JP2924426B2 - Feedback processing condition correction device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for correcting machining conditions by feeding back dimensional information of a workpiece during or after machining, and more particularly to a technique for improving the accuracy of the compensation.

[0002]

2. Description of the Related Art For example, the actual size such as the actual diameter of a machined portion such as an inner cylindrical surface such as a cylinder bore of a vehicle engine or an outer cylindrical surface such as a journal surface of an engine crankshaft accurately matches a target size. In order to perform processing, in-process control, post-process control, hybrid control, and the like are used.

[0003] In-process control is a control for using an in-process measuring tool for measuring the size of a processing portion during processing, and terminating one processing when the measured size thereby reaches a target size. In the in-process control, it is generally assumed that the actual dimension reaches the target dimension when the measured dimension reaches a criterion value (for example, a fixed size point), and a comparison between the measured dimension and the criterion value is performed. The determination as to whether the actual dimensions have reached the target dimensions is made indirectly.

[0004] The post-process control uses a post-process measuring tool for measuring the dimension of a processed portion after processing, and feeds back a dimensional error, which is a difference between the measured dimension and a target dimension, to be used next by the processing tool. This is control for correcting the processing data (for example, NC data). Note that the “post-process measuring tool” here is not always used to measure the dimensions of the processed portion immediately after the end of the processing, and the processed portion processed earlier (the portion processed before In some cases, it is used in a mode of measuring the dimensions of the (worked part).

The hybrid control is a control that combines the in-process control and the post-process control. Generally, the hybrid control is a control for substantially correcting a target dimension in the in-process control by feeding back a dimension error in the post-process control, that is, a dimension error after processing based on a dimension measured by a post-process measuring tool. Here, "substantially correcting the target dimension" means not only directly correcting the target dimension, but also indirectly correcting the target dimension, for example, by correcting the determination reference value. I do.

[0006]

Among these controls, the post-process control and the hybrid control are both controls for correcting the processing conditions of a processing tool by feeding back a dimensional error of a processed portion after processing. However, since both the conventional post-process control and the conventional hybrid control correct the processing data, the determination reference value, and the like as the processing conditions based only on the dimensional error, there is a limit in increasing the correction accuracy. There is a problem. Although the correction of the processing conditions should originally be performed taking into account the factors that affect the actual dimensions of the processing part as much as possible, the conventional post-process control and hybrid control both take only dimensional errors into account. This is because the processing conditions are corrected without performing.

In short, the conventional feedback-type machining condition correction apparatus that corrects machining conditions by feeding back a dimensional error of a machined portion after machining does not sufficiently consider factors that affect the actual dimensions of the machined portion. Therefore, there is a problem that the correction accuracy cannot be sufficiently increased because the processing conditions are corrected.

The conventional feedback-type processing condition correction apparatus has the following problem. As described above, the post-process measuring tool is not always used in a mode for measuring the dimensions of a processed portion immediately after the end of processing, but is used in a mode for measuring the dimensions of a processed portion processed earlier. It may be done. In some cases, it is used in a mode of measuring the dimensions of a machined part that has been machined many times before the current machining to be performed (hereinafter simply referred to as an old machined part). However, in the conventional feedback-type machining condition correction device used in this case, the more the dimension to be fed back relates to the older machining portion,
In other words, there is a problem that the more distant the past dimensional information is, the lower the correction accuracy of the processed part is. The present invention has been made to solve these problems.

[0009]

SUMMARY OF THE INVENTION In order to solve this problem, the gist of the present invention is to control a processing tool for sequentially processing at least one processing portion set on each of a plurality of workpieces according to processing conditions. As shown in FIG. 1, a feedback-type processing condition correction device connected to the processing tool control means performs (a) a dimensional error of each processing portion and a change tendency thereof during and / or after processing. and dimensional information acquisition unit 1 to acquire the size information respectively, a processing condition correction means 2 for successively determining a correction value (b) processing conditions, and the obtained size error and dimensional error change tendency, a previously determined preceding And determining the current correction value based on the at least one correction value obtained.

The feedback-type machining condition correction apparatus according to the present invention (hereinafter, simply referred to as the apparatus of the present invention) has, for example, only one processing portion set for one work, and each of a plurality of works is not required. In the case of processing with the same processing tool in order, or when a plurality of processing parts are set in one work, each of the plurality of works is in order, and in each work, each of the plurality of processing parts is in the same order. In the case of machining with a tool or a plurality of machining parts are set in one work, each of the plurality of works is sequentially and, in each work, the plurality of machining parts are respectively different processing tools. The present invention can also be applied to the case where processing is performed. In this case, it is possible to implement the present invention by actually determining the correction values of the processing conditions individually for each of the plurality of processing parts. If the size and the size match each other, the correction value of the processing condition is actually determined for at least one of the processing portions, but the determined correction value is diverted for the other processing portions. The present invention can be implemented as omitting the actual determination of the correction value.

The "work" in the present invention includes, for example, a cylindrical surface having a straight surface extending without any change in diameter and serving as a processing portion, or a cylindrical surface extending straight with a change in diameter. Can be selected as a part to be processed. In the former case, since the diameter of each cylindrical surface is usually measured as one dimension, that one dimension is the "dimension" in the present invention. On the other hand, in the latter case, since the dimension is usually measured at each axial position of the cylindrical surface, the dimension at each axial position is the “dimension” in the present invention.

In the present invention, the "correction amount of the processing condition" means the amount by which the processing state of the processing tool is corrected as a result. It also means the amount by which a parameter related to the processing condition is corrected in order to indirectly correct the condition.

The present invention can be applied to post-process control. That is, for example, the dimension information acquiring means 1 acquires the difference between the dimension measured by the post-process measuring tool and the target dimension as a dimension error after processing, and further changes the error (change between a plurality of different processing parts different from each other). ), And the processing condition correction means 2 performs processing as processing conditions based on the dimensional error after the processing and the dimensional error change tendency and at least one correction value before the previous processing. By determining the current correction value of data (for example, NC data), the present invention can be applied to post-process control.

The present invention can also be applied to hybrid control. That is, for example, the dimension information acquiring means 1 acquires the difference between the dimension measured by the post-process measuring tool and the target dimension as a dimension error after processing, and further changes the error (change between a plurality of different processing parts different from each other). ), And the processing condition correction means 2 is used as a processing condition based on the dimensional error after the processing and the dimensional error change tendency and at least one correction value before the previous processing. The present invention can also be applied to hybrid control by determining the current correction value of the reference value to be compared with the dimension measured by the in-process measuring tool.

The present invention can be applied to in-process control. That is, for example, the dimension information acquiring means 1 sequentially measures the dimensions of each processing part (one dimension representing the entire processing part) by the in-process measuring tool (as the processing proceeds for each processing part). Every time the difference between the measured dimension and the target dimension is obtained as a dimensional error during processing, and a change tendency thereof (a change tendency with the passage of time in the same processing part) is also obtained; The condition correcting means 2 determines a current correction value of a processing condition (for example, a feed speed of the processing tool) of the processing tool based on the dimensional error and the dimensional error change tendency and at least one correction value before the previous time. By doing so, the present invention can also be applied to in-process control.

Further, the dimension information acquiring means 1 is operated after the machining of the in-process measuring tool to measure the dimension, and the difference between the measured dimension and the target dimension is acquired as a dimension error after machining. (Change tendency between a plurality of different machined parts) is obtained, and the machining condition correcting means 2 is used to determine the dimensional error after the machining and the dimensional error change tendency and at least one The present invention is applied to in-process control by determining a current correction value of a criterion value to be compared with a dimension measured by an in-process measuring tool as a processing condition based on the correction value of You can also. In other words, this application example is an example in which the processing conditions for another processing part are corrected based on the dimensional information about a certain processing part, whereas the previous application example is for processing a certain processing part. it is the example of correcting their working conditions based on its size information medium top that is being performed.

In each of the examples of application to the post-process control, the hybrid control, and the in-process control, various types can be adopted as the correction value determination format of the processing condition correction means 2. For example, a candidate correction value which is a candidate value of the current true correction value of the processing condition is determined from the dimensional error and the tendency of the dimensional error change, and the current candidate correction value is determined as at least one candidate correction value or a genuine value of the previous correction. The correction based on the correction value may be used to determine the genuine correction value, or the genuine correction may be performed based on the dimensional error, the dimensional error change tendency, and at least one candidate correction value or genuine correction value before the previous time. It is possible to adopt a format for determining a value. In another embodiment of the present invention
Indicates that the processing condition correction means determines that the acquired dimensional error and
Determine the machining conditions based on the method error change tendency
The candidate correction values that are the candidate values of the genuine correction values to be determined are sequentially determined.
And the determined candidate correction value and the previous
Based on the determined at least one candidate correction value,
The regression line of the candidate correction values for their determination order
Is determined, and the determined candidate correction value of the current time is determined.
Corrected to a value that satisfies the regression line
The positive correction value is determined. Yet another embodiment
In the above, the dimensional information acquiring means may
Machining order for dimensions for multiple machined parts after finishing
Measurement in the same order as
To obtain dimensional errors in order, and
Of the dimensional error
As the slope when changing with respect to the acquisition order
It is assumed.

[0018]

In the apparatus of the present invention, the dimensional information obtaining means 1 obtains the dimensional error of each processed part and its change tendency as dimensional information during and / or after the processing, and corrects the processing condition. by means 2, a dimension error and dimensional error change trend obtained, the correction value of the current is determined based on at least one correction value of the previous previously.

As described above, in the apparatus of the present invention, the processing conditions are corrected based not only on the dimensional error but also on the change tendency thereof. In this case, the correction is made more appropriately than before.

Generally, a correction value for realizing a truly suitable processing condition generally changes while exhibiting certain characteristics as the processing proceeds. For example, one of the manifestations is that the correction value does not change so rapidly as the processing proceeds, but changes while exhibiting the characteristic of continuity.

Based on these facts, in the apparatus of the present invention, not only the dimensional error and the dimensional error change tendency but also at least the previous and previous dimensional errors are determined so that the correction value is determined in consideration of the characteristics of the correction value. The current correction value is determined based on one correction value.

It should be noted that the processing condition correction means 2 may determine the correction value of the processing condition by using fuzzy inference at least in part. If fuzzy inference is used, rules that are empirically recognized as existing between input information such as dimensional errors and processing conditions, that is, so-called empirical rules, are asked whether they show linearity or nonlinearity. Instead, it can be faithfully expressed, and the correction value of the processing condition can be determined accordingly. Therefore, when the processing condition correcting means 2 is employed, the processing conditions are more appropriately corrected in relation to various factors relating to the processing. However, the present invention can be implemented as a method of determining a correction value of a processing condition using a control theory other than fuzzy inference, for example, a PID control theory or a modern control theory.

[0023]

As described above, according to the present invention, it is possible to more appropriately correct the processing conditions in relation to various factors relating to the processing than in the past, so that the size of the work is independent of those factors. The effect is obtained that the accuracy can be relatively easily increased.

Further, according to the present invention, since the correction value can be determined in consideration of the characteristics of the correction value, even if only the dimensional error and the dimensional error change tendency relating to the old machined part can be obtained, There is also obtained an effect that the processing conditions can be accurately corrected.

In particular, in the case where the present invention is implemented as a method for determining the correction value of the processing condition using fuzzy inference, the processing condition is corrected in accordance with the rule of thumb relating to the processing. A unique effect is obtained in that the processing conditions can be more appropriately corrected in relation to various factors relating to the processing.

In particular, when the present invention is applied to conventional post-process control, the following specific effects can be obtained. Conventional post-process control includes a difference in dimensional error between two processing parts that are successively processed by a processing tool,
In other words, there is a problem that it is difficult to reduce so-called adjacent variation. However, if the present invention is applied to the post-process control, the processing conditions will be corrected so that the variation between the adjacent processes is reduced. Therefore, the problem of the conventional post-process control is solved, and the unique effect that the post-process control can be made intelligent can be obtained.

[0027]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A working system including a feedback type fixed point correcting apparatus according to one embodiment of the present invention will be described in detail with reference to the drawings.

The present processing system is used for an automobile engine.
A cylinder block having a plurality of cylinder bores is provided as a workpiece to be machined, and a machining hole for machining an inner cylindrical surface of each cylinder bore of each cylinder block is provided for honing the machining hole. That is,
In the present embodiment, a work in which a plurality of processing holes as processing parts are set is one aspect of the “work” in the present invention.

As shown in FIG. 2, this processing system uses a horn as a processing tool at each position of a transfer line (indicated by a white arrow in the drawing) where a plurality of works are transferred in a line. A machine 10 for honing each of the cylinder bores is arranged, and a total number measuring machine 14 for measuring the inner diameter of each machined hole for all the honed workpieces is arranged at a position downstream of the machine.

The processing machine 10 is provided with processing tools for each processing hole (the same number as the number of processing holes in each work).
A driving device for driving each processing tool is also provided for each processing hole. An in-process measuring head 18 is provided inside a horn of each processing tool (this is one mode of the in-process measuring tool).
Is incorporated. Each in-process measuring head 18 moves with the horn, and directly measures the inside diameter of a processing hole by an air micrometer method during processing.

On the other hand, the all-in-one measuring device 14 includes a post-process measuring head 20 (which is an embodiment of the post-process measuring tool) for directly measuring the inner diameter of the processed hole by an electric micrometer method for each processed hole. Have.

Processing machine 10 and in-process measuring head 18
Are connected to the sizing device 22, respectively. The counting device 14 is connected to a control device 28 mainly composed of a computer. The control device 28 is connected to the sizing device 22 and further to a data bank 32 as a device for storing data.

Next, the operation will be described. However, a plurality of processing holes in each work are simultaneously processed by different processing tools, and the operation is common between the processing holes. Therefore, the operation of one of the plurality of processing holes will be representatively described. It shall be. Hereinafter, when the processing machine 10, the in-process measurement head 18, the sizing device 22, the post-process measurement head 20, the control device 28, and the data bank 32 are referred to, it means a portion related to one certain processing hole. I do.

The sizing device 22 has its sizing point calibrated by an operator prior to a series of processing. In particular,
With the master work finished to the same size as the target size being measured by the in-process measuring head 18,
The reference voltage representing the sizing point is measured by the in-process measuring head 18.
Is calibrated by the operator to exactly match the output voltage (representing the measured dimensions) from, so that the difference between them is zero. Then, the sizing device 22 sequentially monitors the inner diameter of the hole being machined through the in-process measuring head 18, and determines that the actual dimension has reached the target dimension when the dimension measured thereby reaches the sizing point. Predictably, a control signal for terminating one processing is output to the processing machine 10 (more precisely, the driving device). Further, when a correction value of the sizing point is input from outside, the sizing device 22 corrects the sizing point in accordance with the correction value, and the inner diameter of the processing hole formed by the processing machine 10 is actually a fixed tolerance. To fit within. That is, in the present embodiment, the fixed size point is one aspect of the determination reference value, and is also one aspect of the “processing condition” in the present invention.

The sizing device 22 determines the sizing point correction value and
Is output from the control device 28. Control device 2
8 is a schematic illustration of the processing value of the machined hole by the 100-measuring machine 14 input as measurement data, and based on the measurement data, the correction value of the fixed point is determined using fuzzy inference while utilizing the data bank 32. To decide.

Specifically, the control device 28 stores in advance the program represented by the flowchart of FIG. 3 in the ROM of the computer thereof, and executes the program to determine the correction value of the fixed point. To decide.

The controller 28 further stores data for fuzzy inference in its ROM in advance. The data for fuzzy inference are as follows: (a) Inference program,
(b) a plurality of membership functions relating to an error value R which is a difference between a measured value X measured by the post-process measuring head 20 and a target value A ₀ , (c) a differential value T of the error value R (hereinafter [005]
8)), a plurality of membership functions relating to the correction value U of the fixed point, and (e) a relation between the error value R, the differential value T and the correction value U. Is defined by a plurality of fuzzy rules.

As for the error value R, as it increases from negative to positive, "NB", "NM", "N
There are seven fuzzy labels that change in the order of "S", "ZO", "PS", "PM" and "PB".
Each membership function is represented graphically in FIG. As for the differential value T, as it increases from negative to positive, “NB”, “NS”,
Five fuzzy labels which sequentially change to "ZO", "PS" and "PB" are prepared, and the respective membership functions are represented by a graph in FIG. Note that “T = tan θ” in FIG.
The differential value T is taken on the horizontal axis of the graph, and [00
58], the differential value T or the least squares
When the slope of the regression line is θ, it is calculated as tan θ.
It represents that it is. Regarding the correction value U, as it increases from negative to positive, “N
B ”,“ NM ”,“ NS ”,“ ZO ”,“ PS ”,“ P
Seven fuzzy labels which sequentially change to "M" and "PB" are prepared, and the respective membership functions are represented by a graph in FIG. In addition, when the correction value U increases, the sizing point increases and the hole diameter becomes larger. Conversely, when the correction value U decreases, the sizing point decreases and the hole diameter decreases. Become.

A plurality of fuzzy rules are shown in the following table.

[0040]

[Table 1]

An example of this fuzzy rule is represented by If R = NS and T = PS then U = ZO.

This fuzzy rule is designed based on the following concept. In-process measurement head 18
Dimensions must be measured under severe conditions such as vibrations due to machining, machining distortion of the work, and temperature mismatch with the dimension measurement object. In addition, the in-process measuring head 18 is not calibrated as frequently as the total measuring device 14, and outputs the measurement result of the machined hole as an electric signal, and the electric signal drifts. Due to these circumstances, the in-process measuring head 18 cannot perform dimension measurement with a precision as high as that of the 100% measuring machine 14, and the in-process measuring head 18 generates an output signal from the in-process measuring head 18 based on such a cause. The error is usually a component having a relatively long period (large swell).

The long-period component of the error is usually caused mostly by the in-process measuring head 18 itself, but there is also the one caused by the 100-percent measuring machine 14 itself.

Based on such circumstances, the fuzzy rule is such that the correction value U decreases as the fuzzy label of the error value R increases (hereinafter, the error value R simply increases. The same applies to other parameters). Of course, the correction value U is designed to decrease as the differential value T increases.

This is specifically expressed in the table (hereinafter referred to as a fuzzy rule table) as follows. That is, for example, when the differential value T is “NS”, the correction value U becomes “PB”, “PM”, “PS”, “ZO”, “NS”, and “NM” as the error value R increases. ], And when the error value R is “NM”, as the differential value T increases in the order of “NS”, “ZO” and “PS”, the correction value U becomes “ It is expressed as a decrease from "PM" to "PS".

The in-process measuring head 18 may fail for some reason, and in this case, its measuring accuracy is suddenly and greatly reduced. Nevertheless, if the correction value U is determined by relying on the output signal from the in-process measuring head 18, the actual dimensional accuracy of the machined hole may deviate from the allowable tolerance range.

In view of such circumstances, the fuzzy rules are divided into a case where the measured value X by the post-process measuring head 20 suddenly decreases and becomes considerably small, and a case where the measured value X suddenly increases and becomes considerably large. , The correction amount U is designed to sufficiently approach zero. In this way, when the in-process measuring head 18 breaks down,
Since the output signal from that point is ignored and machining is performed with the expectation that the fixed point up to the previous time is appropriate this time, the dimensional accuracy of the machined hole is not so strongly affected by the failure of the in-process measuring head 18. Can be kept high.

This is specifically expressed in the fuzzy rule table as follows. That is,
When the error value R is “NB” or “NM” and the differential value T is “NB”, the error value R is “PM” or “PB” and the differential value T is “PB”. Is expressed as that the correction value U is “ZO”.

The program shown in FIG. 3 will be described. In brief, the program first calculates a current error value R _i based on a current measurement value X _i , calculates a differential value T _i thereof, and calculates the error value R _i and the differential value R _i. T _i
This time the true correction value U _i is calculated using fuzzy inference based on
^A candidate correction value U _i , which is a candidate value of ^* , is determined, and is corrected using a past candidate correction value U to obtain a genuine correction value U _i
^* To determine.

Specifically, first, in step S1 (hereinafter, referred to as step S1)
It is simply represented by S1. The same applies to other steps)
, A target value A ₀ of the inner diameter of the machined hole, parameters such as a constant ω and n _max relating to a moving average described later are input from a specific memory (not shown). Subsequently, in S2, the total measurement device 14 outputs the current measurement value X _i (i = 0, 1,...).
) Is input, and in S3, a plurality of measurement values X acquired earlier than the current time from the data bank 32 (see FIG. 7).
_i-1 , X _i-2 ,... (hereinafter simply referred to as past measured values X) are input.

[0051] Then, in S4, based on past measurement value X and the current measured value X _i, to remove its neighboring variation between (short-period component) from the current measured value X _i, the current measurement value A moving average value P _i is calculated for X _i . In particular,

[0052]

(Equation 1)

This is calculated using the following equation. Here, “ω _i ” is a weight coefficient relating to the current measurement value X _i ,
“N _max ” means the number of past measurement values X, respectively.

In the present embodiment, on the transport line between the processing machine 10 and the total measuring machine 14,
There is a work that waits for measurement by 4. The number of workpieces waiting in this way means a dead time in a control system that uses the dimensional information based on the total number measuring device 14 as an input signal and the correction value U of the fixed dimension point as an output signal. Explaining the concept of the dead time concretely, if the number of waiting workpieces is 0,
Since the total counting machine 14 measures the dimension of the work immediately after the end of the machining, the dead time is 1, whereas if the number of waiting works is Y (> 0), the total counting machine 14 In No. 14, since the dimension of the workpiece processed (Y + 1) times before by the processing machine 10 is measured, the dead time is (Y + 1).

In this embodiment, the number of workpieces to be on standby is set to 19 and the dead time is set to 20, and each value of the weighting factor ω _i and the number n
The value of _max and the characteristics of the fuzzy rule and the membership function are set in advance.

In this step, while the number of measured values X currently stored in the data bank 32 does not reach n _max , the current measured value X _i is used as it is as the current moving average value P _i. It has become so.

Therefore, for example, for a plurality of measured values X _i shown in the graph of FIG. 8 (a), a plurality of moving average values P _i shown in the graph of FIG. 8 (b) are calculated. become.

When the execution of S4 is completed, in S5,
An error value R _{i of the} current moving average value P _{i from} the target value A ₀ is calculated. Subsequently, in S6, the data bank 32
, A moving average value P for the past m (≧ 2) times is input, and a least squares regression line is calculated from them and the current moving average value P _i, and its differential value T _i (that is, the calculated minimum value) is calculated. The tan θ when the slope of the square regression line is θ (radian) is calculated.

In this step, while the number of moving average values P currently stored in the data bank 32 does not reach m, the current differential value _Ti is set to 0. Past of the moving average value P required for the calculation of the differential value T _i
This is because not all of them are available yet.

That is, in this embodiment, the measured value X _i obtained one by one for each machined hole is one mode of “dimension” in the present invention, and the error value R _i is “dimension error”.
And the differential value T _i is an aspect of “change tendency”. However, to be precise, the deviation of the current measured value X _{i from} the target value A ₀ is not regarded as “dimensional error”, but the deviation of the current moving average value P _{i from} the target value A _0. Is a “dimensional error”, it can be considered that the current moving average value _Pi is one aspect of the “dimensional” in the present invention.

When the current error value R _i and the differential value T _i are calculated as described above, in S7, the current candidate correction value U _i is calculated based on them by using fuzzy inference. . Note that this calculation method is well known and is not indispensable for understanding the present invention, so that the description will be omitted.

Thereafter, in S8, the data bank 32
, The past two candidate correction values U _i−1 and U _i−2 are input. Then, in S9, a least square regression line is calculated from them and the current candidate correction value U _i , By correcting the candidate correction value U _i to a value assumed by the calculated least squares regression line, the current true correction value U _{i is} corrected.
^* Is calculated. That is, in this step, based on the fact that the genuine correction value U ^* shows continuity that it changes smoothly with the progress of machining, the current candidate correction value U _i is calculated using two past candidate correction values U. Is corrected, the current true correction value U _i ^* is calculated.

In this step, while the number of candidate correction values U currently stored in the data bank 32 does not reach two, the current candidate correction value U _i is used as it is as the current true correction value U _i. ^* It is supposed to be.

Thereafter, in S10, the current measured value X
_i is stored in the data bank 32, the current moving average value P _i is also stored in S11, the current candidate correction value U _i is also stored in S12, and the current authentic correction value U _i ^* is also stored in S13. Is done. Subsequently, in S14, the current authenticity correction value U _i ^* is output to the sizing device 22, and in S15, each of the plurality of fuzzy rules used in the calculation of the degree of conformity in the current execution of S7. Are also stored in the data bank 32. This completes one execution of the program, and then the next execution is performed in steps S2 and subsequent steps.

The number of the fuzzy rule used for calculating the degree of conformity among the plurality of fuzzy rules is stored for the following reason. That is, if the number of the used fuzzy rule is stored, for example, after a series of processing is completed, the type and frequency of the fuzzy rule used in the processing are determined, and the content of the fuzzy rule is used using the same. And the characteristics of each membership function can be tuned easily and accurately.

As is clear from the above description, in the present embodiment, since the genuine correction value U ^* is determined in consideration of not only the error value R but also the differential value T, the above-described in-process measuring head is used. 18, measurement error based on signal drift, etc.
The effect is obtained that the diameter of the machined hole can be machined with high accuracy irrespective of disturbance such as a failure.

The applicant performed the following simulation to confirm the performance of the present embodiment. That is,
Based on a plurality of actual measurement values X _i , the current candidate correction value U _i is obtained from the error value R _i and the differential value T _i using fuzzy inference.
Then, in order to correct the current candidate correction value U _i in consideration of continuity, the current candidate correction value U _i is corrected by the past candidate correction value U to obtain a current true correction value U _i ^* , A first simulation of performing processing in a pseudo manner was performed. Further, for comparison with the present embodiment, a simulation of performing a pseudo process using the determined plurality of candidate correction values U _i as they are (that is, using the continuity thereof without considering their continuity) is also performed. went.

FIG. 9 shows a plurality of measured values X which are the simulation results of the present embodiment, and FIG. 10 shows a plurality of measured values X which are the simulation results of the comparative example. In each of these graphs, the width of the three-sigma limit of the plurality of measured values X as the simulation results is expressed as “difference from the target value”.

As is apparent from these graphs, the gradient of the increase in the “difference from the target value” with respect to the increase in the dead time is smaller in the present embodiment than in the comparative example. In this case where the length is 20, the “difference from the target value” is smaller in the present embodiment than in the comparative example. That is, in the present embodiment, based on the fact that the genuine correction value U ^* indicates continuity, the current genuine correction value U _i
^{Since *} is determined, the effect that the dimensional variation is stabilized almost independently of the increase in the dead time can be obtained.

In this embodiment, the following special effects can be obtained.

In this embodiment, the processing system used by the applicant before the completion of the present invention, that is, the processing machine 10, the in-process measuring head 18, the sizing device 22, the total measuring machine 1
The present invention is implemented by adding a control device 28, a data bank 32, a communication cable, and the like, which are relatively simple and inexpensive to install, to a device mainly composed of 4 or the like. As described above, in the present embodiment, since the main hardware such as the processing machine 10 is kept as it is and the sizing accuracy is improved by software, there is also an effect that the present invention can be implemented simply and inexpensively. can get.

The total number measuring machine 14 is originally disposed at the final position of a certain process, and its main purpose is to prevent a work having a dimensional defect from flowing out to the next process. Slightly different in character from post-process measuring tools dedicated to control. However, since it is common in measuring the dimensions of the work after processing, in the present embodiment, the feedback-type fixed point correction is performed by using the existing 100-measuring machine 14 as a post-process measuring tool. It has become. In other words, the present invention can be implemented by using a post-process measuring tool in addition to the 100-measuring device 14, but in the present embodiment, the feedback result is obtained by effectively utilizing the measurement result of the 100-measuring device 14. Since the fixed point correction of the equation is enabled, the effect of reducing the burden on the implementation of the present invention is also obtained.

If the hole to be machined by the machine 10 is to be fitted with another part without a gap, such as a cylinder bore, the dimensional accuracy of the machined hole and the dimension of the mating partner Accuracy is required.
In the conventional processing system, the variation in the dimensional accuracy of the processing hole cannot be suppressed so small. Therefore, as a countermeasure, the machined hole is divided into a plurality of ranks according to the actual dimensional accuracy, and correspondingly, the mating partner is also divided into a plurality of ranks according to the actual dimensional accuracy and belongs to a matching rank. A work and a mating partner are selected and combined with each other. Therefore, a device for engraving around a processing hole to identify a rank, a device for reading the engraving, a device for selecting a processing hole according to the engraving, a shelf for storing the selected work, and the like are indispensable. However, in the present embodiment, the variation in the dimensional accuracy of the machined hole can be sufficiently suppressed, so that the ranking is not necessary or the number of the ranks is reduced even if necessary. Also, the effect that the cost, labor, space, and the like required for the device can be completely or partially omitted can be obtained.

As is clear from the above description, in the present embodiment, the sizing device 22 constitutes one embodiment of the "working tool control means" of the present invention,
Of S1 to S6, S10, and S11 form one mode of “dimension information acquisition unit” in cooperation with the 100-percentage measuring instrument 14, and among the control devices 28, S7 to S9 and S12 to S14 of FIG. Is a part of "processing condition correcting means".

While the embodiment of the present invention has been described in detail with reference to the drawings, the present invention can be implemented in other modes.

For example, the following steps can be added to the program of FIG. That is, if the absolute value of the differential value T exceeds the threshold once or continuously more than once, it is determined that the in-process measuring head 18 may be faulty, and the operator is notified of the possibility. Can be added to the warning step. In this way, the effect is obtained that the failure can be detected simply and early and an early countermeasure can be taken without the operator diagnosing the presence or absence of the failure of the in-process measuring head 18 each time.

The added step is performed, for example, in order to make effective use of the fact that a plurality of measured values X _i are stored in the data bank 32 and to use the current measured value X _i and the previous measured value X _i. It is also possible to calculate a difference from _i−1 , that is, a variation between adjacent components, and detect a failure of the in-process measuring head 18 based on the difference.

The present invention can be implemented in various modified and improved forms based on the knowledge of those skilled in the art without departing from the scope of the claims.

[Brief description of the drawings]

FIG. 1 is a block diagram conceptually showing the configuration of the present invention.

FIG. 2 is a diagram showing a configuration of a machining system including a feedback type fixed point correction device according to an embodiment of the present invention.

FIG. 3 is a computer ROM of the control device in FIG. 2;
3 is a flowchart showing a program stored in the program.

FIG. 4 is a graph showing a plurality of membership functions stored for an error value R in a ROM of the control device in FIG. 2;

5 is a graph showing a plurality of membership functions stored with respect to a differential value T in a ROM of the control device in FIG. 2;

6 is a graph showing a plurality of membership functions stored for a correction value U in a ROM of the control device in FIG. 2;

7 is a computer RAM of the control device in FIG. 2;
FIG. 3 is a diagram conceptually showing the configuration of FIG.

FIG. 8 is a graph showing an example of a measured value in the above embodiment and a moving average value obtained for the example.

FIG. 9 is a graph showing the performance of the above embodiment.

FIG. 10 is a graph showing the performance of a comparative example with the above embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 10 processing machine 14 total measuring machine 18 in-process measuring head 20 post-process measuring head 22 sizing device 28 control device 32 data bank

──────────────────────────────────────────────────続 き Continued on the front page (58) Fields surveyed (Int.Cl. ⁶ , DB name) G05B 19/404 B23Q 15/04 G05D 3/12

Claims (4)

(57) [Claims] 1. A processing tool control means for controlling a processing tool for sequentially processing at least one processing portion set on each of a plurality of workpieces in accordance with processing conditions, and (a) during and after processing. At least one of the following: dimensional information obtaining means for obtaining the dimensional error of each processing part and its change tendency as dimensional information; and (b) processing condition correcting means for sequentially determining the correction value of the processing condition. Feedback type machining, comprising: determining the current correction value based on the acquired dimensional error and the dimensional error change tendency and at least one correction value determined before the previous time. Condition correction device. 2. The processing condition correction means according to claim 1, wherein
Based on the dimensional error and the dimensional error change tendency, the processing conditions
Correction, which is a candidate for a genuine correction value to be determined for
The values are determined one after another, and the current candidate correction value is determined.
To at least one candidate correction value previously determined.
Or based on the genuine correction value,
2. The filter according to claim 1, wherein the authenticity correction value is determined.
Feedback type processing condition correction device. 3. The processing condition correcting means according to claim 1, wherein
Based on the dimensional error and the dimensional error change tendency, the processing conditions
Correction, which is a candidate for a genuine correction value to be determined for
The values are determined one after another, and the current candidate correction value is determined.
And at least one candidate correction value determined before
Of the candidate correction values based on the order of their determination.
Regression line, and the determined candidate correction value
To a value that satisfies the determined regression line.
The present invention determines the genuine correction value this time.
Or the feedback-type processing condition correction device according to 2. 4. The processing tool according to claim 1, wherein
Machining dimensions for multiple machining sites where machining has been completed
By measuring in the same order as the order,
Dimensional errors are obtained in order for
Change tendency for several dimensional errors,
Obtain as a slope when changing with respect to their acquisition order
The file according to any one of claims 1 to 3, wherein
Feedback processing condition correction device.