JPH0741471B2 - EDM control device - Google Patents

Description

Detailed Description of the Invention [Industrial applications]

The present invention relates to an electric discharge machining control device which always maintains an optimum electric discharge machining state in electric discharge machining and realizes control such that the electric discharge machining efficiency is maximized.

[Prior art]

In the electrical discharge machining, the machining conditions that can maintain the stable electrical discharge machining state and improve the machining efficiency are the pause time, the discharge duration, the servo reference voltage, and the axis feed rate during machining (gain of the axis feed rate during machining). Fig. 17 is a block diagram showing a conventional electric discharge machining control apparatus using a pause time control method, in which 1 is an electric discharge machining process including an electric discharge phenomenon, and 2 is an electric discharge machining process.
Is a status amount of the electric discharge machining process, 3 is a machining power supply, 4 is a status detector for detecting the status quantity 2, 5 is its detection value, 6 is a pause time setting device for setting a pause time, 7 is its command value, 8 Is a pause time controller for controlling the pause time by the command value from the pause time setter 6 and the detected value 5 of the situation quantity of the electric discharge machining process from the situation detector 4, and 9 is the pause time manipulated variable. Next, the operation of the conventional electric discharge machining control device configured as described above will be described. The operator sets the down time as one of the machining conditions of the electric discharge machining by the down time setting device 6, and performs the machining. The gap between the machining electrode and the workpiece during electrical discharge machining is generally 10 μm
It is as narrow as several tens of μm, and as the machining progresses, machining scraps generated by electrical discharge machining become localized between the machining electrode and the workpiece, and the electrical discharge concentrates on that part, causing secondary discharge,
Abnormal discharge is likely to occur. This kind of situation occurs because the amount of processed powder exceeds the discharge capacity,
As a method of preventing this, an abnormal state is detected, and the pause time is lengthened in accordance with this to control the generation of machining powder. The status detector 4 detects the status quantity 2 of the process and gives the occurrence of abnormal discharge to the pause time controller 8 as a detection value 5. The down time controller 8 controls the down time based on the command value from the down time setting device 6 and the detected value 5 of the status amount of the electric discharge machining process from the status detector 4. The rest time operation amount 9 is given to the machining discharge power source 3. As is clear from the above description, the control of the down time is indispensable for always maintaining a stable electric discharge machining state,
From the standpoint of machining efficiency, time that does not contribute to machining is wasted, and it is important to perform optimal downtime control in order to improve machining efficiency. In order to realize this optimal downtime control, it is necessary to determine the conditions for increasing or decreasing the downtime for the combination of the pulse conditions of the machining power supply, the machining power supply shape, the machining electrode and the material of the workpiece, etc. Yes, and in general, it depends on the technique possessed by skilled workers. In particular, skilled workers monitor the instability of the electrical discharge machining state and change the downtime accordingly. FIG. 18 is a block diagram showing a conventional electric discharge machining control device using a discharge duration control method. In the figure, 1 is an electric discharge machining process including an electric discharge phenomenon, 2 is a status quantity of the electric discharge machining process, 3 is a machining power supply, 4 is a status detector for detecting the status quantity 2, and 10 is a discharge duration for setting a discharge duration. A setter 11 is a discharge duration controller for controlling the discharge duration according to the command value from the discharge duration setter 10 and the detected value 5 of the status quantity of the electric discharge machining process from the status detector 4, 13 is the discharge sustaining manipulated variable. Next, the operation will be described. In particular, when machining is performed using a graphite material as an electrode material, the operator sets the discharge duration as one of the machining conditions of the electrical discharge machining with the electrical discharge duration setting device 10 to perform the machining. Due to the characteristics of electric discharge machining, the longer the electric discharge duration,
Although the amount of consumption of the electrode is reduced and the machining accuracy is improved, on the contrary, if the discharge duration is set too long, the discharge concentrates on the corner portion of the electrode, and secondary discharge or abnormal discharge is likely to occur. Furthermore, when using graphite electrodes, etc.
As shown in FIG. 19, projections 41 are formed at the corners of the electrode 40, which sometimes makes it impossible to process. To prevent this, by detecting abnormal discharge and shortening the discharge duration,
The protrusions formed on the electrode corners are consumed, and the machining is continued under the same conditions until the machining becomes stable. After all the protrusions are removed, the condition is returned to the previous condition and the machining is continued. By repeating this operation, the electrode consumption can be suppressed to the minimum and the processing time can be optimized.
The situation detector 4 that detects the abnormal discharge detects, for example, the amplitude of the movement of the electrode during machining and the amount of increase from the machining deepest value. The status detector 4 detects the status quantity 2 of the process and gives the occurrence of abnormal discharge to the discharge duration controller 12 as a detection value 5. The discharge duration controller 12 controls the discharge duration based on the command value from the discharge duration setting device 10 and the detected value 5 of the status quantity of the electric discharge machining process from the status detector 4. The discharge duration operation amount 13 is given to the machining power supply 3. As is clear from the above description, control of the discharge duration is indispensable for maintaining stable machining, but from the viewpoint of machining accuracy, it is not always positive, and optimal discharge maintenance time control should be performed. Is important for improving processing efficiency. To achieve this optimal discharge duration control,
It is necessary to determine the conditions for increasing or decreasing the discharge duration for the combination of the pulse conditions of the machining power supply, the machining electrode shape, the machining electrode and the material of the workpiece, etc. It depends very much on help. In particular, a skilled worker monitors the degree of instability of the electric discharge machining state and adjusts the electric discharge duration accordingly. FIG. 20 is a block diagram showing a conventional electric discharge machining control device using a servo reference voltage control method. In the figure, 1 is an electric discharge machining process including a segment phenomenon, 2 is a state quantity of the electric discharge machining process, 14 is an electrode control system, and 15 is a machining gap adjusted between the machining electrode and the workpiece by the electrode control system. The distance 4 is a situation detector for detecting the situation quantity 2, 5 is its detection value, 16 is a servo reference voltage setter for setting a servo reference voltage for electrical discharge machining, 17 is its command value (Vref), 18
Is an arithmetic unit that takes out the deviation between the command value 17 (Vref) and the detected value 5. The deviation 18a is given as an input to the electrode control system 14, and the electrode control system 14 processes the deviation 18a to zero. Adjust the gap distance signal 15. FIG. 21 and FIG. 22 V _M is the average gap voltage indicates the relationship between the servo reference voltage command value Vref and machining gap voltage waveform, T _M is the unloading time, FIG. 21 servo reference voltage Command value Vref
22 is high and FIG. 22 is low. 21 when the servo reference voltage command value Vref as shown in FIG high is higher average voltage V _M of the machining gap, unloading time T _M is long, i.e. interpole discharge waiting time after the voltage application increases the As a result, the machining gap distance 15 becomes large. Further, when the servo reference voltage command value Vref as shown in FIG. 22 is low, the average gap voltage V _M is low, unloading time T _M is short, that is, the discharge waiting time after the application of a voltage to the machining gap As a result, the machining gap distance 15 becomes smaller. Therefore, the servo reference voltage command value
When Vref is increased, the distance 15 of the machining gap is increased, which facilitates the removal of machining scraps and stabilizes machining, but the machining speed becomes slow. On the contrary, if the servo reference voltage command value Vref is lowered, the distance 15 of the machining gap becomes small, it becomes difficult to remove the machining waste, and the machining becomes unstable, but the machining speed becomes faster. Normally, at the beginning of machining, the operator determines the servo reference voltage command value based on the machining contents such as machining depth, electrode shape, machining liquid supply method, material of the electrode and workpiece, etc. Set the reference voltage setter 16. As is clear from the above description, the servo reference voltage command value is indispensable for maintaining the stable electric discharge machining state of machining and for increasing the machining speed, and it is necessary to set the optimum servo reference voltage value. It is important for improving efficiency. In order to obtain this optimum servo reference voltage value, the machining depth that changes from moment to moment, machining electrode pulse conditions, the machining electrode supply method in which the machining electrode faces the work piece, and the machining electrode and the work piece It is necessary to determine the combination of materials, etc., and in general, it depends to a great extent on the technique of a skilled worker. Particularly, a skilled worker monitors the degree of instability of the electric discharge machining state and changes the servo reference voltage setting value accordingly. FIG. 23 is a block diagram showing a conventional electric discharge machining control device using a speed gain control method during machining. In the figure,
1 is an electric discharge machining process including an electric discharge phenomenon, 2 is a status quantity of the electric discharge machining process, 4 is a status detector for detecting the status of the process, 5 is a detection value of the status quantity, 43 is an electrode control system, 43a is a machining control system The response speed of machining controlled by, 44 is the speed gain setter that sets the gain of the axis feed speed during machining, 44a is the set value that is output from this time, 45 is the speed gain for multiplying the axis feed speed command value. An amplifier, 45a is an amplified output amplified by a speed gain, 46 is a calculation unit that feeds back the detected value 5 of the situation amount to the servo voltage command value to perform calculation, and 46a is the calculation output. Next, the operation will be described. In the electric discharge machining, first, the velocity gain, which is one of the machining conditions, is set by the velocity gain setter 44. At the time of electric discharge machining, the reference servo reference voltage command value and the detected value of the situation amount of the electric discharge machining process 1 detected by the situation detector 4 are added to the calculation unit 46 to perform calculation, and the calculation output 46a is output to the amplifier 45. To do. In the amplifier 45, and amplified in accordance with the input A (operation output 46a) gain is set at a speed gain setting unit 44 K, it gives the amplified output 45a a (A K ₌ KA) to the electrode control system 43. In the electrode control system 43, the forward / backward response speed 43a of the electrode is changed according to the given output 45a, thereby controlling the electrode of the machining process 1. Figures 24 and 25 show the voltage waveform between the poles and the gain setting value.
In the figure, T _M represents the no-load time. When the speed gain is low, the no-load time T _M fluctuates as shown in FIG. This is because when the value of the velocity gain is low, the response speed of the electrode becomes slow, so that it becomes impossible to follow the variation between the electrodes generated in the previous electric discharge machining process, and it takes time to correct the variation amount. Therefore, the processing speed becomes slow. In addition, when the velocity gain is high, the electrode response speed becomes faster as shown in Fig. 25, and it becomes easier to follow the variation between the electrodes. It also takes less time to solve. Therefore, the 25th
As shown in the figure, the no-load time T _M , that is, the waiting time until discharge is shortened, and the machining speed is increased. However, if the velocity gain is set too high, on the contrary, the electrode will excessively respond to the variation between the electrodes, and the electrodes will hunt, making the machining impossible. The setting value of the speed gain by the speed gain setter 44 is normally set by the operator at the start of processing such as the machining depth, the shape of the electrode,
It is set based on the processing content according to the method of supplying the processing liquid and the combination of the electrode and the material of the workpiece. As is clear from the above description, the speed gain is indispensable for maintaining a stable electric discharge machining state and increasing the machining speed, and setting the optimum speed gain is important for improving machining efficiency. . In order to obtain this optimum speed gain, changes in the machining depth, pulse conditions of the machining power supply, the area where the machining electrode faces the workpiece, the method of supplying the machining fluid, the combination of the machining electrode and the material of the workpiece However, it depends to a great extent on the technique of the skilled worker, and therefore the skilled worker monitors the instability of the electrical discharge machining state and sets the speed gain corresponding to this. The current situation is that it has been changed.

[Problems to be Solved by the Invention]

Since the conventional electric discharge machining control device is configured as described above, it is necessary for the skilled worker to perform optimum rest time control, optimum discharge duration control, optimum servo reference voltage control, or optimum speed gain control. If an attempt is made to set the pause time, discharge duration, servo reference voltage, or speed gain according to a person's own method, the qualitative ambiguous expression included in the method is changed to an appropriate pause time control method, discharge duration control method. , It is difficult to describe as a servo reference voltage control method or a shaft feed speed control method, and individual differences occur. Further, when controlling the down time, the discharge duration, the servo reference voltage or the speed gain automatically according to the instability of the electric discharge machining state on behalf of the skilled worker, the judgment criterion of the skilled worker can be appropriately determined. However, there was a problem in improving the electric discharge machining efficiency. The present invention has been made in order to solve the above-mentioned problems, and the machining conditions effective for maintaining the electric discharge machining state optimally and controlling the electric discharge machining efficiency to the maximum (pause time, It is possible to properly and easily describe the methods of skilled workers concerning the setting of discharge duration, servo reference voltage, speed gain) and criteria for determining the degree of instability of the electrical discharge machining. It is an object to obtain an electric discharge machining control device that can automatically execute and appropriately change it.

[Means for Solving the Problems]

The electric discharge machining control device according to the present invention includes a control means for controlling at least one of machining conditions such as a pause time, a discharge duration time, a servo reference voltage, and a shaft feed speed during machining, and a machining condition of the control means. Knowledge storage means for storing a method effective for controlling, status detection means for detecting at least a machining status required for the method from an electric discharge machining process, and present or past machining status detected by the status detection means A plurality of results obtained from the situation storing means for storing at least one of the quantities, the quantity of the machining situation read from the situation storing means, and the method read from the knowledge storing means in relation to these quantities. And an inference means for obtaining a command value for controlling the processing conditions by the control means. [Operation] In the present invention, a method effective for controlling machining conditions such as pause time, discharge duration time, servo reference voltage, speed gain, etc. is stored in the knowledge storage means. The required processing situation is detected by the situation detecting means, the detected value is stored in the situation storing means, and the reasoning means obtains it from the method from the knowledge storing means and the amount of the processing situation from the situation storing means. A plurality of results obtained are combined to obtain a command value for executing and controlling the optimum machining conditions and adaptively changing the result, and the control means is controlled by this. EDM that maximizes EDM efficiency can be realized, and
It is possible to properly and easily describe a method that includes a qualitative and vague expression of a skilled worker, which is effective for performing optimum control, and automatically execute optimum machining conditions and adaptive changes. It will be possible to do.

【Example】

Embodiments of the present invention will be described below with reference to the drawings. Embodiment 1 FIGS. 1 to 4 show a first embodiment of an electric discharge machining control apparatus according to the present invention. Particularly, a downtime for keeping the electric discharge state optimum and maximizing the electric discharge machining efficiency. It is an electric discharge machining control device for realizing control. FIG. 1 shows an overall configuration diagram. An electric discharge machining process 1 including an electric discharge phenomenon, a machining power supply 3 for supplying machining electric power to the electric discharge machining process 1, and a pause time setting device for setting a pause time of the electric discharge machining. 6 and a pause time controller 8 for controlling the pause time based on the pause time command value 7 from the pause time setter 6 and a command value 23a from an inference unit 23 described later, and output from the pause time controller 8 The resting time manipulated variable 9 is input to the machining power supply 3. Effective method to control at least the down time from EDM process 1 (see Fig. 2)
A plurality of process status quantities 19 required for the output are output to the status detector 20. The situation detector 20 detects the amount of machining situation required for the method, and the detected situation detection value 20a is stored in the situation storage unit 21. The situation storage unit 21 displays the situation detection value 20a.
At least one of the present or past is stored, and the situation quantity 24 read from this is input to the inference unit 23. 22 is an effective method for controlling the pause time (see Fig. 2)
The method 22a read from the knowledge storage unit 22 is input to the inference unit 23. The inference unit 23 includes the situation quantity 24 from the situation storage unit 21 and the knowledge storage unit.
This is to comprehensively determine the optimum pause time control and adaptive change based on the method 22a from 22.
The command value 23a for optimum rest time control determined in 23 is given to the rest time controller 8. In FIG. 1, the part surrounded by the broken line is for generating the optimum command value 23a for the pause time control, and constitutes the main part of this embodiment. Next, the operation of the present embodiment configured as described above, that is, the operation for generating the optimum command value 23a for the pause time control will be described. FIG. 2 shows an example of an effective method for controlling the pause time. To describe such methods I and II appropriately and easily, we use fuzzy sets. For example, the fuzzy set shown in FIG. 3 can be used to store the methods I and II shown in FIG. 2 in the knowledge storage unit 22 in a rule format including an IF antecedent part and a THEN consequent part. That is, "effective discharge efficiency is small", "amplitude of electrode movement during machining is large", which is included in the method I of FIG.
"Deep machining depth", "Large down time", or "Large effective discharge efficiency" included in Method II,
A qualitative vague expression such as "the amplitude of the electrode movement during machining is small", "the machining depth is shallow", and "the down time is small" is described by the membership function as shown in Fig. 3. For example, in the method I, the feature of "the effective discharge rate is small" is that if the effective discharge rate is 25% or less, the value of the membership function is "1" in the sense that it is completely satisfied, and the effective discharge rate is 75%. If the above is satisfied, the value of the membership function is set to "0" in the sense that it is not satisfied at all, and the effective discharge rate is
The value of the membership function is "0-1" in the sense that if it is 25-75%, it is satisfied between 0 and 1. Similarly,
Of "electrode movement amplitude", "machining depth", "pause time" in method I and "effective discharge rate", "electrode movement amplitude", "machining depth", "pause time" in method II A qualitative ambiguous expression can be described appropriately and easily by the membership function. On the other hand, the situation storage unit 21 stores a situation detection value 20a obtained by using the situation detector 20 to measure the machining situation quantity 19 required for the method stored in the knowledge storage unit 22. In the case shown in FIG. 2, the effective discharge rate, the amplitude of the electrode movement, the working depth, and the like. The inference unit 23 performs fuzzy inference based on the method stored in the knowledge storage unit 22 and the situation stored in the situation storage unit 21 according to the procedure shown in FIG.
Determine a. In the figure, 24a, 24b, and 24c are detected values of "effective discharge rate", "amplitude of electrode movement during machining", and "machining depth" stored in the status storage unit 21. In fuzzy reasoning, in each method I and II,
24 examines whether or not the qualitative vague expression of the antecedent part described by the membership function is satisfied, and in the antecedent part, the value of the membership function with the smallest satisfaction (Method I (Detected value 24a in method II, detected value 24c in method II)
Then, the upper limit of the membership function in the consequent part is cut, the membership functions are synthesized so that the membership function always has the largest function value, and the area centroid of the synthesized membership function is Find position CC. This value is the optimum command value 23a for the pause time control. In the present embodiment as described above, an effective technique for performing the downtime control is stored in the knowledge storage unit, and at least the machining status amount required for the technique is stored in the situation storage unit. Since the reasoning unit can comprehensively determine the command value of the pause time from the situation quantity, it is possible to appropriately and easily describe the methods of skilled workers concerning the pause time control, and to optimize the pause time control by those methods. And adaptive changes can be made automatically. In FIG. 4 of the first embodiment described above, the antecedent part of the method describes three machining situations and the consequent part describes one downtime control amount, but there is nothing to limit these. Needless to say, the optimum command value for the pause time control is similarly obtained even when the number of methods is increased. Further, in this embodiment, although it has not been described that the pause time is adaptively changed depending on the instability of the electric discharge machining state, it can be realized by the same idea as described above. In the first embodiment, the fuzzy set is used in the knowledge storage unit and the fuzzy inference is performed in the inference unit. However, it is natural to use the knowledge representation and the inference method used in other general expert systems. It is possible and has the same effect as the above embodiment. Embodiment 2 FIGS. 5 to 8 show a second embodiment of the electric discharge machining control apparatus according to the present invention. Particularly, the electric discharge machining which keeps the idle machining state optimum and maximizes the electric discharge machining efficiency. This is an electric discharge machining control device for realizing duration control. FIG. 5 shows an overall configuration diagram. An electric discharge machining process 1 including an electric discharge phenomenon, a machining power supply 3 for supplying machining electric power to the electric discharge machining process 1, and a discharge duration for setting a discharge duration of the electric discharge machining. A setter 10, and a discharge duration controller 12 that controls the discharge duration based on a discharge duration command value 11 from the discharge duration setter 10 and a command value 30a from an inference unit 30 described later, and discharge The discharge duration manipulated variable 13 output from the duration controller 12 is input to the machining power supply 3. The electric discharge machining process 1 outputs at least a plurality of machining status quantities 25 required for an effective method (see FIG. 6) for performing discharge duration control, and the status quantities 25 are input to the status detector 26. To be done. The situation detector 26 detects the amount of machining situation required for the above method, and the detected situation detection value 26a is stored in the situation storage unit 27. The situation storage unit 27 displays the situation detection value 26a.
The present status is stored in at least one of the present and the past, and the situation quantity 28 continuously extracted from this is input to the inference unit 30. Reference numeral 29 is a knowledge storage unit that stores an effective method (see FIG. 6) for performing discharge duration control, and the method 29a read from this knowledge storage unit 29 is input to the inference unit 30. The inference unit 30 includes the situation amount 28 from the situation storage unit 27 and the knowledge storage unit.
The optimum discharge duration control and the adaptive change are comprehensively determined based on the method 29a from 29.The command value 30 for the optimum discharge duration control determined by the inference unit 30 is determined.
a is provided to the discharge duration controller 12. In FIG. 5, the part surrounded by the broken line is for generating the optimum command value 30a for discharge duration control, and constitutes the main part of this embodiment. Next, the operation of the present embodiment configured as described above, that is, the operation for generating the optimum discharge duration control command value 30a will be described. FIG. 6 shows an example of an effective method for controlling the discharge duration. To describe such methods I and II appropriately and easily, we use fuzzy sets. For example, using the fuzzy set shown in FIG. 7, the methods I and II shown in FIG.
Knowledge storage 29 in rule format consisting of antecedent and THEN consequent
Can be described in. That is, "effective discharge efficiency is small", "amplitude of electrode movement during machining is large", which is included in the method I of FIG.
“Deep machining depth”, “Short discharge duration, or included in Method II,“ Great effective discharge efficiency ”,
Describe the qualitative vague expressions such as "the amplitude of the electrode movement during machining is small", "the machining depth is shallow", and "the discharge duration is large" by the membership function as shown in Fig. 7. . For example, in the method I, the feature of "the effective discharge rate is small" is that if the effective discharge rate is 25% or less, the value of the membership function is "1" in the sense that it is completely satisfied, and the effective discharge rate is 75% or more. If so, the value of the membership function is set to “0” in the sense that it is not satisfied at all, and the effective discharge rate is
The value of the membership function is "0-1" in the sense that if it is 25-75%, it is satisfied between 0 and 1. Similarly,
Method I "Amplitude of electrode movement", "Machining depth", "Discharge duration" and Method II "Effective discharge rate", "Amplitude of electrode movement", "Machining depth", "Discharge duration" The qualitative and ambiguous expression of "" can be properly and easily described by the membership function. On the other hand, the situation storage unit 27 stores the situation detection value 26a obtained by the situation detector 26 detecting the machining situation amount 25 necessary for the method stored in the knowledge storage unit 29. In the case shown in FIG. 6, the effective discharge rate, the amplitude of the electrode movement, the working depth, and the like. In the inference unit 30, the fuzzy inference is performed from the method stored in the knowledge storage unit 29 and the situation stored in the situation storage unit 27 according to the procedure shown in FIG.
Decide 30a. In the figure, 28a, 28b, 28c are detected values of "effective discharge rate", "amplitude of electrode movement during machining", and "machining depth" stored in the status storage unit 27. In fuzzy reasoning, in each method I and II, we check how much these situation quantities 28 satisfy the qualitative ambiguous expression of the antecedent part described by the membership function, and find out in the antecedent part. The value of the membership function with the lowest degree of satisfaction (detected value 28a in method I, detected value 28c in method II)
Then, cut the upper limit of the membership function of the consequent part,
Membership functions are synthesized so as to have the largest function value among the respective membership functions, and the area center of gravity position CG of the synthesized membership functions is obtained. This value is the optimum discharge duration control command value 30a
Is. In the present embodiment as described above, an effective method for performing discharge duration control is stored in the knowledge storage unit, and at least the processing situation amount necessary for the method is stored in the situation storage unit. Since the command value of the discharge duration can be comprehensively determined by the inference unit from the machining situation amount, the optimum discharge duration control can be executed and adaptively adjusted by the technique possessed by the skilled worker regarding the discharge duration control. Changes can be made automatically. In FIG. 8 of the second embodiment, although the antecedent part of the method describes three machining situations and the consequent part describes one downtime control amount, there is nothing to limit these. . Needless to say, the optimum command value for the discharge duration machining is also obtained as the number of methods increases. Further, in this embodiment, although the electric discharge duration is not adaptively changed according to the degree of instability of the electric discharge machining state, the same idea as described above can be realized. In the second embodiment, the fuzzy set is used in the knowledge storage unit and the inference unit performs the fuzzy inference. However, the knowledge representation and the inference method used in other general expert systems may be used. Obviously, this is possible, and the same effect as that of the above-described embodiment can be obtained. Third Embodiment FIG. 9 to FIG. 12 show a third embodiment of the electric discharge machining control device according to the present invention. Particularly, a servo that maintains an optimum electric discharge machining state and maximizes the electric discharge machining efficiency. It is an electric discharge machining control device for realizing reference voltage machining. FIG. 9 shows an overall configuration diagram, 1 is an electric discharge machining process including an electric discharge phenomenon, 2 is a situation quantity output from the electric discharge machining process 4, 4 is a situation detector for detecting the situation quantity 2, The detection value 5 output from this is input to the calculation unit 18. 16 is a servo reference voltage setter for setting the servo reference voltage for electric discharge machining, and the command value output from this
17 (Vref) is input to the calculation unit 18, and in the calculation unit 18, the deviation 18a between the detected value 5 and the command value 17 is extracted and input to the electrode control system 14. The electrode machining system 14 adjusts the machining gap distance signal 15 so that the deviation 18a becomes zero. From the electric discharge machining process 1 and the electrode control system 14, a machining status amount 31 required for at least an effective method for controlling the servo reference voltage (see FIG. 10) is output. Input to the container 32. The situation detector 32 detects the amount of machining situation required for the above method, and the detected situation detection value 32a is stored in the situation storage unit 33. The situation storage unit 33 displays the situation detection value 32a.
At least one of the present and past is stored, and the situation quantity 34 read from this is input to the inference unit 36. Reference numeral 35 is a knowledge storage unit that stores an effective method (see FIG. 10) for performing servo reference voltage control.
The method 35a read from is input to the inference unit 36. The inference unit 36 includes the situation quantity 34 from the situation storage unit 33 and the knowledge storage unit.
The optimum servo reference voltage control and adaptive change are comprehensively determined based on the method 35a from 35.The command value 36a for the optimum servo reference voltage control determined by the inference unit 36 is the servo reference. It is given to the voltage controller 37. Also,
From the servo reference voltage controller 37,
The servo reference voltage manipulated variable 37a is output for 16. In FIG. 9, the portion enclosed by the broken line is for generating the optimum command value 36a for servo reference voltage control.
It constitutes the main part of this embodiment. Next, the operation of the present embodiment configured as described above, that is, the operation for generating the optimum servo reference voltage command value 36a will be described. FIG. 10 shows an example of an effective method for controlling the servo reference voltage. To describe such methods I and II appropriately and easily, we use fuzzy sets. For example, using the fuzzy sets shown in FIG. 11, the methods I and II shown in FIG.
It can be described in the storage unit 35 in a rule format including an IF antecedent part and a THEN consequent part. That is, "effective discharge efficiency is small", "amplitude of electrode movement during machining is large", which is included in the method I of FIG.
“Deep machining depth”, “High servo reference voltage”, or included in Method II, “Great effective discharge efficiency”, “Small amplitude of electrode movement during machining”, “Machining depth” A qualitative ambiguous expression such as "shallowness" and "servo reference voltage is low" is described by the membership function as shown in Fig. 11. For example, in the method I, the feature of "the effective discharge rate is small" is that if the effective discharge rate is 25% or less, it is completely satisfied. Therefore, the membership function value is set to "1" and the effective discharge rate is 75%. If the above is satisfied, the value of the membership function is set to "0" in the sense that it is not satisfied at all, and the effective discharge rate is
The value of the membership function is "0-1" in the sense that if it is 25-75%, it is satisfied between 0 and 1. Similarly,
Method I "Amplitude of electrode movement", "Machining depth", "Servo reference voltage" and Method II "effective discharge rate", "Amplitude of electrode movement", "Machining depth", "Servo reference voltage""
The qualitative and ambiguous expression of can be described appropriately and easily by the membership function. On the other hand, the situation storage unit 33 stores the situation detection value 32a obtained by the situation detector 32 detecting the machining situation amount 31 necessary for the method stored in the knowledge storage unit 35. In the case shown in FIG. 10, the effective discharge rate, the amplitude of the electrode movement, the working depth, etc. The inference unit 36 performs fuzzy inference according to the procedure shown in FIG. 12 from the method stored in the knowledge storage unit 35 and the situation stored in the situation storage unit 33 to determine the optimum command value 36a for the pause time control. In the figure, 34a, 34b, and 34c are situation storage units 3.
These are the detected values of "effective discharge rate", "amplitude of electrode movement during machining" and "machining depth" stored in 3. In fuzzy reasoning, these situation variables 34
Is satisfied with the qualitative vague expression in the antecedent part described by the membership function, and the value of the membership function with the least satisfaction in the antecedent part (detected value 34a, In Method II, the upper limit of the membership function in the consequent part is cut with the detected value 34c), and the membership function is synthesized so that it always has the largest function numerical value among the membership functions. The area center of gravity CG of the membership function is calculated. This value is the optimum servo reference voltage control command value 36a. In the present embodiment as described above, a method effective for performing the servo reference voltage control is stored in the knowledge storage unit, and at least the machining situation amount necessary for the method is stored in the situation storage unit. Since it is configured so that the command value of the servo reference voltage can be comprehensively determined by the inference unit from the machining state amount, it is possible to appropriately and easily describe the technique of the skilled worker concerning the servo reference voltage control, and use the technique. The optimum servo reference voltage processing can be automatically performed and adaptive change can be automatically performed. In FIG. 12 of the third embodiment, the antecedent part of the method describes three processing situations and the consequent part describes one servo reference voltage control amount, but there is nothing to limit these. . Needless to say, a similar optimum command value for servo reference voltage control can be obtained even if the number of methods is increased. Furthermore, although the servo reference voltage is not adaptively changed according to the degree of instability of the electric discharge machining in this embodiment, it can be realized by the same idea as described above. Further, in the third embodiment, the fuzzy set is used in the knowledge storage unit and the inference unit performs the fuzzy inference, but it is naturally possible to use the knowledge representation and the inference method used in other general expert systems. Thus, the same effect as that of the third embodiment can be obtained. Embodiment 4 FIGS. 13 to 16 show a fourth embodiment of the electric discharge machining control apparatus according to the present invention, and particularly a speed for keeping the electric discharge machining state optimum and maximizing the electric discharge machining efficiency. It is an electric discharge machining control device that realizes gain control. FIG. 13 shows the overall configuration. Reference numeral 1 is an electric discharge machining process including an electric discharge phenomenon, 2 is a situation quantity output from the electric discharge machining process 1, and 4 is a situation detector for detecting the situation quantity 2. Reference numeral 5 is a detected value of the situation amount output from the situation detector 4, and this data is input to the calculation unit 46. The calculation unit 46 calculates the servo voltage command value and the detected value 5, and the output 46a of the calculation result is input to the amplifier 45. Reference numeral 44 is a speed gain setting device for setting the feed speed gain at the time of processing, and the setting device 44a is input to the amplifier 45. The amplifier 45 is a calculation unit 46
Electrode amplifies and outputs _{45a { "A K" (KA} )} with an amplification factor of the operation result output "A" speed gain setting value set by the speed gain setting unit 44 (46a) "K" (44a ) of It is given to the control system 43. The electrode control system 43 drives and controls the electrodes according to the given value, and changes and adjusts the response speed 43a of the electrodes. From the electric discharge machining process 1 and the speed gain setter 44, at least the method necessary for controlling the speed gain (14th
The processing status quantity 47 necessary for performing the processing) is output, and this status quantity 47 is input to the status machine 48. The situation detector 48 detects the amount of machining situation required for the above method, and the detected situation detection value 48a is stored in the situation storage unit 49. The situation storage unit 49 uses the situation detection value 48a.
At least one of the present and past is stored, and the situation quantity 50 read from this is input to the inference unit 51. A knowledge storage unit 52 stores a method (see FIG. 14) effective for controlling the gain of the shaft feed speed during machining. The method 52a read from the knowledge storage unit 52 is stored in the inference unit 51. Is entered. The inference unit 51 includes a situation amount 50 from the situation storage unit 49 and a knowledge storage unit.
The optimum speed gain control and adaptive change are comprehensively determined based on the method 52a from 52, and the optimum speed gain control command value 51a determined by the inference unit 51 is determined.
Is given to the speed gain controller 53. The speed gain controller 53 outputs a speed gain manipulated variable 53a to the speed gain setter 44. The part surrounded by the broken line in FIG. 13 is for generating the optimum speed gain control command value 51a, and constitutes the main part of the present embodiment. Next, the operation of the present embodiment configured as described above, that is, the operation for generating the optimum speed gain control command value 53a will be described. FIG. 14 shows an example of an effective method for performing speed gain control. To describe such methods I to IV appropriately and easily, a fuzzy set is used. For example, using the fuzzy sets shown in FIG. 15, techniques I to IV shown in FIG.
Storage section 52 in a rule format consisting of IF antecedent and THEN consequent
Can be described in. That is, “increasing velocity gain”, “small effective discharge rate”, “large amplitude of electrode movement during machining”, “low velocity gain” or technique II included in technique I of FIG. “Increased velocity gain”, “Large effective discharge rate”, “Small electrode amplitude during machining”, “High velocity gain” or “Reduced velocity gain” included in Method III. , “Effective discharge rate is small”, “Amplitude of electrode being processed is large”, “Velocity gain is high” or method
Included in IV are "decrease velocity gain", "large effective discharge rate", "small amplitude of electrode movement during machining",
A qualitative vague expression such as "the velocity gain is low" is described by the membership function as shown in Fig. 15. For example, in the method I, the feature of "the effective discharge rate is small" is that if the effective discharge rate is 25% or less, it is completely satisfied. Therefore, the membership function value is set to "1" and the effective discharge rate is 75%. If the above is satisfied, the value of the membership function is set to "0" in the sense that it is not satisfied at all, and the effective discharge rate is
The value of the membership function is "0-1" in the sense that if it is 25-75%, it is satisfied between 0 and 1. Similarly,
"Amplitude of electrode movement", "amount of change in velocity gain" and "velocity gain" of method I and "effective discharge rate", "amplitude of electrode movement", "amount of change of velocity gain" in method II, " A qualitative ambiguous expression such as "speed gain" can be appropriately and easily described by the membership function. On the other hand, the situation storage unit 49 stores the situation detection value 48a obtained by detecting the machining situation quantity 50 necessary for the method stored in the knowledge storage unit 52 by the situation detector 48. In the case as shown in FIG. 14, it is the variation amount of the velocity gain, the effective discharge rate, the amplitude of the electrode movement, the processing gain, and the like. The inference unit 51 performs fuzzy inference based on the method stored in the knowledge storage unit 52 and the situation amount stored in the situation storage unit 49 according to the procedure shown in FIG. 16 to determine the optimum speed gain command value 51a. In the figure, 50a, 50b, and 50c are status storage units 4
The detected values of “Variation of velocity gain”, “Effective discharge rate”, “Amplitude of electrode movement during machining”, “Effective discharge rate”, and “Amplitude of electrode movement during machining” stored in 9 is there.
In fuzzy reasoning, in each of methods I and II, we check how much these situation quantities 50 satisfy the qualitative vague expression of the antecedent part described by the membership function, and find out how much in the antecedent part. The upper limit of the membership function in the consequent part is cut by the value of the membership function with a low degree of satisfaction (detection value 50b in method I, detection value 50c in method II), and it is always the best among the membership functions. The membership function is synthesized so that it has a large function value, and the area gravity center position CG of the synthesized membership function
Ask for. This value is the optimum speed gain control command value 51a
Is. In the present embodiment as described above, a method effective for performing velocity gain control is stored in the knowledge storage unit, and at least the machining situation amount required for the method is stored in the situation storage unit. Since the command value of the servo reference voltage can be comprehensively determined by the inference unit from the processing situation amount, the technique of a skilled worker regarding the speed gain control can be described appropriately and easily, and the optimum technique can be obtained by those techniques. The speed gain control can be automatically executed and adaptively changed. In FIG. 16 of the above-mentioned embodiment, the antecedent part of the method describes three processing situations and the consequent part describes one speed gain control amount, but there is nothing to limit these. Needless to say, even if the number of methods is increased, an optimum command value for speed gain control is similarly obtained. Further, in this embodiment, the speed gain is not adaptively changed according to the degree of instability in the electric discharge machining, but it can be realized by the same idea as described above. Further, in the above embodiment, the fuzzy set is used in the knowledge storage unit and the inference unit performs the fuzzy inference, but it is naturally possible to use the knowledge expression and the inference method used in other general expert systems. There is the same effect as the above-mentioned embodiment.

【The invention's effect】

As described above, according to the present invention, the effective means for controlling the machining conditions such as the pause time, the discharge duration time, the servo reference voltage, and the speed gain is stored in the knowledge means, and at least the method is required. Since the amount of processing status is detected and stored in the status storage means, and the instruction value of the processing condition can be comprehensively determined in the inference unit from the method and the processing status quantity from these storage means, It is possible to describe appropriately and easily the techniques possessed by a skilled worker for optimal machining condition control such as pause time, discharge duration, servo reference voltage, speed gain, etc. There is an effect that various changes can be automatically made.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram showing a first embodiment of an electric discharge machining control device according to the present invention, and FIG. 2 is an explanatory diagram showing an example of an effective method for performing a pause time control in the first embodiment,
FIG. 3 is an explanatory view showing an example in which the method of FIG. 2 is described using a fuzzy set, FIG. 4 is an explanatory view showing a process of fuzzy inference by the method of FIG. 2, and FIG. The whole block diagram which shows the 2nd Example of the electric discharge machining control apparatus by this invention, FIG. 6 is the explanatory view which shows an example of the effective method for performing the discharge duration control in 2nd Example, FIG. Is an explanatory view showing an example of the case where the method of FIG. 6 is described by using a fuzzy set, FIG. 8 is an explanatory view showing a process of fuzzy reasoning by the method of FIG. 6, and FIG. 9 is according to the present invention. FIG. 10 is an overall configuration diagram showing a third embodiment of an electric discharge machining control device, FIG. 10 is an explanatory diagram showing an example of an effective method for performing servo reference voltage control in the third embodiment, and FIG. 11 is a fuzzy set. Explanatory diagram showing an example when the method of FIG. 10 is described using
FIG. 12 is an explanatory diagram showing the process of fuzzy inference by the method of FIG. 10, and FIG. 13 is a fourth diagram of the electric discharge machining control device according to the present invention.
FIG. 14 is an explanatory diagram showing an example of an effective method for performing velocity gain control in the embodiment of FIG. 13, FIG. 15 is a diagram showing the use of a fuzzy set, and FIG. Explanatory diagram showing an example of describing the method of Fig. 14, Fig. 16
Explanatory diagram showing the process of fuzzy inference by the method of Fig. 14, No. 17
FIG. 18 is a block diagram of a conventional electric discharge machining control apparatus by a pause time control, FIG. 18 is a block diagram of a conventional electric discharge machining control apparatus by an electric discharge duration control, and FIG. 19 is an explanatory view showing a state of an electrode corner section. FIG. 20 is a block diagram of a conventional electric discharge machining control device by servo reference voltage control, FIGS. 21 and 22 are voltage waveform diagrams between electrodes, and FIG. 23 is an embodiment of a conventional electric discharge machining control device by speed gain control. Fig. 24 is an overall configuration diagram showing
The figure shows a voltage waveform between electrodes when the velocity gain is low, and FIG. 25 is a voltage waveform diagram between electrodes when the velocity gain is high. 1 ... EDM process, 3 ... Machining power supply, 6 ... Rest time setting device, 8 ... Rest time controller, 10 ... Discharge duration setting device, 12 ... Discharge duration controller, 16 ... Servo reference voltage setter, 18,46 ... Calculator, 14,43 ... Electrode control system, 20,26,32,48 ... Situation detector, 21,27,33,49 ... Situation storage, 22 , 29,35,52 …… Knowledge memory, 23,30,36,51 ……
Inference section, 37 ... Servo reference voltage control section. The same reference numerals in the drawings indicate the same or corresponding parts.

Claims (3)

[Claims] 1. In electric discharge machining, control means for controlling at least one of machining conditions such as pause time, electric discharge duration, servo reference voltage, and axis feed speed during machining, and control of machining conditions by said control means. Knowledge storage means for storing a method effective for performing the above, a status detection means for detecting at least a machining status required for the method from an electric discharge machining process, and a current or past machining status detected by the status detection means. A plurality of results obtained from the status storing means for storing at least one of the quantities, the quantity of the processing status extracted from the status storing means, and the method read from the knowledge storing means in relation to these quantities. And an inference means for obtaining a command value for controlling the machining conditions by the control means. 2. The knowledge storage means according to claim 1, wherein a method effective for performing pause time control, discharge duration control, and servo reference voltage control describes a condition part to be judged and the condition part. 2. An electric discharge machining control apparatus having a knowledge storage unit that stores a rule that is composed of a consequent unit that describes the contents to be executed when is satisfied or not satisfied. 3. The electric discharge machining control apparatus according to claim 1, wherein the knowledge storage means expresses the antecedent part and consequent part of some or all of the rules by fuzzy sets, and this expression and the membership function for the fuzzy sets. , Or the relationship between the member functions and the functions, and the inference means obtains from the processing situation stored in the situation storage means by fuzzy synthesis and the method related to these situations stored in the knowledge storage means. An electric discharge machining control device, wherein a plurality of results obtained are combined to obtain command values for pause time control, discharge duration control, and servo reference voltage control.