CN110376963B - Closed-loop control precision machining method and system based on in-situ detection - Google Patents
Closed-loop control precision machining method and system based on in-situ detection Download PDFInfo
- Publication number
- CN110376963B CN110376963B CN201910646784.8A CN201910646784A CN110376963B CN 110376963 B CN110376963 B CN 110376963B CN 201910646784 A CN201910646784 A CN 201910646784A CN 110376963 B CN110376963 B CN 110376963B
- Authority
- CN
- China
- Prior art keywords
- workpiece
- processing
- machining
- error
- key point
- 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
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
- G05B19/401—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 measuring, e.g. calibration and initialisation, measuring workpiece for machining purposes
- G05B19/4015—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 measuring, e.g. calibration and initialisation, measuring workpiece for machining purposes going to a reference at the beginning of machine cycle, e.g. for calibration
-
- 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
- G05B2219/34—Director, elements to supervisory
- G05B2219/34242—For measurement only
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Numerical Control (AREA)
Abstract
The invention discloses a closed-loop control precision machining method and a closed-loop control precision machining system based on in-situ detection, and belongs to the field of precision machining. The method comprises the following steps: s1, carrying out a primary processing procedure by the machine tool cutter according to a preset processing path; s2, acquiring the external dimension, the feature key point dimension and the processing parameter of the workpiece in situ, inputting the external dimension, the feature key point dimension and the processing parameter into the neural network intelligent compensation model in real time, calculating an error compensation value C of the cutter, and optimizing the processing path of the next step according to the error compensation value C so as to perform real-time dynamic compensation on the processing of the next step; and S3, taking the next machining path optimized in the step S2 as a preset machining path in the step S1, and repeating the steps S1 and S2 until the machining is finished. The invention obtains the local key size in the processing process in real time through in-situ measurement, calculates the processing error, calculates the processing compensation value based on the real-time fitting of the neural network algorithm, and automatically optimizes the processing path, thereby reducing the processing error and realizing the precise processing of real-time closed-loop control.
Description
Technical Field
The invention belongs to the field of precision machining, and particularly relates to a real-time closed-loop control precision part machining system and method based on an in-situ detection technology.
Background
With the development of machining and manufacturing technology, the production requirements of mechanical parts are higher and higher, for example, in the fields of die manufacturing, lithium battery equipment and the like, the machining of the shape structure of a workpiece tends to be complicated and precise. At present, the processing modes of the numerical control machine tool mainly comprise open-loop, semi-closed-loop and closed-loop processing. The open-loop processing is that after the processing is finished, workers manually measure and feed back unqualified products to process personnel, and process parameters are adjusted; the semi-closed loop machining adopts a pulse encoder and other parts on a machine tool to feed back parameters such as the rotating speed of a motor and the like, and controls the machining process; and the closed-loop processing mode adopts the feedback position of elements such as a photoelectric sensor, a limiter and the like to realize closed-loop control. However, the machining method does not consider the machining and cutting process of the numerical control machine tool, the detection result has time lag, the machining efficiency is reduced, and the precise machining of real-time closed-loop control cannot be realized.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a closed-loop control precision machining method and a closed-loop control precision machining system based on in-situ measurement, and aims to perform in-situ measurement through image acquisition and laser ranging, obtain local key dimensions in the machining process in real time, calculate machining errors, fit and calculate machining compensation values in real time based on a neural network algorithm, and automatically optimize a machining path, so that the machining errors are reduced, and real-time closed-loop control precision machining is realized.
In order to achieve the above object, according to one aspect of the present invention, there is provided a closed-loop control precision machining method based on in-situ detection, comprising the steps of:
s1, carrying out a primary processing procedure by the machine tool cutter according to a preset processing path;
s2, after the step S1 is completed, or in real time during the execution of the step S1, acquiring the external dimension, the dimension of the shape key point and the processing parameter of the workpiece in situ, inputting the external dimension, the dimension of the shape key point and the processing parameter into the intelligent compensation model of the neural network in real time, calculating an error compensation value C of the cutter, and optimizing the processing path of the next step according to the error compensation value C so as to perform real-time dynamic compensation on the processing of the next step; wherein:
the overall dimension, the dimension of the morphology key points and the processing parameters comprise: shape S of workpiece, outer dimension H, and material M of workpiecepThe accuracy requirement of the workpiece A and the material quality of the cutter MtSpeed V of main shaft of machine toolmTool feed speed VtCutting fluid F and machining step WsAnd processing error data; the machining error data includes shape error of the workpieceSAnd the shape and size errors of the workpieceH，SAndHthe difference values of the measured values and the design values of the shape S and the feature size H of the workpiece are respectively.
The neural network intelligent compensation model is obtained by taking the external dimension and the shape key point dimension of a processed workpiece as input and taking a corresponding error compensation value C as output and training a neural network
And S3, taking the next machining path optimized in the step S2 as a preset machining path in the step S1, and repeating the steps S1 and S2 until the machining is finished.
Further, in step S2, key points to be collected on the surface of the workpiece are preset, during measurement, the outline size of the workpiece is obtained as the shape S of the workpiece by taking a picture with a camera, and the height size of the key points on the surface of the workpiece is obtained as the shape size H corresponding to the key points by laser ranging.
Further, in step S2, laser ranging is performed on the corresponding position on the surface of the workpiece according to a preset key point to obtain a height error of the key point, data prediction interpolation is performed based on the radial basis function, more interpolation key points are generated according to the preset key point, then the height error of the interpolation key point is calculated, and the height error of the preset key point and the interpolation key point is used as an error of the feature size HHAnd performing compensation calculation.
Further, in step S2, a key point is set every i mm from the start point to the end point on the machining path, N key points are set in advance, and the coordinate value x of the preset key point is obtained as { x ═ x {1,x2,…xi,…xN}; selecting the radial basis function as:
the interpolation expression for the radial basis function is then:
wherein r ═ x-xiIs coordinate x and coordinate value xiA second norm of (d); sigma represents the expansion constant of the radial basis function, reflects the width of the radial basis function image, and the smaller sigma is, the narrower width is, and the more selective the function is; c. C0Is a constant term, c1Is a coefficient of a first order term, λiTo correspond to xiRadial basis function of
The coefficient of (a);
according to the preset N key points and the measured shape and size errors of the ith key point
C in f (x) is calculated by matrix SVD decomposition or least square method0、c1And λiTherefore, an interpolation expression is obtained, and the shape and size errors of other interpolation key points on the machining path are further solved.
In order to achieve the above object, the present invention further provides a closed-loop control precision machining system based on in-situ detection, comprising: the system comprises an industrial camera, an industrial personal computer, a laser range finder and a closed-loop control program module; wherein:
the industrial camera is used for shooting a plane image of the workpiece, and the laser range finder is used for carrying out laser range finding on a preset key point on the surface of the workpiece to obtain the height of the workpiece;
the industrial personal computer is used for controlling the work of the industrial camera and the laser range finder and calling a closed-loop control program module to perform error compensation;
when the closed-loop control program module is called by an industrial personal computer, the closed-loop control precision machining method based on in-situ detection is executed.
Further, the industrial camera is provided with an optical lens and an annular light source; the annular light source consists of a plurality of illuminating lamps which are annularly arranged; all the illuminating lamps are uniformly distributed around the optical lens in a ring shape.
In general, compared with the prior art, the above technical solution contemplated by the present invention can obtain the following beneficial effects:
(1) the machine tool machining and cutting process is considered in size detection and closed-loop control, and the anti-interference capability of a measuring system on external disturbance such as cutting fluid, chips and the like is improved through a size detection method of machine vision and laser ranging. And the shape error is directly obtained by the difference value between the plane figure calculation obtained by the vision and the design value, and the shape error is obtained by the laser ranging, so that compared with the traditional three-dimensional modeling error calculation method, the method can also greatly improve the operation efficiency, and further improve the real-time performance of closed-loop control.
(2) The size error measurement and the machine tool control are combined to form a closed loop, so that the machining process is controlled in real time, the machining deviation is compensated in time, the machining efficiency and the precision are improved, and certain instructive significance is realized on other precision machining scenes with similar conditions.
(3) According to the method, only errors of a plurality of key points need to be actually measured, the positions of interpolation key points can be obtained through an interpolation method, so that the errors of the interpolation key points can be obtained, the size detection speed can be improved, the shape and size errors of the workpiece can be calculated only through the errors of the key points, the detection efficiency is improved, and the detection precision can be guaranteed.
Drawings
FIG. 1 is a schematic view of the mounting positions of a camera, a lens and a ring light source in a preferred embodiment of the invention;
FIG. 2 is a schematic plan view of the annular light source of FIG. 1;
FIG. 3 is a schematic illustration of machine vision planform dimension measurement in a preferred embodiment of the present invention;
FIG. 4 is a schematic view of laser measurement point selection according to the present invention;
FIG. 5 is a schematic diagram of the radial basis function interpolation calculation of the present invention;
FIG. 6 is a schematic diagram of an in-situ inspection and closed loop processing system according to the present invention;
fig. 7 is a schematic flow diagram of a preferred embodiment of the present invention.
The same reference numbers will be used throughout the drawings to refer to the same or like elements or structures, wherein:
1-industrial camera, 2-optical lens, 3-annular light source and 4-LED lamp.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
A closed-loop process control system and method using in-place measurement techniques is disclosed. The arrangement of the industrial camera and the light source of the measuring part is shown in figure 1, and the precise measuring point of the laser measuring head is selected as shown in figure 2. The measuring part of the system is controlled by control software to work and read feedback images and measurement data, then the feedback images and the measurement data are compared with theoretical process dimensions to calculate machining errors, the shape dimension errors are densified by an interpolation algorithm, the control software calculates the machine tool cutter compensation by using a neural network intelligent model and then feeds the machine tool cutter compensation back to a numerical control machine, and the machine tool cutter corrects the machining feeding of the next step according to the machining compensation.
The working principle of the image acquisition part is that an annular light source arranged on a camera is controlled by a control software system through a light source controller to emit flash light with preset brightness, the camera is controlled by software to shoot at the same time, an obtained image is transmitted to the software, the key size of a workpiece is measured through image preprocessing, image segmentation, edge fitting and other technologies, and a processing error is calculated. Because the machine vision can measure all the key sizes of the whole measurement part at the same time and is insensitive to the interference of cutting fluid and cutting chips, all the sizes of the part of the workpiece can be measured quickly and accurately, but the monocular machine vision measurement component cannot sense the depth information of the workpiece, so the height (depth) appearance size processing error of the workpiece is measured by using a laser measuring head. The laser measuring head and the machine vision measuring component work simultaneously, the laser measuring head uses a triangular measuring method to accurately measure the size processing error of the key point of the height (depth) appearance of the workpiece, the result is returned to the control software, and the software calculates the processing error of more key points in the height (depth) appearance of the workpiece through a radial basis function interpolation algorithm. The software processes the two machining errors in a unified way, the compensation value is calculated by a neural network intelligent adjustment algorithm and is converted into a compensation result required by a CNC system, and finally a servo system of the machine tool is controlled to realize the next machining parameter correction of the machine tool, so that the accuracy of the workpiece size is ensured.
The system, method and specific applications of the present invention are described in detail below with reference to a specific embodiment.
As shown in fig. 1, 2 and 6, the present invention provides a closed-loop control precision machining system based on in-situ measurement, which adopts an in-situ measurement tool of an industrial camera and a laser ranging head to measure the local critical dimension of a product in the machining process in real time, and matches with a related software system to compare the process dimension with the machining result dimension, calculate the machining error, and calculate the machining compensation value by fitting through a neural network algorithm in the matching software, and automatically optimize the machining path to reduce the machining error. The working states of the industrial camera and the measuring head are controlled by the PLC controller and exchange data with corresponding software systems.
Specifically, the present embodiment uses a combined measurement method based on machine vision and a laser ranging head, which includes hardware facilities such as an industrial camera, an industrial lens, an annular light source, an industrial personal computer, a laser ranging head, and other corresponding fixtures. Preferably, in this embodiment:
the industrial camera is a common industrial camera, the pixel resolution is selected based on the overall dimension of a processed workpiece, and in order to meet the requirements of rapidity, accuracy and the like of measurement, the system adopts a 2000W pixel CMOS camera;
the camera is provided with a common optical lens, the size of the focal length needs to meet the size of a workpiece, and a focal length fixed-focus lens of 8mm is adopted;
the light source is a common annular light source, a white light source is adopted in the system and used for providing illumination when an industrial camera is used for shooting the surface of a workpiece, and the brightness is controlled by a matched software system on an industrial personal computer through a light source controller. The annular light source is arranged in front of the industrial lens and used for collecting images along with the camera.
The laser range finder is an industrial laser range finder, the size is measured by adopting a high-precision triangular method so as to realize high-precision measurement, and the measurement precision can reach 1 mu m.
The machine vision measurement assembly is mainly used for measuring the local key size of a workpiece, when the machine vision measurement assembly works, supporting software on an industrial personal computer controls a light source to flash and a camera to take a picture to obtain a picture of the workpiece, the picture is transmitted to control software through an Ethernet line, the software processes the image of the workpiece through a sub-pixel-level precision image processing method such as a centroid method and edge fitting, the measured local key size of the workpiece is obtained through measurement, the measured local key size is compared with the theoretical size of a machining process, machining errors are solved, and the machining errors are used by an intelligent optimization algorithm in the software. The pixel calculation precision can reach micron level under the image processing technology of twenty million pixel resolution ratio at sub-pixel level precision.
As shown in fig. 3, the measurement of the machine vision is limited to a two-dimensional plane, and only the profile size can be measured, and the depth information of the workpiece cannot be measured, so that the defect is compensated by using a laser measuring head, which measures the profile height (depth) size of the workpiece and calculates a corresponding error value.
As shown in fig. 4 and 5, when the laser ranging head measures the surface topography of a workpiece, the corresponding position of the surface of the workpiece is measured according to a key point preset by software, the processing path and the size of the workpiece input in advance in the software are compared to calculate the error of the key point, and then a precise interpolator based on a radial basis function in the software is used for data prediction interpolation, so that the error values of more key points in the surface topography of the workpiece are calculated, and are recorded by the software and are compensated and calculated by an intelligent algorithm in the software. In the machining process, the error values on one machining path are correlated, so that the specific shape and size of the machining path do not need to be accurately fitted, and only the relation between the machining errors needs to be searched. Setting N key points on a machining path, setting a sampling point every i mm from the starting point to the ending point of the machining path, and automatically setting the value of i according to the machining condition to obtain the coordinate value x of each key point, wherein the coordinate value x is { x ═ x }1,x2,……xN(e.g., abscissa values in fig. 4). Selecting a polynomial basis function as follows:
the interpolation expression for the radial basis function is then:
wherein r ═ x-xiIs coordinate x and coordinate value xiA second norm of (d); sigma represents the expansion constant of the radial basis function, reflects the width of the radial basis function image, and the smaller sigma is, the narrower width is, and the more selective the function is; c. C0Is a constant term, c1Is a coefficient of a first order term, λiTo correspond to xiRadial basis function of
The coefficient of (a);
according to the preset N key points and the measured shape and size errors of the ith key point
C in f (x) is calculated by matrix SVD decomposition or least square method0、c1And λiTherefore, an interpolation expression is obtained, and the shape and size errors of other interpolation key points on the machining path are further solved. The method can improve the size detection speed, can calculate the shape and size errors of the workpiece by only needing errors of a plurality of key points, improves the detection efficiency and ensures the detection precision.
The control software system uses an intelligent compensation model algorithm based on a neural network, and has a plurality of error influence factors in the machining process of the machine tool, including the shape, the external dimension, the material quality of the workpiece, the accuracy requirement of the workpiece, the rotating speed of a machine tool spindle, the feeding speed of a cutter, the material of the machining cutter, cutting fluid and the like. The neural network algorithm is based on industrial processing big data, classifies and collates processing process data accumulated in the past processing production practice, and analyzes the relation between the generation of errors and processing conditions in the processing process. The obtained data is processed according to the shape S, the external dimension H and the material M of the workpiecepWorkpiece accuracy requirement A, knifeWith material quality MtSpeed V of main shaft of machine toolmTool feed speed VtCutting fluid F and machining step WsAnd processing error data generated under these processing conditions as input data, and an error compensation value C set by a process specialist during these processing as output. And establishing a three-layer neural network with 9 output layer nodes, 12 hidden layer nodes and 1 output layer nodes. And training the accumulated processing data by using a gradient descent algorithm of the SGD to obtain a neural network model for predicting the compensation value.
The control software adopts closed-loop control with error control, the size error and the shape error obtained by calculation are used as input data, and a trained neural network intelligent algorithm is adopted to calculate the required compensation value of the machine tool. The compensation value calculated by the network model is amplified to control a servo motor to drive moving parts such as a machine tool cutter and the like to compensate and feed to the command position, and the cutter feeding compensation precision is improved by using an interpolation operation technology.
The lower limit of the sampling frequency of the combined measuring system is in the machine vision part, the maximum sampling frame rate of the industrial camera is 25FPS, and the sampling frequency of the laser ranging head can reach thousands of times per second. The sampling frequency and the shape key point of the measuring system are also calculated by another neural network model in the control software, and real-time calculation and adjustment are carried out according to the actual state of the workpiece, the processing requirement, the rotating speed of the main shaft of the machine tool, the feeding speed of the cutter and other processing parameters, so as to ensure the timeliness and the accuracy of measurement.
The workflow of the present invention is illustrated below in conjunction with FIG. 7:
a preparation stage: fixing a workpiece to be detected on a machine tool workbench, and initializing a detection system;
the method comprises the following steps of calibrating the size of information of a machine vision measuring system by taking the surface of a workpiece to be measured as a reference, and calibrating and zeroing a laser measuring head by taking the upper surface of a workbench as a reference;
and (3) a processing stage:
s1, processing the workpiece to be measured according to a preset processing route after the numerical control machine tool finishes tool setting;
s2, error measurement and compensation calculation:
s21, when the machine tool is used for machining, the control software measures the size of the workpiece according to the preset sampling frequency and the sampling point position; the method comprises the following steps:
s211, the control software sends an instruction to enable the machine vision measuring assembly to work, the annular light source emits preset flash under the control of the controller, and the industrial camera captures an image of the surface of the workpiece and transmits the image to the control software;
s212, the control software quickly calculates the overall dimension of the workpiece according to the returned image, calculates the error and records the error; simultaneously controlling the laser measuring head to work;
s213, measuring the preset key point size of the height (depth) appearance of the workpiece by the laser measuring head under the control of the sampling frequency of the control software, measuring the machining error of the key point once by the measuring head along the machining path from the machining starting point according to the set size interval (such as 10mm), and returning the result to the control software;
s214, the control software uses a radial basis function interpolation algorithm to solve an interpolation expression according to the height (depth) measurement error result, then more key points are selected on the machining path, 2 points are selected in the interval distance of the previous key points, and more morphology key point errors are calculated.
S22, the neural network model in the control software takes these error values and the processing condition parameters as inputs, i.e. the input vector X ═ S, H, Mp,A,Mt,Vm,Vt,F,Ws,]Intelligently calculating a compensation value required by the cutter according to the neural network model, outputting y as C, feeding the converted signal back to the numerical control machine tool, and correcting the next processing;
and S3, the numerical control machine tool circulates the processing-measuring process until finishing the processing of all the working procedures.
In other embodiments, the sampling frequency of the measurement system and the selection of the morphology key points can be intelligently and automatically regulated and controlled by control software, and can be regulated according to the IT grade required by the machining accuracy grade and the rotating speed V of the main shaft of the machine toolmAnd the tool feed speed VtPerforming the processing according to the processing parametersThe time is adjusted by calculation to ensure the timeliness and accuracy of measurement.
In general, the method adopts an in-situ measuring tool of an industrial camera and a laser ranging head to measure the local key dimension of a product in the processing process in real time, and is matched with a relevant software system to compare the process dimension with the processing result dimension, calculate the processing error, calculate the processing compensation value through the neural network algorithm fitting in the matched software, automatically optimize the processing path, reduce the processing error, have extremely high control and processing precision, are particularly suitable for the processing process of precision parts in the industries of molds and lithium battery equipment, and can realize the precision processing of complex curved surfaces of core parts.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (3)
1. A closed-loop control precision machining method based on in-situ detection is characterized by comprising the following steps:
s1, carrying out a primary processing procedure by the machine tool cutter according to a preset processing path;
s2, after the step S1 is completed, or in real time during the execution of the step S1, acquiring the external dimension, the dimension of the shape key point and the processing parameter of the workpiece in situ, inputting the external dimension, the dimension of the shape key point and the processing parameter into the intelligent compensation model of the neural network in real time, calculating an error compensation value C of the cutter, and optimizing the processing path of the next step according to the error compensation value C so as to perform real-time dynamic compensation on the processing of the next step; wherein:
the overall dimension, the dimension of the morphology key points and the processing parameters comprise: shape S of workpiece, outer dimension H, and material M of workpiecepThe accuracy requirement of the workpiece A and the material quality of the cutter MtSpeed V of main shaft of machine toolmTool feed speed VtCutting fluid F and machining step WsAnd processing error data; the machining error data includes shape error of the workpieceSAnd the shape and size errors of the workpieceH，SAndHthe difference values of the measured values and the design values of the shape S and the shape dimension H of the workpiece are respectively;
the neural network intelligent compensation model is obtained by training a neural network by taking the external dimension and the shape key point dimension of a processed workpiece as input and taking a corresponding error compensation value C as output;
s3, taking the next machining path optimized in the step S2 as a preset machining path in the step S1, and repeating the steps S1 and S2 until the machining is finished;
in the step S2, key points needing to be collected on the surface of the workpiece are preset, during measurement, the outline size of the workpiece is obtained by taking a picture through a camera and is used as the shape S of the workpiece, and the height size of the key points on the surface of the workpiece is obtained through laser ranging and is used as the shape size H corresponding to the key points;
in step S2, laser ranging is performed on the corresponding position on the surface of the workpiece according to a preset key point to find the height error of the key point, data prediction interpolation is performed based on the radial basis function, more interpolation key points are generated according to the preset key point, then the height error of the interpolation key point is calculated, and the height error of the preset key point and the interpolation key point is used as the error of the feature size HHPerforming compensation calculation;
in step S2, a key point is set every i mm from the start point to the end point on the machining path, N key points are set in advance, and the coordinate value x of the preset key point is obtained as { x ═ x {1,x2,…xi,…xN}; selecting the radial basis function as:
the interpolation expression for the radial basis function is then:
wherein r ═ x-xiIs coordinate x and coordinate value xiA second norm of (d); sigma represents the expansion constant of the radial basis function, reflects the width of the radial basis function image, and the smaller sigma is, the narrower width is, and the more selective the function is; c. C0Is a constant term, c1Is a coefficient of a first order term, λiTo correspond to xiRadial basis function of
The coefficient of (a);
according to the preset N key points and the measured shape and size errors of the ith key point
C in f (x) is calculated by matrix SVD decomposition or least square method0、c1And λiTherefore, an interpolation expression is obtained, and the shape and size errors of other interpolation key points on the machining path are further solved.
2. A closed-loop control precision machining system based on in-situ detection is characterized by comprising: the system comprises an industrial camera, an industrial personal computer, a laser range finder and a closed-loop control program module; wherein:
the industrial camera is used for shooting a plane image of the workpiece, and the laser range finder is used for carrying out laser range finding on a preset key point on the surface of the workpiece to obtain the height of the workpiece;
the industrial personal computer is used for controlling the work of the industrial camera and the laser range finder and calling a closed-loop control program module to perform error compensation;
wherein the closed-loop control program module, when invoked by the industrial control machine, performs the closed-loop control precision machining method based on in-situ detection as recited in claim 1.
3. A closed-loop control precision machining system based on in-situ detection as claimed in claim 2, wherein the industrial camera is provided with an optical lens and an annular light source; the annular light source consists of a plurality of illuminating lamps which are annularly arranged; all the illuminating lamps are uniformly distributed around the optical lens in a ring shape.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910646784.8A CN110376963B (en) | 2019-07-17 | 2019-07-17 | Closed-loop control precision machining method and system based on in-situ detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910646784.8A CN110376963B (en) | 2019-07-17 | 2019-07-17 | Closed-loop control precision machining method and system based on in-situ detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110376963A CN110376963A (en) | 2019-10-25 |
| CN110376963B true CN110376963B (en) | 2020-12-08 |
Family
ID=68253863
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910646784.8A Active CN110376963B (en) | 2019-07-17 | 2019-07-17 | Closed-loop control precision machining method and system based on in-situ detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110376963B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110989494A (en) * | 2019-11-29 | 2020-04-10 | 上海交通大学 | Thin-wall part machining error measuring and compensating method based on trigger type measuring head |
| CN111338289A (en) * | 2020-03-02 | 2020-06-26 | 珠海格力智能装备有限公司 | Machine tool precision analysis method and device, precision detector and machine tool machining method |
| CN112059726B (en) * | 2020-08-31 | 2022-05-24 | 西安精雕精密机械工程有限公司 | On-machine measurement and compensation processing method of workpiece with multiple reference characteristics |
| CN112198838B (en) * | 2020-10-12 | 2021-08-10 | 湖南汽车工程职业学院 | Intelligent detection system for working condition parameters of machine tool |
| CN112255964B (en) * | 2020-10-12 | 2021-08-13 | 湖南汽车工程职业学院 | Machine tool machining cutter remote temperature control system based on big data |
| CN112859742B (en) * | 2021-01-12 | 2021-08-06 | 广州盛硕科技有限公司 | Numerical control machine tool control system based on visual identification |
| CN113093648A (en) * | 2021-03-31 | 2021-07-09 | 西门子（中国）有限公司 | Automatic machining system and method for motor base |
| CN113304971B (en) * | 2021-04-26 | 2022-12-13 | 深圳市世宗自动化设备有限公司 | 3D dynamic guiding dispensing compensation method, device and equipment and storage medium thereof |
| CN115933534B (en) * | 2023-02-09 | 2023-11-07 | 山东山科世鑫科技有限公司 | Numerical control intelligent detection system and method based on Internet of things |
| CN117260379B (en) * | 2023-11-21 | 2024-02-23 | 靖江市恒友汽车部件制造有限公司 | On-line control method for machining diameter of automobile part |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102773524A (en) * | 2012-08-01 | 2012-11-14 | 西北工业大学 | Hole making method with alignment based on machine vision |
| CN104933220B (en) * | 2015-05-11 | 2018-10-09 | 东莞市凌英模具塑胶有限公司 | The high-accuracy manufacturing method of complex-curved automobile injection mold and injection mold |
| CN106584209A (en) * | 2016-11-10 | 2017-04-26 | 哈尔滨理工大学 | Real-time online monitoring method for tool wear based on cloud manufacturing |
| CN107132814B (en) * | 2017-02-24 | 2019-12-13 | 上海交通大学 | open numerical control system based on machine vision state monitoring and error compensation |
-
2019
- 2019-07-17 CN CN201910646784.8A patent/CN110376963B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110376963A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110376963B (en) | Closed-loop control precision machining method and system based on in-situ detection | |
| CN110434671B (en) | Cast member surface machining track calibration method based on characteristic measurement | |
| CN109489580B (en) | On-machine point cloud detection and compensation method for surface processing of aero-engine blade | |
| CN112964186B (en) | Device and method for measuring clearance in automatic shaft hole assembly process | |
| CN108127206B (en) | Laser brazing process transplanting method and laser brazing device capable of transplanting data | |
| CN101199994A (en) | Intelligent laser cladding forming metal parts | |
| WO2023138241A1 (en) | Machine tool rapid compensation system based on principle of laser interference, and compensation method thereof | |
| CN109341532A (en) | A kind of part coordinate scaling method based on structure feature towards automatic assembling | |
| US11809155B2 (en) | Morphic manufacturing | |
| Feng et al. | A novel feature-guided trajectory generation method based on point cloud for robotic grinding of freeform welds | |
| CN108032011A (en) | Initial point guiding device and method are stitched based on laser structure flush weld | |
| CN107932185A (en) | The numerically-controlled machine tool closed loop system of view-based access control model | |
| CN109759897A (en) | A kind of measurement of the horizontal assembly of large size shell and aligning method | |
| CN111069973B (en) | Method and device for quickly aligning complex-shape casting | |
| RU2528923C2 (en) | Article adaptive machining at nc machines | |
| CN113137627B (en) | Machining and positioning method for cooling air film hole of flame tube of aircraft engine | |
| CN107544428A (en) | A kind of closed-loop Digit Control Machine Tool processing method of view-based access control model | |
| CN207171229U (en) | Automatic laser welder with CCD detections | |
| CN108170095A (en) | The processing method of the numerically-controlled machine tool closed loop system of view-based access control model | |
| US20220134567A1 (en) | Robot system, robot control device, and robot control program | |
| CN107991990B (en) | Vision-based full-closed loop system of numerical control machine tool | |
| CN118314138B (en) | Laser processing method and system based on machine vision | |
| Dong et al. | A Vision Algorithm for Robot Seam Tracking Based on Laser Ranging | |
| CN111210469B (en) | Digital image correlation technology-based robot machining displacement parameter measurement method | |
| TWI762182B (en) | Automatic processing method and automatic processing system |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220104 Address after: 430000 No. 07 (b452), floor 3, building 3, phase III, international enterprise center, No. 1, Guanggu Avenue, Donghu New Technology Development Zone, Wuhan, Hubei (Wuhan area of free trade zone) Patentee after: Wuhan moding Technology Co.,Ltd. Address before: 430074 Hubei Province, Wuhan city Hongshan District Luoyu Road No. 1037 Patentee before: HUAZHONG University OF SCIENCE AND TECHNOLOGY |
|
| TR01 | Transfer of patent right |