CN115204321B - Precision fluctuation control method and system for automatic lathe machining - Google Patents
Precision fluctuation control method and system for automatic lathe machining Download PDFInfo
- Publication number
- CN115204321B CN115204321B CN202211119770.9A CN202211119770A CN115204321B CN 115204321 B CN115204321 B CN 115204321B CN 202211119770 A CN202211119770 A CN 202211119770A CN 115204321 B CN115204321 B CN 115204321B
- Authority
- CN
- China
- Prior art keywords
- sample
- precision
- workpiece
- data
- fluctuation
- 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.)
- Active
Links
- 238000003754 machining Methods 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 72
- 238000012545 processing Methods 0.000 claims abstract description 60
- 238000007689 inspection Methods 0.000 claims abstract description 35
- 238000013145 classification model Methods 0.000 claims abstract description 28
- 239000000470 constituent Substances 0.000 claims abstract description 5
- 238000012512 characterization method Methods 0.000 claims description 27
- 230000008439 repair process Effects 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 12
- 238000012544 monitoring process Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000001276 controlling effect Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 2
- 239000000047 product Substances 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 4
- 239000002356 single layer Substances 0.000 description 4
- 125000006850 spacer group Chemical group 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 239000012467 final product Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012958 reprocessing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Numerical Control (AREA)
Abstract
The invention discloses a precision fluctuation control method and a system for automatic lathe machining, wherein the method comprises the following steps that S1, a quality inspection device is arranged at a workpiece outlet end of an automatic lathe; s2, creating machining precision sample data by using sample workpieces with known fluctuation control categories, and constructing a fluctuation control classification model based on the machining precision sample data; and S3, re-processing the target workpiece according to the fluctuation control category so that the processing precision of the target workpiece meets the processing requirement. According to the invention, the processing precision representation data is obtained by carrying out weight summation processing on the size precision representation data, the position precision representation data and the shape precision representation data of the target workpiece, so that the automatic regulation and control of the constituent elements of the processing precision is realized for various workpiece types adapting to the target workpiece, the application of quality inspection on various types of workpieces can be realized, the adaptability of quality inspection during expansion is realized, in addition, a fluctuation control classification model is established, the fluctuation control category of the target workpiece can be accurately obtained, and the quality inspection accuracy is improved.
Description
Technical Field
The invention relates to the technical field of lathe machining, in particular to a precision fluctuation control method and system for automatic lathe machining.
Background
Various devices in high-voltage transmission comprise pins, equalizing rings, anti-vibration hammers and spacing bars after knurling, an automatic lathe is required to be used for producing the components, product precision deviation refers to the fact that products are affected by various deviation sources in the machining process, and as the components pass through a production line, dimensional precision deviation is continuously generated, accumulated and transmitted, and finally deviation difference values of actual precision of the products relative to design precision are formed. As the production process becomes more and more complex, the product is processed through one process, and the quality of the product is affected by multiple sources of deviation. The processing system of the complex product is often a multi-source multi-process system combining parallel and serial, the quality of the final product is affected by a plurality of deviation sources in all processes in the processing process, besides various deviations in single processes, such as the influence of material characteristics of parts, tooling equipment, element characteristics of clamps and the like on the deviation of the product to different degrees, complex coupling relations exist among different processes, and the deviation can be introduced, so that the quality deviation is continuously generated, transmitted, increased, subtracted, accumulated and transmitted, and the precision deviation of the final product is formed. Because the product precision deviation is one of the most important factors which directly influence the product quality, the productivity, the market response time and the like, the workpiece precision quality inspection is inevitably required in the production process, and the workpiece to be repaired is screened out and repaired to meet the designed standard precision.
In the prior art, quality inspection of components such as pins, equalizing rings, damper hammers, spacing bars and the like after knurling is generally designed according to different component requirements, and the quality inspection of the components is performed by only using a single classification method, so that the accuracy is low.
Disclosure of Invention
The invention aims to provide a precision fluctuation control method and a system for automatic lathe machining, which are used for solving the technical problems that different quality inspection schemes are required to be designed for different parts in the prior art, and quality inspection of the parts is carried out by only using a single classification method, so that the precision is low.
In order to solve the technical problems, the invention specifically provides the following technical scheme:
the precision fluctuation control method for automatic lathe machining comprises the following steps:
step S1, arranging a quality inspection device at a workpiece outlet end of an automatic lathe, wherein the quality inspection device is used for monitoring processing precision characterization data of a target workpiece positioned at the workpiece outlet end of the automatic lathe;
s2, creating machining precision sample data by using sample workpieces with known fluctuation control categories, and constructing a fluctuation control classification model based on the machining precision sample data, wherein the fluctuation control classification model is used for obtaining the fluctuation control category of a target workpiece according to machining precision characterization data;
and S3, re-processing the target workpiece according to the fluctuation control category so that the processing precision of the target workpiece meets the processing requirement.
As a preferred embodiment of the present invention, in the step S1, the method for monitoring the machining precision characterization data includes:
the quality inspection device acquires data of machining size, machining position and machining shape of the target workpiece to obtain size precision representation data, position precision representation data and shape precision representation data respectively;
characterizing data, positions, for the dimensional accuracyThe processing precision characterization data are obtained by carrying out weight summation processing on the precision characterization data and the shape precision characterization data, so that the forming elements for realizing autonomous regulation and control of the processing precision are used for various workpiece types adapting to target workpieces, wherein the weight summation processing formula is as follows:，
wherein S is characterized by processing precision characterization data,、/>and->Respectively characterizing as dimension precision characterizing data, position precision characterizing data and shape precision characterizing data, +.>、/>And->The characteristics are respectively the weight of the size precision characteristic data, the weight of the position precision characteristic data and the weight of the shape precision characteristic data.
As a preferred embodiment of the present invention, in the step S2, the method for creating the machining precision sample data includes:
obtaining a plurality of sample workpieces, wherein the plurality of sample workpieces comprise positive sample workpieces and negative sample workpieces with the same number, the positive sample workpieces are characterized in that the fluctuation control type is a repaired workpiece, and the negative sample workpieces are characterized in that the fluctuation control type is an unrepaired workpiece;
processing precision meter for sample workpiece in sequenceCollecting the sign data, taking the processing precision characterization data of the sample workpiece as sample characteristics of the sample workpiece, taking the fluctuation control category as a sample label of the sample workpiece, and constructing by combining the sample characteristics and the sample label to obtain the processing precision sample data, wherein the characterization method of the processing precision sample data comprises the following steps:；
in the method, in the process of the invention,machining accuracy sample data, characterized by sample workpiece i, < >>Sample feature/machining accuracy characterization data characterizing sample workpiece i, +.>Sample tag/wave control class characterized as sample workpiece i, i is characterized as sample workpiece differentiation number, +.>N is characterized by the total number of sample workpieces, +.>Wave control class/sample tag characterized as sample workpiece i is repair, < >>The wave control class/sample tag characterized as sample workpiece i is not repaired.
As a preferred embodiment of the present invention, in the step S2, the method for creating the fluctuation control classification model includes:
setting a fluctuation thresholdAnd is combined withConstructing a fluctuation classifier G with a fluctuation threshold, wherein the construction method of the fluctuation classifier comprises the following steps:
step S201, setting the iteration number k, initializing to 1, and setting a sample fluctuation weight for the sample workpiece i；
Step S202, if the processing precision of the sample workpiece i represents the dataLess than fluctuation threshold->Then two classification classifiers->Marking the classification result of the sample tag of sample workpiece i as +.>Or (b)；
Characterization data of processing precision of sample workpiece iGreater than fluctuation threshold->Then two classification classifiersMarking the classification result of the sample tag of sample workpiece i as +.>Or (b)；
Step S203, sequentially obtaining n-1 classification classifiersAnd sequentially calculating n-1 classification classifiers +.>Classification error rate +.>Wherein, the method comprises the steps of, wherein,
if it isOutputting the current fluctuation classifier G as a fluctuation control classification model, and stopping iteration;
if it isSelecting the classification error rate +.>Minimum two-class classifier->Marked +.>The optimal error rate is marked +.>And calculates the optimal classifier mark as +.>Classifier fluctuation weights of +.>Sample fluctuation weight for the sample workpiece i>Updating to obtain sample fluctuation weight of sample workpiece i>And update the fluctuation classifier to +.>K is added with 1, and the step S202 is returned;
wherein, l and j are measurement constants, which have no substantial meaning,、/>sample fluctuation weight of sample workpiece i characterized by kth, k+1th iteration, +.>The j-th classifier of the classification, characterized by the k-th iteration, ">Classification error rate of the jth classification classifier characterized by the kth iteration, +.>A bi-classification classifier characterized by the kth iteration +.>Classification results of sample tags for sample workpiece i.
As a preferable mode of the invention, the classifier fluctuates weightThe calculation method of (1) comprises the following steps:
calculating the two-classification classifierClassification error rate +.>Wherein, the calculation formula of the classification error rate is: />；
In the method, in the process of the invention,sample label, characterized by sample workpiece i true for the kth iteration, < >>A bi-classification classifier characterized by the kth iteration +.>Classification result of sample label of sample workpiece i, if->And->Inequality->If->And->Equal +.>；
Calculating an optimal error rateWherein->The minimum function is obtained;
computing an optimal classifierClassifier fluctuation weights of +.>。
As a preferable mode of the invention, the sample wave weightThe updating method of (1) comprises the following steps:
when k=1, the sample is fluctuated by weightSet to->；
At k>1, weighting the sample fluctuation according to the updated formulaUpdating to obtain sample fluctuation weightThe update formula is as follows: />；
In the method, in the process of the invention,，/>normalized constant characterized by the kth iteration, < +.>The optimal classifier for the kth iteration is marked +.>Is used to determine the classifier fluctuating weights.
As a preferred embodiment of the present invention, the method for obtaining the surge control category includes:
inputting the machining precision representation data S of the target workpiece into a fluctuation control classification model to obtain a classification result of a sample label of the target workpieceWherein, the method comprises the steps of, wherein,
if it isThe fluctuation control category of the target workpiece is repair;
if it isThe fluctuation control class of the target workpiece is not repaired.
As a preferred embodiment of the present invention, in the step S3, the method for reworking the target workpiece includes:
comparing the size precision representation data, the position precision representation data and the shape precision representation data in the machining precision representation data of the target workpiece with the fluctuation control class as repair with the size precision representation data, the position precision representation data and the shape precision representation data of the standard workpiece respectively to obtain the repair field of the target workpiece with the fluctuation control class as repair;
and placing the target workpiece with the fluctuation control category being repaired in a repair neighborhood for re-machining so as to realize the repair of the target workpiece until the machining precision reaches the machining requirement of the standard component.
As a preferable mode of the present invention, the dimensional accuracy characterizing data, the position accuracy characterizing data, and the shape accuracy characterizing data are normalized before the weight summation is performed.
As a preferred embodiment of the present invention, there is provided a control system according to the method for controlling precision fluctuation of automatic lathe machining, comprising:
the quality inspection device is used for monitoring processing precision characterization data of a target workpiece positioned at the workpiece outlet end of the automatic lathe;
the model building unit is used for building machining precision sample data from sample workpieces with known fluctuation control categories, and building a fluctuation control classification model based on the machining precision sample data, wherein the fluctuation control classification model is used for obtaining the fluctuation control category of the target workpiece according to machining precision characterization data;
and the re-machining unit re-machines the target workpiece according to the fluctuation control type so that the machining precision of the target workpiece meets the machining requirement.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the processing precision representation data is obtained by carrying out weight summation processing on the size precision representation data, the position precision representation data and the shape precision representation data of the target workpiece, so that the automatic regulation and control of the constituent elements of the processing precision is realized for various workpiece types adapting to the target workpiece, the application of quality inspection on various types of workpieces can be realized, the adaptability of quality inspection during expansion is realized, in addition, a fluctuation control classification model is established, the fluctuation control category of the target workpiece can be accurately obtained, and the quality inspection accuracy is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It will be apparent to those of ordinary skill in the art that the drawings in the following description are exemplary only and that other implementations can be obtained from the extensions of the drawings provided without inventive effort.
FIG. 1 is a flowchart of a precision fluctuation control method according to an embodiment of the present invention;
fig. 2 is a block diagram of a control system according to an embodiment of the present invention.
Reference numerals in the drawings are respectively as follows:
1-a quality inspection device; 2-a model building unit; 3-a reprocessing unit.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in figure 1, various workpieces in high-voltage transmission comprise pins, equalizing rings, damper hammers, spacing bars and the like after knurling, the focus points of quality inspection of the workpieces after production are different, some focus on dimensional accuracy, some focus on position accuracy and shape accuracy, some focus on shape accuracy and dimensional accuracy, and the focus on position accuracy, and in the traditional quality inspection, the workpieces are classified according to types, namely, each type of workpieces are formulated into a type of quality inspection scheme and are mutually independent and not movable for use, so that the efficiency is low.
The precision fluctuation control method for automatic lathe machining comprises the following steps:
step S1, arranging a quality inspection device at a workpiece outlet end of an automatic lathe, wherein the quality inspection device is used for monitoring processing precision characterization data of a target workpiece positioned at the workpiece outlet end of the automatic lathe;
the machining precision is the degree of coincidence between the three geometric parameters of the actual size, shape and position of the machined workpiece surface and the ideal geometric parameters required by the drawing, and basically three aspects can be summarized, namely: 1. dimensional accuracy refers to the degree of coincidence between the actual size of the machined part and the center of a tolerance zone of the part size; 2. shape accuracy, which is the degree of coincidence between the actual geometry of the surface of the machined part and the ideal geometry; 3. positional accuracy refers to the difference in actual positional accuracy between the relevant surfaces of the part after machining.
In order to expand the applicable category range of workpiece quality inspection, the size precision, the shape precision and the position precision are monitored simultaneously, and weights are set for the size precision, the shape precision and the position precision according to the important attention precision of the workpiece category, so that a three-aspect feature of simultaneously representing the precision is obtained, and the three-aspect feature data of the focus attention of the precision is also provided, and the three-aspect feature data of the focus attention of the precision are specifically as follows:
in step S1, the method for monitoring the machining precision characterization data includes:
the quality inspection device acquires data of machining size, machining position and machining shape of the target workpiece to obtain size precision representation data, position precision representation data and shape precision representation data respectively;
carrying out weight summation processing on the size precision representation data, the position precision representation data and the shape precision representation data to obtain processing precision representation data so as to realize that constituent elements for independently regulating and controlling the processing precision are used for various workpiece types adapting to target workpieces, wherein the weight summation processing formula is as follows:，
wherein S is characterized by processing precision characterization data,、/>and->Respectively characterizing as dimension precision characterizing data, position precision characterizing data and shape precision characterizing data, +.>、/>And->The characteristics are respectively the weight of the size precision characteristic data, the weight of the position precision characteristic data and the weight of the shape precision characteristic data.
For example, if the shape accuracy and the dimensional accuracy are focused on the spacer during quality inspection and the position accuracy is focused on the spacer, the processing accuracy of the spacer represents the structure of the data S、/>The value is set higher and higher>When the value setting is low and the pin focuses on the dimensional accuracy, the shape accuracy and the position accuracy during quality inspection, the processing accuracy of the spacer represents the constitution of the data SThe value is set higher and higher>、/>The value setting is low, and when quality inspection is performed on different types of workpieces, only the weight of the size precision representation data, the weight of the position precision representation data and the weight of the shape precision representation data are required to be adjusted, so that the method can be applied to various types of workpieces in an expanded mode, and the application mobility is high.
S2, creating machining precision sample data by using sample workpieces with known fluctuation control categories, and constructing a fluctuation control classification model based on the machining precision sample data, wherein the fluctuation control classification model is used for obtaining the fluctuation control category of the target workpiece according to machining precision characterization data;
in step S2, the method for creating the machining precision sample data includes:
obtaining a plurality of sample workpieces, wherein the plurality of sample workpieces comprise positive sample workpieces and negative sample workpieces with the same number, the positive sample workpieces are characterized in that the fluctuation control type is a repaired workpiece, and the negative sample workpieces are characterized in that the fluctuation control type is an unrepaired workpiece;
the method comprises the steps of sequentially collecting machining precision representation data of a sample workpiece, taking the machining precision representation data of the sample workpiece as sample characteristics of the sample workpiece, taking a fluctuation control class as a sample label of the sample workpiece, and constructing by combining the sample characteristics and the sample label to obtain the machining precision sample data, wherein the method for representing the machining precision sample data comprises the following steps:
；
in the method, in the process of the invention,machining accuracy sample data, characterized by sample workpiece i, < >>Sample feature/machining accuracy characterization data characterizing sample workpiece i, +.>Sample tag/wave control class characterized as sample workpiece i, i is characterized as sample workpiece differentiation number, +.>N is characterized by the total number of sample workpieces, +.>Wave control class/sample tag characterized as sample workpiece i is repair, < >>The wave control class/sample tag characterized as sample workpiece i is not repaired.
Conventionally, a single-layer two-classification classifier is directly built based on processing precision sample data of a sample workpiece to classify fluctuation control categories of a target workpiece, but the classification precision of the single-layer two-classification classifier is low, so that an iteration algorithm is utilized in the embodiment to build a plurality of single-layer two-classification classifiers, and the single-layer two-classification classifiers are combined to be built into a high-precision multi-layer two-classification classifier to serve as a fluctuation control classification model, so that the fluctuation control category classification precision of the target workpiece is effectively improved, and the method comprises the following steps of:
in step S2, the method for creating the fluctuation control classification model includes:
setting a fluctuation thresholdAnd constructing a fluctuation classifier G with a fluctuation threshold, wherein the construction method of the fluctuation classifier comprises the following steps:
step S201, setting the iteration number k, initializing to 1, and setting a sample fluctuation weight for the sample workpiece i；
Step S202, if the processing precision of the sample workpiece i represents the dataLess than fluctuation threshold->Then two classification classifiers->Marking the classification result of the sample tag of sample workpiece i as +.>Or (b)；
Characterization data of processing precision of sample workpiece iGreater than fluctuation threshold->Then two classification classifiersMarking the classification result of the sample tag of sample workpiece i as +.>Or (b)；
Step S203, sequentially obtaining n-1 classification classifiersAnd sequentially calculating n-1 classification classifiers +.>Classification error rate +.>Wherein, the method comprises the steps of, wherein,
if it isOutputting the current fluctuation classifier G as a fluctuation control classification model, and stopping iteration;
if it isSelecting the classification error rate +.>Minimum two-class classifier->Marked +.>The optimal error rate is marked +.>And calculates the optimal classifier mark as +.>Classifier fluctuation weights of +.>Sample fluctuation weight for sample workpiece i>Updating to obtain sample fluctuation weight of sample workpiece i>And update the fluctuation classifier to +.>K is added with 1, and the step S202 is returned;
wherein, l and j are measurement constants, which have no substantial meaning,、/>sample fluctuation weight of sample workpiece i characterized by kth, k+1th iteration, +.>The j-th classifier of the classification, characterized by the k-th iteration, ">Classification error rate of the jth classification classifier characterized by the kth iteration, +.>A bi-classification classifier characterized by the kth iteration +.>Classification results of sample tags for sample workpiece i.
Classifier fluctuation weightsThe calculation method of (1) comprises the following steps:
computing a classification classifierClassification error rate +.>Wherein, the calculation formula of the classification error rate is: />；
In the method, in the process of the invention,sample label, characterized by sample workpiece i true for the kth iteration, < >>A bi-classification classifier characterized by the kth iteration +.>Classification result of sample label of sample workpiece i, if->And->Inequality->If->And->Equal +.>；
Calculating an optimal error rateWherein->The minimum function is obtained;
computing an optimal classifierClassifier fluctuation weights of +.>。
Sample fluctuation weightThe updating method of (1) comprises the following steps:
when k=1, the sample is fluctuated by weightSet to->；
At k>1, weighting the sample fluctuation according to the updated formulaUpdating to obtain sample fluctuation weight->The update formula is: />；
In the method, in the process of the invention,，/>normalized constant characterized by the kth iteration, < +.>The optimal classifier for the kth iteration is marked +.>Is used to determine the classifier fluctuating weights.
The implementation provides a fluctuation control classification modelAssuming sample workpieces 1,2,3,4,5,6, the machining precision sample data are respectively，/>，，/>，/>，。
Iteration 1:
1. selecting an optimal two-class classifier:
sample weight initialization toI.e.1/6, fluctuation threshold +.>=0.5,1.5,2.5,3.5,4.5；
If it is=0.5, resulting in a classification classifier +.>：S<0.5, out= 1;S>0.5, out= -1, at which time the sample workpiece 1,2,3,4,5,6 +.>，/>，/>，/>，/>，And->To->The error rate is then the error in classifying the sample workpieces 2 and 5；
If it is=1.5, resulting in a classification classifier +.>Out= 1;S>1.5, out= -1, at which time the sample workpiece 1,2,3,4,5,6 +.>，/>，/>，/>，/>，And->To->The error rate is the error rate when compared with the sample workpiece 5；
If it is=2.5, resulting in a classification classifier +.>：S<2.5, out= 1;S>2.5, out= -1, at which time the sample workpiece 1,2,3,4,5,6 +.>，/>，/>，/>，/>，And->To->The error rate is then the error in classifying the sample workpieces 3 and 5；
If it is=3.5, resulting in a classification classifier +.>：S<3.5, out= 1;S>3.5, out= -1, at which time the sample workpiece 1,2,3,4,5,6 +.>，/>，/>，/>，/>，And->To->Phase sampleThe workpieces 3,4 and 5 are classified as wrong, the error rate is that；
If it is=4.5, resulting in a classification classifier +.>：S<4.5, out= 1;S>4.5, out= -1, at which time the sample workpiece 1,2,3,4,5,6 +.>，/>，/>，/>，/>，/>And->To->The error rate is +.>；
Due toError rate is at least +.1.5>Then the optimal classification classifier ++>Is classified into two classes of classifierOut= 1;S>1.5, out= -1.
2. Calculating the weight of the optimal two-class classifier:；
3. updating the sample weight:
the sum of the weights of the sample workpieces 1,2,3,4,5,6 is；
Sample workpiece i=1, 2,3,4,6, label i If the classification is correct, the sample weight is:
；
sample workpiece i=5, label i Classification errors, the sample weights are:
；
at this time, the ripple classifier g= 0.8047The error rate of the fluctuation classifier G is +.>。
Iteration 2:
1. selecting an optimal two-class classifier:
if it is=0.5, resulting in a classification classifier +.>：S>0.5, out= 1;S<0.5, out= -1, the error rate is then；
If it is=1.5, resulting in a classification classifier +.>Out= 1;S>1.5, out= -1, error rate is +.>；
If it is=2.5, resulting in a classification classifier +.>：S>2.5, out= 1;S<2.5, out= -1, the error rate is then；
If it is=3.5, resulting in a classification classifier +.>Out= 1;S<3.5, out= -1, error rate is +.>；
If it is=4.5, resulting in a classification classifier +.>Out= 1;S>4.5, out= -1, error rate is +.>；
Due toError rate is at least +.4.5>Then the optimal classification classifier ++>Is classified into two classes of classifierOut= 1;S>4.5, out= -1.
2. Calculating the weight of the optimal two-class classifier:。
3. updating the sample weight:
the sum of the weights of the sample workpieces 1,2,3,4,5,6 is；
Sample workpiece i=1, 2,6, label i Classification is positiveThe sample weights are:
；
sample workpiece i=5, label i If the classification is correct, the sample weight is:
；
sample workpiece i=3, 4, label i Classification errors, the sample weights are:
；
at this time, the wave classifierThe error rate of the fluctuation classifier G is。
Iteration 3:
1. selecting an optimal two-class classifier:
if it is=0.5, resulting in a classification classifier +.>Out= 1;S>0.5, out= -1, error rate is +.>；
If it is=1.5, resulting in a classification classifier +.>Out= 1;S>1.5, out= -1, error rate is +.>；
If it is=2.5, resulting in a classification classifier +.>Out= 1;S<2.5, out= -1, error rate is +.>；
If it is=3.5, resulting in a classification classifier +.>Out= 1;S<3.5, out= -1, error rate is +.>；
If it is=4.5, resulting in a classification classifier +.>Out= 1;S>4.5, out= -1, error rate is +.>；/>
Due toWhen the error rate is 0.1875 at least when the value is 3.5, the optimal classification classifier is +.>Is classified into two classes of classifierOut= 1;S<3.5, out= -1.
2. Calculating the weight of the optimal two-class classifier:；
3. updating the sample weight:
the sum of the weights of the sample workpieces 1,2,3,4,5,6 is；
Sample workpiece i=3, 4, label i If the classification is correct, the sample weight is:
；
sample workpiece i=5, label i If the classification is correct, the sample weight is:
；
sample workpiece i=1, 2,6, label i Classification errors, the sample weights are:
；
at this time, the wave classifierThen wave classifier GError rate is +.>。
Is S<1.5, out= 1;S>1.5, out= -1.
Is S<4.5, out= 1;S>4.5, out= -1.
Is S>3.5, out= 1;S<3.5, out= -1.
Classifying all sample workpieces i according to the fluctuation classifier G, wherein the average classification is correct, if the error rate of the fluctuation classifier G is 0/6=0, stopping iteration, and if the error rate of the fluctuation classifier G is 0/6=0, stopping iterationAs the fluctuation control classification model, the classification error rate of the fluctuation control classification model on the sample workpiece is 0, so that the high accuracy of classification of the target workpiece is ensured.
The method for obtaining the fluctuation control category comprises the following steps:
inputting the machining precision representation data S of the target workpiece into a fluctuation control classification model to obtain a classification result of a sample label of the target workpieceWherein, the method comprises the steps of, wherein,
if it isThe fluctuation control category of the target workpiece is repair;
if it isThe fluctuation control class of the target workpiece is not repaired.
And S3, re-processing the target workpiece according to the fluctuation control category so that the processing precision of the target workpiece meets the processing requirement.
In step S3, the method for reprocessing the target workpiece includes:
comparing the size precision representation data, the position precision representation data and the shape precision representation data in the machining precision representation data of the target workpiece with the fluctuation control class as repair with the size precision representation data, the position precision representation data and the shape precision representation data of the standard workpiece respectively to obtain the repair field of the target workpiece with the fluctuation control class as repair;
and placing the target workpiece with the fluctuation control category being repaired in a repair neighborhood for re-machining so as to realize the repair of the target workpiece until the machining precision reaches the machining requirement of the standard part, and only repairing a certain repair field without integral repair, thereby improving the repair efficiency.
The dimension precision representation data, the position precision representation data and the shape precision representation data are normalized before the weight summation is carried out, and the data dimension is unified.
As shown in fig. 2, based on the above-mentioned precision fluctuation control method for automatic lathe machining, the present invention provides a control system comprising:
the quality inspection device 1 is used for monitoring processing precision characterization data of a target workpiece positioned at a workpiece outlet end of the automatic lathe;
a model building unit 2 for creating machining precision sample data from sample workpieces with known fluctuation control categories, and building a fluctuation control classification model based on the machining precision sample data, wherein the fluctuation control classification model is used for obtaining the fluctuation control category of the target workpiece according to machining precision characterization data;
and a re-processing unit 3 re-processes the target workpiece according to the fluctuation control category so that the processing precision of the target workpiece meets the processing requirement.
According to the invention, the processing precision representation data is obtained by carrying out weight summation processing on the size precision representation data, the position precision representation data and the shape precision representation data of the target workpiece, so that the automatic regulation and control of the constituent elements of the processing precision is realized for various workpiece types adapting to the target workpiece, the application of quality inspection on various types of workpieces can be realized, the adaptability of quality inspection during expansion is realized, in addition, a fluctuation control classification model is established, the fluctuation control category of the target workpiece can be accurately obtained, and the quality inspection accuracy is improved.
The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application, the scope of which is defined by the claims. Various modifications and equivalent arrangements may be made to the present application by those skilled in the art, which modifications and equivalents are also considered to be within the scope of the present application.
Claims (9)
1. The precision fluctuation control method for automatic lathe machining is characterized by comprising the following steps of:
step S1, arranging a quality inspection device at a workpiece outlet end of an automatic lathe, wherein the quality inspection device is used for monitoring processing precision characterization data of a target workpiece positioned at the workpiece outlet end of the automatic lathe;
s2, creating machining precision sample data by using sample workpieces with known fluctuation control categories, and constructing a fluctuation control classification model based on the machining precision sample data, wherein the fluctuation control classification model is used for obtaining the fluctuation control category of a target workpiece according to machining precision characterization data;
s3, re-machining the target workpiece according to the fluctuation control category so that the machining precision of the target workpiece meets the machining requirement;
in step S2, the method for creating the fluctuation control classification model includes:
setting a fluctuation thresholdAnd constructing a fluctuation classifier G with a fluctuation threshold, wherein the construction method of the fluctuation classifier comprises the following steps:
step S201. Setting the iteration number k, initializing to 1, and setting a sample fluctuation weight e for a sample workpiece i ik ；
Step S202, if the processing precision of the sample workpiece i represents the data S i Less than the fluctuation threshold S d Then the classification classifier { g jk |j∈[1,n-1]Marking the classification result of the sample label of the sample workpiece i as out ik =1 or out ik ＝-1；
If the processing precision of the sample workpiece i represents data S i Greater than the fluctuation threshold S d Then the classification classifier { g jk |j∈[1,n-1]Marking the classification result of the sample label of the sample workpiece i as out ik =1 or out ik ＝-1；
Step S203, sequentially obtaining n-1 classification classifiers { g } jk |j∈[1,n-1]N-1 classification classifiers { g } are sequentially calculated jk |j∈[1,n-1]Classification error Rate { p } jk |j∈[1,n-1]And } wherein,
if it isOutputting the current fluctuation classifier G as a fluctuation control classification model, and stopping iteration;
if it isThen the classification error rate p is selected jk Minimum two-classification classifier g jk Marked g as optimal classifier k The optimal error rate is marked as p k And calculate the optimal classifier label as g k Classifier fluctuation weights E k Sample fluctuation weight e for sample workpiece i ik Updating to obtain a sample fluctuation weight e of the sample workpiece i ik+1 And update the wave classifier tok is added with 1 automatically, and the step S202 is returned;
wherein l and j are measurement constants, and e is not significant ik 、e ik+1 Sample fluctuation weight, g, of sample workpiece i characterized by the kth, k+1th iteration jk The j-th classifier, p, characterized by the k-th iteration jk Classification error rate, out, of the jth classification classifier characterized by the kth iteration ik Two-classification classifier g characterized by kth iteration jk Classification results of sample tags for sample workpiece i.
2. The precision fluctuation control method for automatic lathe machining according to claim 1, wherein: in the step S1, the method for monitoring the machining precision characterization data includes:
the quality inspection device acquires data of machining size, machining position and machining shape of the target workpiece to obtain size precision representation data, position precision representation data and shape precision representation data respectively;
performing weight summation processing on the size precision representation data, the position precision representation data and the shape precision representation data to obtain the processing precision representation data so as to realize that constituent elements for independently regulating and controlling the processing precision are used for various workpiece types adapting to target workpieces, wherein the weight summation processing formula is as follows:
S＝ω 1 *s 1 +ω 2 *s 2 +ω 3 *s 3 ，
wherein S represents machining precision representation data, S 1 、s 2 Sum s 3 Respectively characterizing as size precision characterizing data, position precision characterizing data and shape precision characterizing data, omega 1 、ω 2 And omega 3 The characteristics are respectively the weight of the size precision characteristic data, the weight of the position precision characteristic data and the weight of the shape precision characteristic data.
3. The precision fluctuation control method for automatic lathe machining according to claim 2, wherein: in the step S2, the method for creating the machining precision sample data includes:
obtaining a plurality of sample workpieces, wherein the plurality of sample workpieces comprise positive sample workpieces and negative sample workpieces with the same number, the positive sample workpieces are characterized in that the fluctuation control type is a repaired workpiece, and the negative sample workpieces are characterized in that the fluctuation control type is an unrepaired workpiece;
the method comprises the steps of sequentially collecting processing precision representation data of a sample workpiece, taking the processing precision representation data of the sample workpiece as sample characteristics of the sample workpiece, taking the fluctuation control category as a sample label of the sample workpiece, and combining the sample characteristics and the sample label to construct processing precision sample data, wherein the processing precision sample data representation method comprises the following steps:
Z i ＝(S i ,label i )；
wherein Z is i Machining precision sample data characterized as sample workpiece i, S i Sample feature/machining precision characterization data characterizing sample workpiece i, label i Sample tag/wave control class characterized as sample artifact i, i is characterized as sample artifact's distinguishing number, i ε [1, n]N is characterized by the total number of sample workpieces, label i Wave control class/sample tag for repair characterized by =1 for sample workpiece i, label i = -1 is characterized as the ripple control class/sample tag of sample workpiece i is not repaired.
4. A precision fluctuation control method for automatic lathe machining according to claim 3, wherein: the classifier fluctuation weight E k The calculation method of (1) comprises the following steps:
calculating the two-classification classifier { g } jk |j∈[1,n-1]Classification error Rate { p } jk |j∈[1,n-1]And the calculation formula of the classification error rate is as follows:
in the label ik Sample tag, out, characterizing the sample workpiece, i, as true of the kth iteration ik Two-classification classifier g characterized by kth iteration jk For sample workerIf out, the classification result of the sample label of the part i ik And label ik Not equal, then I (out ik ≠label ik ) =1, if out ik And label ik Equal, then I (out ik ≠label ik )＝0；
Calculating an optimal error rate p k ＝min(p jk ) Wherein min is a minimum function;
calculating an optimal classifier g k Classifier fluctuation weights of (a)
5. The method for controlling precision fluctuation of automatic lathe machining according to claim 4, wherein: the sample fluctuation weight e ik The updating method of (1) comprises the following steps:
when k=1, the sample is fluctuated by a weight e ik Is set as
At k>1, the sample is fluctuated by a weight e according to an updated formula ik Updating to obtain sample fluctuation weight e ik+1 The update formula is as follows:
in the method, in the process of the invention,Y k normalized constant characterized by the kth iteration, E k The optimal classifier for the kth iteration is labeled g k Is used to determine the classifier fluctuating weights.
6. The method for controlling precision fluctuation of automatic lathe machining according to claim 5, wherein: the method for obtaining the fluctuation control category comprises the following steps:
inputting the machining precision representation data S of the target workpiece into a fluctuation control classification model to obtain a classification result out of a sample label of the target workpiece s Wherein, the method comprises the steps of, wherein,
if out s =1, then the fluctuation control class of the target workpiece is repair;
if out s = -1, then the fluctuation control category of the target workpiece is not repaired.
7. The method according to claim 6, wherein in the step S3, the method for reworking the target workpiece includes:
comparing the size precision representation data, the position precision representation data and the shape precision representation data in the machining precision representation data of the target workpiece with the fluctuation control class as repair with the size precision representation data, the position precision representation data and the shape precision representation data of the standard workpiece respectively to obtain the repair field of the target workpiece with the fluctuation control class as repair;
and placing the target workpiece with the fluctuation control category being repaired in a repair neighborhood for re-machining so as to realize the repair of the target workpiece until the machining precision reaches the machining requirement of the standard component.
8. The method according to claim 7, wherein the dimensional accuracy characterizing data, the position accuracy characterizing data, and the shape accuracy characterizing data are normalized before the weight summing.
9. A control system of the precision fluctuation control method of automatic lathe machining according to any one of claims 1 to 8, characterized by comprising:
the quality inspection device (1) is used for monitoring processing precision characterization data of a target workpiece positioned at a workpiece outlet end of the automatic lathe;
the model building unit (2) is used for building machining precision sample data from sample workpieces with known fluctuation control categories, and building a fluctuation control classification model based on the machining precision sample data, wherein the fluctuation control classification model is used for obtaining the fluctuation control category of a target workpiece according to machining precision characterization data;
and the re-machining unit (3) re-machines the target workpiece according to the fluctuation control type so that the machining precision of the target workpiece meets the machining requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211119770.9A CN115204321B (en) | 2022-09-15 | 2022-09-15 | Precision fluctuation control method and system for automatic lathe machining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211119770.9A CN115204321B (en) | 2022-09-15 | 2022-09-15 | Precision fluctuation control method and system for automatic lathe machining |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115204321A CN115204321A (en) | 2022-10-18 |
| CN115204321B true CN115204321B (en) | 2023-07-25 |
Family
ID=83572528
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211119770.9A Active CN115204321B (en) | 2022-09-15 | 2022-09-15 | Precision fluctuation control method and system for automatic lathe machining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115204321B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113569957A (en) * | 2021-07-29 | 2021-10-29 | 中国工商银行股份有限公司 | Object type identification method and device of business object and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109472057B (en) * | 2018-10-16 | 2023-05-23 | 浙江大学 | Product processing quality prediction device and method based on cross-process implicit parameter memory |
| JP7458183B2 (en) * | 2019-12-26 | 2024-03-29 | 川崎重工業株式会社 | Energy demand fluctuation pattern classification method and system |
-
2022
- 2022-09-15 CN CN202211119770.9A patent/CN115204321B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113569957A (en) * | 2021-07-29 | 2021-10-29 | 中国工商银行股份有限公司 | Object type identification method and device of business object and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115204321A (en) | 2022-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6285494B2 (en) | Measurement sample extraction method with sampling rate determination mechanism and computer program product thereof | |
| CN113762410B (en) | Key welding process parameter prediction method based on machine learning | |
| ElMaraghy et al. | Integrated inspection and machining for maximum conformance to design tolerances | |
| TWI663569B (en) | Quality prediction method for multi-workstation system and system thereof | |
| CN115204321B (en) | Precision fluctuation control method and system for automatic lathe machining | |
| CN112580935A (en) | Industrial product production process traceability analysis method based on machine vision | |
| US20210311440A1 (en) | Systems, Methods, and Media for Manufacturing Processes | |
| CN108537249B (en) | Industrial process data clustering method for density peak clustering | |
| Blumberg et al. | Deformation error compensation of industrial robots in single point incremental forming by means of data-driven stiffness model | |
| Cheung et al. | A framework of a virtual machining and inspection system for diamond turning of precision optics | |
| Bedoui et al. | New method for the systematic determination of the model’s base of time varying delay system | |
| CN109885027B (en) | Industrial process fault diagnosis method based on bidirectional two-dimensional sparse orthogonal discriminant analysis | |
| TWI692695B (en) | Feedback system for improving process quality of product and method thereof | |
| Lutz et al. | Towards a digital twin: Simulation and residual stress analysis in aerospace composite structures assembly | |
| EP4118512A1 (en) | Systems, methods, and media for manufacturing processes | |
| Hashemi et al. | To improve machining fixture design: A case based reasoning paradigm | |
| Vukelić et al. | Computer-aided design and optimization of fixtures for plastic parts machining | |
| Khan et al. | Finding correlations between 3-d surfaces: A study in asymmetric incremental sheet forming | |
| Lu | Investigation of control of the incremental forming processes | |
| Huseynov | Agent technology at computing option of flexible manufacture system element and composed structure | |
| CN115502443B (en) | Intelligent drilling and compounding integrated machine for multi-station fixed ball valve seat ring | |
| CN114186359B (en) | Intelligent analysis method and system for manufacturability of mechanical product | |
| CN117891203B (en) | Method and device for regulating and controlling reentrant manufacturing system, storage medium and electronic equipment | |
| CN115179445B (en) | CNC (computer numerical control) processing method for vacuum cavity sealing piece based on space positioning error model | |
| JP2828870B2 (en) | How to adjust the characteristics of assemblies |
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 |