JPH04135207A - Numerical controller with automatic producing function of machining condition - Google Patents

Abstract

PURPOSE:To automatically produce the optimum machining condition data from various actual input condition data by obtaining the machining condition data after correcting the reference value with the corrected variable set to the reference value of the machining condition data. CONSTITUTION:The reference value of the machining condition data by a reference value computing means based on the input condition data. At the same time, the input condition data is inputted and a corrected variable set to the reference value of the machining condition data is outputted through a 1st neutral network. Then the corrected variable is added to the reference value of the machining condition and the proper machining condition data is calculated by a 1st corrected variable computing means. Based on the machining condition data, a work is actually machined and the machining error data is obtained to the request value for machining the work. This error data is stored in a storage means and then inputted to a 2nd neural network. Then the corrected variable is calculated to the machining condition data. A 2nd correction means adds the corrected variable to the machining condition data. As a result, a proper corrected variable of the machining condition data is automatically decided according to various machining condition data.

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a numerical control device having a function of automatically creating processing condition data.

[Prior art]

Conventionally, for example, in grinding using a numerically controlled grinder, input condition data such as finished diameter of the workpiece, machining allowance, rigidity coefficient, part type, etc. are used to determine the workpiece for each grinding mode such as rough grinding, fine grinding, equipment, etc. There is known an apparatus that uses a computer to automatically determine machining conditions such as the number of revolutions, the grinding start position, the grinding feed rate, and the post-grinding feed stop time.

[Problem to be solved by the invention]

However, in reality, it is known from experience that automatically determined processing condition data is not optimal data for various input condition data. In such cases, when the operator looks at the automatically determined machining condition data and finds that it is not appropriate, the operator may correct the value based on the operator's experience or intuition, or perform the actual machining using the machining condition data. The operator judged the resulting machining results and corrected the machining condition data based on experience and intuition. How to correct which machining condition data from input condition data, or how to correct which machining condition data from the machining error obtained from the machining result when machining a workpiece with certain machining condition data. The correct correction depends on the experience and intuition of the operator, and the correct correction is not always performed. For this reason, it was necessary to machine the workpiece using the corrected machining condition data and to correct the machining condition data again according to the machining results. Furthermore, even if the operator corrects the machining condition data correctly, there is a problem in that the causal relationship between the machining condition data at that time and the input condition data or machining error data is not reflected in the subsequent correction of the machining condition data. Ta. Therefore, there is a problem that it takes time to create appropriate processing condition data. The present invention has been made to solve the above problems, and its purpose is to automatically create optimal processing condition data from various actual input condition data.

[Means to solve the problem]

As shown in FIG. 12, the configuration of the invention for solving the above problems includes an input condition data storage means for storing input condition data, and a reference value for calculating a reference value of machining condition data based on the input condition data. a first neural network that receives input condition data as input and outputs a correction amount for the reference value of each processing condition data; a first correction means for calculating the machining error data; an error data storage means for storing the machining result data obtained when machining the workpiece according to the machining condition data and machining error data for the required value; The second neural network outputs a correction amount for the data, and a correction means further corrects the machining condition data using the correction amount outputted from the second neural network.

[Effect]

In the present invention, the reference value calculation means determines the reference value of the machining condition data from the input condition data using a theoretical formula or the like. Further, the first neural network inputs the input condition data and outputs a correction amount for the reference value of the machining condition data. Then, by the first correction amount calculation means,
The correction amount is added to the reference value of the machining conditions to calculate appropriate machining condition data. Furthermore, the workpiece is actually machined using the machining condition data, and machining result data such as surface accuracy, dimensional accuracy, and machining time are measured. Then, the deviation of the machining result data from the required value, that is, the machining error data is determined,
It is stored in the storage means. Next, the machining error data is input to the second neural network, and a correction amount for the machining condition data is calculated. Thereafter, the second correction means further adds the correction amount to the machining condition data to obtain final appropriate machining condition data. The coupling coefficients of the first neural network are learned by inputting various input condition data and using a correction amount of processing condition data optimal for the input condition data as a teacher signal. Furthermore, when the output of the first neural network is not appropriate, the coupling coefficient is learned using the appropriate output at that time as a teacher signal. This appropriate output can be obtained by performing learning that takes into account the amount of correction by the output of the second neural network, thereby making it possible to obtain machining condition data that makes the machining error data zero using only the first neural network. Furthermore, the coupling coefficient of the second neural network is:
The machining error data obtained as a result of machining with various machining condition data is input, and an appropriate correction amount of the machining condition data corresponding to the machining error data is learned as a teacher signal. Furthermore, when the output of the second neural network is not appropriate, the coupling coefficient is learned using the appropriate output at that time as a teacher signal. In this way, the operator's experience and intuition are memorized in the form of the coupling coefficients of the two neural networks, so the appropriate amount of correction for machining condition data corresponding to various input condition data is automatically determined. .

【Example】

The present invention will be described below based on specific examples. (1) Structure of Grinding Machine FIG. 1 is a block diagram showing the entire mechanical structure of a numerically controlled grinding machine having an automatic processing condition determination device according to the present invention. 50 is a grinding machine, and a table 52 that slides on the bed 51 of the grinding machine 5o is provided above the bed 51. The table 52 is moved in the left-right direction in the drawing by driving the table feed motor 53. Also, on the table 52 are a headstock 54 and a psychological table 56.
are arranged, the headstock 54 has a main shaft 55, and the psychological table 56 has a mind axis 57. The workpiece W is pivotally supported by a main shaft 55 and a Shinshin shaft 57,
It is rotated by the rotation of the main shaft 55. This rotation of the main shaft 55 is performed by a main shaft motor 59 disposed on the head stock 54. On the other hand, the grinding wheel 6o for grinding the workpiece W is fixed to a drive shaft that is rotationally driven by a grinding wheel drive motor 62 provided on a grinding wheel head 61. Further, the grindstone head 61 is controlled to move in the vertical direction of the drawing by a grindstone head feed motor 63. Table feed motor 53, grindstone feed motor 63,
A numerical control device 3o is provided to drive and control the main shaft motor 59, grinding wheel drive motor 62, and the like. (2) Configuration of numerical control device The numerical control device 30 mainly consists of
It is composed of a CPU 31, a ROM 32, a RAM 33, and an IP (interface) 34. The RAM 33 has an NC data area 331 for storing NC programs, an input condition data area 332 for storing input condition data such as finishing diameter, machining allowance, rigidity coefficient, part type, division on fabric, operator name, etc., and workpiece rotation speed. , a machining condition data area 333 that stores machining condition data such as a grinding start position, a grinding feed rate, and a feed stop time after grinding, a neural network area 334 that stores a neural network calculation program, and a connection that stores neural network coupling coefficients. A coefficient area 335 is provided. In addition, the RAM 33 includes a required value area 336 that stores required values of machining results determined from required machining specifications, a reference value area 337 that stores automatically determined standard values of machining condition data, and calculations using a neural network. A correction amount area 338 for storing the correction amount of the processed machining condition data, a machining result data area 339 for storing machining result data, and a machining error data area 340 for storing machining error data are formed. Note that the RAM 33 is backed up by a battery, and the learned coupling coefficients are stored therein. Further, the ROM 32 includes a control program area 321 storing a control program for operating the numerically controlled grinding machine according to NC data, an automatic determination program area 322 storing a main automatic determination program for determining machining condition data, and a neural network. A learning program area 323 storing a program for learning coupling coefficients is formed. In addition, the numerical control device 30 is connected to the operation panel 20 via the IF 54.
is installed. Operation panel 2 of the operation panel 20
1, there are provided a keyboard 22 for inputting data and a CRT display device 23 for displaying data. Further, an FD drive device 36 is connected to the numerical control device 30 via an IF 55, and an FD (flexible disk) 37, which is a storage medium, is read and written via the FD drive device 36. (3) Operation Next, the processing procedure of the CPU 31 used in the device of this embodiment will be explained based on a flowchart. 1. General operation of the entire apparatus The general operation of the apparatus will be explained with reference to FIG. 11. In step S1, a first neural network calculation program used to actually obtain machining condition data is created from the FD37a that stores neural network calculation programs corresponding to many types of machining and their learned coupling coefficient data. and its coupling coefficient data are stored in RAM3.
It can be loaded into 3. Hereinafter, the calculation program of the first neural network will also be referred to as a condition matching engine. In step S2, input condition data is input to the RAM 33. In step S3, the reference value of the machining condition data is automatically determined from the input condition data. In step S4, a correction amount of the machining condition data is calculated by a calculation program (condition adaptation engine) of the first neural network that receives the input condition data. In step S5, the processing condition data corrected by the corrected quantum reference value is displayed on the CRT 23. In step S6, the machining condition data is judged based on the operator's experience and intuition to determine whether it is appropriate. If it is not appropriate, the processing condition data is corrected in step S7. Further, in step S8, the coupling coefficient of the first neural network is learned using the corrected correction amount as a teacher signal. By repeatedly performing learning and calculation of the machining condition data, the machining condition data is determined to be appropriate in step S6. Incidentally, since the first neural network is supplied in a trained form, in most cases, appropriate machining condition data can be obtained by calculating -degrees. In step S9, the workpiece is actually machined using the machining condition data. In step SIO, machining results such as finished dimensions of the workpiece are measured, and machining error data indicating an error from the required value is calculated. In step Su, the machining error data is evaluated, and if appropriate, machining is completed in step S12. Furthermore, if the machining results are not appropriate, it is necessary to correct the machining condition data. In step S13, the above machining error data is stored in RAM3.
3 is input. In step S14, the second neural network calculation program corresponding to the processing type and its learned coupling coefficient are input to the RAM 33 from the FD 37b. The calculation program of the second neural network is also called a machining diagnosis engine. In step S15, a second
The neural network is activated to calculate the correction amount of the machining condition data. In step S16, new machining condition data is calculated using concave machining condition data + correction amount, and the influence of the new machining condition data on the machining result is predicted. Step S1? In this step, the modified machining condition data is evaluated based on the predicted machining results. If the processing condition data is not appropriate, step S18
Then, the coupling coefficient of the second neural network is learned using the appropriate correction amount as a teacher signal. By repeating learning and calculation of machining condition data, step S1? Therefore, the machining condition data is determined to be appropriate. Incidentally, since the second neural network is supplied in a trained form, in most cases, appropriate machining condition data can be obtained by calculating -degrees. Then, the process returns to step S6, and if the correction amount by the second neural network is large in the corrected machining condition data, in step S8, the deviation of the machining condition data from the reference value is used as a teacher signal, and the first The coupling coefficients of the neural network are also learned. Therefore, as learning progresses, the correction amount by the second neural network is gradually added to the coupling coefficient of the first neural network, so that when learning progresses, only the first neural network
It becomes possible to obtain optimal processing condition data. That is, the machining error data gradually becomes smaller, and the amount of correction by the second neural network also gradually decreases. In addition, the coupling coefficient data of the first and second neural networks learned as described above during actual processing is F.
The data is output to D37a, 37b, and the corresponding coupling coefficient data stored in FD37a, 37b is updated. As described above, in this embodiment, the machining condition data is corrected by the condition matching engine and the machining diagnosis engine. FIG. 3 shows a flowchart of a program that inputs a predetermined neural network calculation program and coupling coefficient data from the FD 37 of the storage medium and stores them in the RAM 33. The FD 37 stores a large number of first and second neural network calculation programs and learned coupling coefficients corresponding to the type of processing (for example, grinding of an engine camshaft, etc.). In step 700, a list of all neural network names is displayed on the CRT 23. In step 702, the operator looks at the list and selects the first and second neural networks (condition matching engine and processing diagnosis engine) corresponding to the type of processing to be performed. Next, in step 704, the calculation programs of the two selected neural networks and their coupling coefficient data are transferred from the FD 37 to the neural network area 334 and the coupling coefficient area 335 of the RAM 33.
is stored in a predetermined area. 3. Input of input condition data FIG. 4 is a flowchart of the main program that automatically generates machining condition data from input condition data. In step 100, input condition data input from the keyboard 22 is read and stored in the input condition data area 332 of the RAM 33. In this embodiment, the input condition data includes finished diameter data D1. Removal allowance data 02. Rigidity coefficient data D31 Part type data D, Rough finishing division data Ds, A worker name data D6. B is worker name data D, and C is worker name data D. In the next step 102, the above input condition data (D to D
, ) and other input condition data, the reference values v1 to V of the machining condition data are calculated. In this embodiment, the machining condition data includes coarse grinding rotation speed data, fine grinding rotation speed data of 2+, and machine string rotation speed data of 5. Rough grinding start diameter data 1. 7. Fine grinding start diameter data Ks, equipment start diameter data Ks, coarse grinding feed speed data. Seiken feed speed data Km, equipment feed speed data 9. 1° for feed stop time data after rough grinding, K for feed stop time data after equipment
It is composed of l+. The calculation of this reference value is performed as follows. Regarding the rotation speed data for each grinding mode, the circumferential speed of the workpiece is determined in advance as a function of the required surface roughness of the workpiece with respect to the circumferential speed of the grinding wheel during a certain grinding. The circumferential speed of the workpiece is calculated from the commanded surface roughness of the workpiece, and the rotation speed data of the workpiece is calculated from the circumferential speed of the workpiece and the diameter of the workpiece. Regarding the feed rate data for each grinding mode, the depth of cut per rotation of the workpiece of the grinding wheel is determined in advance as a function of dimensional tolerance. The depth of cut is calculated from the commanded dimensional tolerance of each grinding mode, and the grinding feed rate data is calculated from the rotation speed data. The grinding start diameter data for each grinding mode has a standard feed amount set for each grinding mode, and is calculated based on the relationship between the commanded finishing diameter and this feed amount. Regarding the post-grinding feed stop time data, the rotational speed of the workpiece at which the feed is stopped is determined depending on whether fixed-size grinding is performed and whether grinding is performed in divided steps. The post-grinding feed stop rotation speed is determined from the input condition data, and the post-grind feed stop time is calculated using the rotation speed data. Next, in step 104 of FIG. 4, the first neural network is activated to obtain input condition data D, ~D.
, and each correction amount δ1 to δ,1 of each processing condition data (L~) is calculated. The first neural network has the configuration shown in FIG. In this example, the neural network is
It has a three-layer structure: an input layer, an intermediate layer, and an output layer. As is well known in the neural network, each element of the second intermediate layer and the third output layer is defined as an element that performs the following calculation. The output OJ of the j-th element of the first layer is calculated by the following equation. However, 1≧2. 0, = f(1,) f (x)-1/ (1+exp (-x))
-(3) However, V
, is 7Nl of the third arithmetic element in the first layer.
The coupling coefficient between the th elements, O4, represents the output value of the third element in the first layer. In other words, since it is the first layer, it outputs the input as it is without performing any calculations, so the input layer (the first
It is also the input value of the jth element of the layer). Therefore, J=04 (4)
However, D is input condition data input to the third element of the input layer. The specific calculations of the neural network are executed according to the procedure shown in FIG. In step 200, each element of the intermediate layer (second layer) receives an output value II, -D from each element of the input layer (first layer).
, the product-sum operation of the following equation is performed. Next, in step 206, the output value of each element in the output layer is calculated using a sigmoid function, similar to equation (6). This output value becomes the correction amount δ of the machining condition data. That is, the correction amount δ is obtained by the following equation. The third element of the second layer is calculated using the following equation. In this embodiment, the bias is zero. Next, in step 202, the output of each element of the intermediate layer (second layer) is calculated using the sigmoid function of the product-sum function value of the input values of equation (5) using the following equation. jth layer of second layer
The output value of the th element is calculated using the following equation. This output value Oj becomes the input value of each element of the output layer (third layer). Next, in step 204, the output layer (third layer) returns to step 106 in FIG.
By the sum of the correction amounts δ1 to δII determined in , 1 to 1 are determined for the machining condition data. The machining condition data 1 to K11 are stored in the machining condition data area 333 of the RAM 33. Next, in step]08, the machining condition data is changed to ~Kl.
l is displayed on the CRT 23, and the operator looks at the displayed result and, if necessary, inputs a correction value from the keyboard 22.
The value is stored in the machining condition data area 333. Next, in step 110, in the processing condition pig,
-, the NC data is calculated using , and the calculated N
The C data is stored in the NC data area 331 of the RAM 33. 7 Generation of machining error data Next, a workpiece is trial-machined according to this NC data. The machining results of the workpiece are measured and machining result data is obtained. Machining result data includes, for example, total grinding time, finished surface roughness, and finished dimensional accuracy. Finish roundness, degree of cracking, degree of chatter, etc. Next, in step 500 of FIG. 7, which describes a procedure for modifying machining condition data from these measured machining result data, the above-mentioned measured machining result data is inputted from the keyboard 22 by the operator, and the input The value is stored in the processing result data area 339 of the RAM 33. Next, in step 502, the required value of the machining result is input in the same way, and the value is stored in the required value area 336 of the RAM 33.
is memorized. Next, in step 504, machining error data, which is the deviation of the machining result data from the required value, is calculated, and these data are stored in the machining error data area 340 of the RAM 33. The machining error data includes a total grinding time error H1, a finishing roughness error H1, a finishing dimensional error H1, a finishing roundness error, a cracking degree error 1'ls, a chatter degree error H6, etc. 8. Second Neural Network (Machining Diagnosis Engine) Next, in step 506, the machining error data ()1. ~1
(6) is input to start the second neural network. Then, in step 508, the correction amount (61 to 61) of the processing condition data that is the output of the second neural network
δ11) is stored in the correction amount area 338 of the RAM 33. This second neural network has the configuration shown in FIG. In this embodiment, the second neural network has a three-layer structure including an input layer, a middle layer, and an output layer. The arithmetic function of each element of this second neural network is completely the same as the arithmetic function of the first neural network described above. Next, in step 510, each concave machining condition data (
The correction amount (δ1 to δ1.) is added to K, ~kz) to obtain corrected new machining condition data (K, ~kz). These data are stored by rewriting the contents of the machining condition data area 333 of the RAM 33. As described above, in this embodiment, the reference value of the machining condition data is corrected by the first neural network, and further, the second neural network is used to perform correction taking into account machining errors. Therefore, more accurate machining condition data is required. Furthermore, by performing learning of the first neural network even when the amount of correction by the second neural network is large, as the learning progresses, the first neural network alone can obtain suitable processing condition data. The amount of correction can be determined. That is, as the learning of the first neural network progresses, the amount of correction by the second neural network decreases, making it possible to obtain machining condition data without the need for trial machining or the like. 9. Learning of the first neural network Next, we will explain how to learn the coupling coefficients of the first neural network. The coupling coefficients are determined using the well-known back propagation method for the neural network shown in FIG. This learning is performed when it becomes necessary to further learn the already learned coupling coefficients at the time of use. That is,
When the correction amount output by the first neural network is not appropriate, and when the processing error becomes large when processing using the processing condition data corrected by the first neural network, in other words, when the correction amount output by the first neural network is This is a case where the amount of correction to be output is large. At step 600 in FIG. 9, a learning signal for each element of the output layer is calculated using the following equation. However, T is a teacher signal for the correction amount δ, which is the output, and is given from the outside. Further, f' (x) is a derivative of the sigmoid function. Next, in step 602, the learning signal of the intermediate layer is calculated using the following equation. Next, in step 604, the coupling coefficients of each element of the output layer are corrected. The amount of correction is determined by the following formula. fit of the coupling coefficient between the child and the first element of the intermediate layer
This is the amount of change in the second calculation. This is the amount of correction. P and Q are proportionality constants. Therefore, the corrected coupling coefficient can be obtained as follows. Next, the process moves to step 606, and the coupling coefficients of the elements in the intermediate layer are corrected. The amount of correction of the coupling coefficient is determined by the following equation, as in the case of the output layer. Therefore, the corrected coupling coefficient can be obtained as follows. Next, in step 608, it is determined whether the correction amount of the coupling coefficient has become less than or equal to a predetermined value, and it is determined whether the coupling coefficient has converged. If the coupling coefficient has not converged, the process returns to step 600, similar calculations are repeated using the newly corrected coupling coefficient, and the coupling coefficient is again corrected. By repeating such calculations, learning for a certain teacher signal is completed. Furthermore, since the coupling coefficient has changed for other input condition data, learning for other input condition data is also performed. 10. Learning of the neural network of WE 2 Regarding the second neural network that receives machining error data as input, if the correction amount of the output machining condition data is not appropriate, using the teacher signal for the correction amount, Learning takes place as needed. 11. Saving the learned coupling coefficient data As shown in step 800 in FIG.
The coupling coefficient data of the first and second knee coral networks stored in the FD 35 is rewritten in the area where the coupling coefficient data was stored, and the latest learned coupling coefficient data is saved. . As a result, when performing the same type of machining next time, accurate machining condition data will be generated quickly, and the neural network will be more convenient to use. In the above embodiment, the neural network calculation program and its coupling coefficient data necessary for processing are stored on the FD37.
Although the information is input from the main computer via a communication line, the information may also be input from the main computer to the numerical control device. That is, it can also be applied to machining centers and numerical control devices connected to a parent computer via a communication line.

【Effect of the invention】

The present invention uses a first neural network that inputs input condition data and outputs a correction amount for each processing condition data to calculate the correction amount for the reference value of the processing condition data, and corrects the reference value with the correction amount. By doing this, processing condition data can be obtained. Further, a second neural network inputs machining error data obtained when actually machining a workpiece using the machining condition data, and calculates a correction amount for the machining condition data.
The processing condition data is further corrected using the correction amount. Therefore, since the operator's experience and intuition can be learned and stored in the form of coupling coefficients in the two neural networks, appropriate machining condition data that reflects the machining results can be automatically generated from the input condition data.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram showing the overall mechanical configuration of a numerically controlled grinding machine having an automatic processing condition determination device according to a specific embodiment of the present invention. FIG. 2 is a block diagram showing the electrical configuration of the numerical control device and operation panel according to the same embodiment. FIG. 3 is a flowchart showing a procedure for inputting a neural network calculation program and its coupling coefficient data by the CPU used in the device of the embodiment. FIG. 4 is a flowchart showing the procedure for calculating machining condition data by the CPU used in the apparatus of the embodiment. Fifth
The figure is a structural diagram showing the structure of the first neural network according to the same embodiment. FIG. 6 is a flowchart showing the calculation procedure by the first neural network. 7th
The figure is a flowchart showing a procedure for correcting machining condition data based on machining error data by the CPU used in the apparatus of the embodiment. FIG. 8 is a structural diagram showing the structure of the second neural network. FIG. 9 is a flowchart showing the learning procedure of the first neural network. FIG. 10 is a flowchart showing a procedure for saving coupling coefficient data learned by the CPU used in the device of the embodiment in a storage medium. FIG. 11 is a flowchart showing the overall operational concept of this device. FIG. 12 is a block diagram showing the concept of the present invention. 20 Operation panel 21 Operation panel 22 Keyboard 23-CRT display device 30 Numerical control device 31-CP TJ32-ROM 33-RA
M 50 Grinding machine 51 Bed 52 Table 53 Table feed motor 54 Headstock 55 Main spindle 56 Tailstock 57 Tailstock 59 Main spindle motor 60
Grinding wheel 61 Grinding wheel head 62 Grinding wheel drive motor 63 Grinding wheel head feed motor W Workpiece diagram diagram diagram diagram diagram diagram diagram

Claims (1)

[Claims] When machining a workpiece with a numerically controlled machine tool, machining condition data such as the workpiece rotation speed, machining start position, post-machining feed stop time, and tool feed rate in each machining mode are A device that automatically determines a finished diameter of a workpiece, machining allowance, rigidity coefficient, tool type, etc. from input condition data, comprising: an input condition data storage means for storing the input condition data; and a reference value of the machining condition data to the input condition data. a first neural network that receives the input condition data as an input and outputs a correction amount for the reference value of each of the processing condition data; first correction means for calculating the machining condition data by correcting the reference value of the machining condition data based on the amount; and storing machining result data obtained when machining the workpiece according to the machining condition data and machining error data for the required value. a second neural network that inputs the machining error data and outputs a correction amount for the machining condition data; and a second neural network that inputs the machining error data and outputs a correction amount for the machining condition data; A processing condition automatic correction device comprising a correction means for further correction.