and t, sorts the sequences according to the ascending order of the absolute values, selects two sets of data with the smallest absolute value, the two sets of data respectively correspond to the corresponding temperature nodes, and the two GM-DL thermal error compensation models established by the two corresponding temperature nodes are the selected models.
8. The gray theory segmented weighted thermal error modeling method with hysteresis data processing as claimed in claim 7, wherein the model is selected and located between temperature nodes T1 ℃ and T2 ℃, wherein T1 < T < T2, the thermal errors are y1 and y2 respectively by substituting the test sample into GM-DL thermal error compensation models with initial environment temperatures of T1 ℃ and T2 ℃, and the thermal error calculation for different initial environment temperatures is completed by two steps of model selection and weighted calculation on the basis of the previous stage of building the segmented model base.
9. The gray theory segmented weighted thermal error modeling method with hysteresis data processing as claimed in claim 8, wherein the actual thermal error y is:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910592635.8A CN110262393B (en) | 2019-07-03 | 2019-07-03 | Gray theory sectional weighted thermal error modeling method with lag data processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910592635.8A CN110262393B (en) | 2019-07-03 | 2019-07-03 | Gray theory sectional weighted thermal error modeling method with lag data processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110262393A CN110262393A (en) | 2019-09-20 |
| CN110262393B true CN110262393B (en) | 2020-08-18 |
Family
ID=67923993
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910592635.8A Expired - Fee Related CN110262393B (en) | 2019-07-03 | 2019-07-03 | Gray theory sectional weighted thermal error modeling method with lag data processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110262393B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114295095B (en) * | 2021-11-26 | 2023-07-14 | 广西科技大学 | Method for determining optimal number of measuring points for free-form surface detection |
| CN115328025B (en) * | 2022-09-21 | 2024-07-09 | 重庆大学 | Machine tool thermal error double-closed-loop modeling and compensation method based on model applicability evaluation index |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6167634B1 (en) * | 1998-03-28 | 2001-01-02 | Snu Precision Co., Ltd. | Measurement and compensation system for thermal errors in machine tools |
| CN103823991A (en) * | 2014-03-11 | 2014-05-28 | 华中科技大学 | Heavy-duty tool thermal error prediction method taking environmental temperature into account |
| CN103984287A (en) * | 2014-03-12 | 2014-08-13 | 江苏齐航数控机床有限责任公司 | Numerically-controlled machine tool thermal error compensation grey neural network modeling method |
| CN105404237A (en) * | 2015-11-10 | 2016-03-16 | 湖北文理学院 | Numerical control machine tool space error modeling method based on space grid compensation mode |
| CN105700473A (en) * | 2016-04-13 | 2016-06-22 | 合肥工业大学 | Method for curved surface thermal-error compensation of whole workbench of precise numerical-controlled machine tool |
| CN106372337A (en) * | 2016-09-05 | 2017-02-01 | 华中科技大学 | Thermal deformation prediction method of preheating stage of numerical control machine tool |
| CN106736848A (en) * | 2016-12-13 | 2017-05-31 | 西安交通大学 | Numerically controlled lathe Thermal Error measures compensation system and compensation method |
| CN108594761A (en) * | 2018-04-25 | 2018-09-28 | 河北工业大学 | A kind of electro spindle Thermal Error Active Control Method keeping strategy based on initial temperature |
-
2019
- 2019-07-03 CN CN201910592635.8A patent/CN110262393B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6167634B1 (en) * | 1998-03-28 | 2001-01-02 | Snu Precision Co., Ltd. | Measurement and compensation system for thermal errors in machine tools |
| CN103823991A (en) * | 2014-03-11 | 2014-05-28 | 华中科技大学 | Heavy-duty tool thermal error prediction method taking environmental temperature into account |
| CN103984287A (en) * | 2014-03-12 | 2014-08-13 | 江苏齐航数控机床有限责任公司 | Numerically-controlled machine tool thermal error compensation grey neural network modeling method |
| CN105404237A (en) * | 2015-11-10 | 2016-03-16 | 湖北文理学院 | Numerical control machine tool space error modeling method based on space grid compensation mode |
| CN105700473A (en) * | 2016-04-13 | 2016-06-22 | 合肥工业大学 | Method for curved surface thermal-error compensation of whole workbench of precise numerical-controlled machine tool |
| CN106372337A (en) * | 2016-09-05 | 2017-02-01 | 华中科技大学 | Thermal deformation prediction method of preheating stage of numerical control machine tool |
| CN106736848A (en) * | 2016-12-13 | 2017-05-31 | 西安交通大学 | Numerically controlled lathe Thermal Error measures compensation system and compensation method |
| CN108594761A (en) * | 2018-04-25 | 2018-09-28 | 河北工业大学 | A kind of electro spindle Thermal Error Active Control Method keeping strategy based on initial temperature |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110262393A (en) | 2019-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255160B (en) | Neural network-based unit delay prediction method and unit delay sensitivity calculation method | |
| Khotanzad et al. | A neuro-fuzzy approach to short-term load forecasting in a price-sensitive environment | |
| CN109214948A (en) | A kind of method and apparatus of electric system heat load prediction | |
| CN112149879B (en) | New energy medium-and-long-term electric quantity prediction method considering macroscopic volatility classification | |
| CN111478314B (en) | Transient stability evaluation method for power system | |
| CN110262393B (en) | Gray theory sectional weighted thermal error modeling method with lag data processing | |
| CN101446994A (en) | Modeling method of thermal error least squares support vector machine of numerically-controlled machine tool | |
| CN105209984A (en) | Methods for ascertaining a model of a starting variable of a technical system | |
| CN110083125A (en) | A kind of machine tool thermal error modeling method based on deep learning | |
| CN111639111A (en) | Water transfer engineering-oriented multi-source monitoring data deep mining and intelligent analysis method | |
| CN113988373B (en) | Multi-task massive user load prediction method based on multi-channel convolutional neural network | |
| CN114548368A (en) | Modeling method, prediction model and prediction method of lithium battery temperature field prediction model based on multilayer nuclear overrun learning machine | |
| CN112801388B (en) | Power load prediction method and system based on nonlinear time series algorithm | |
| CN112308159A (en) | Image identification and classification method based on prediction increment width learning | |
| CN117289652A (en) | Numerical control machine tool spindle thermal error modeling method based on multi-universe optimization | |
| CN116774086B (en) | Lithium battery health state estimation method based on multi-sensor data fusion | |
| CN116703644A (en) | Attention-RNN-based short-term power load prediction method | |
| CN112170501A (en) | Prediction method for wear crown and thermal crown of roller | |
| CN114117852B (en) | Regional heat load rolling prediction method based on finite difference working domain division | |
| CN111652422A (en) | Heat supply system load prediction method, device and system based on building classification | |
| CN114971090A (en) | Electric heating load prediction method, system, equipment and medium | |
| CN112580844A (en) | Meteorological data processing method, device, equipment and computer readable storage medium | |
| CN116894180B (en) | Product manufacturing quality prediction method based on different composition attention network | |
| CN109782586A (en) | The tight format non-model control method of the different factor of the MISO of parameter self-tuning | |
| CN110909492B (en) | Sewage treatment process soft measurement method based on extreme gradient lifting algorithm |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200818 |