JP6952941B1 - Machining condition setting device, machining condition setting method, and electric discharge machining device - Google Patents

Abstract

The machining condition setting device (30) has the desired machining results for die-sinking electric discharge machining, such as the electrode wear rate, which is the wear rate of the tool electrode, the machining speed, and the machining surface, which is the roughness of the machined surface of the workpiece. An input device (31) that receives the roughness and the surface area ratio of the actual machined surface to the area of a specific area in the machined surface of the workpiece, and the discharge current peak value and discharge current pulse of the pre-acquired die-sinking electric discharge machining. A storage device (32) that stores basic data showing the theoretical correlation of machining results with respect to the combination with width, and machining processing from the first stage to the final stage is performed as multiple times of die-sinking electric discharge machining. It is provided with a calculation device (33) that calculates the machining conditions from the first stage to the final stage when the electric discharge machine is performed based on the machining result received by the input device and the basic data stored in the storage device.

Description

The present disclosure relates to a machining condition setting device for setting machining conditions for die-sinking electric discharge machining, a machining condition setting method, and an electric discharge machining device.

In die-sinking electric discharge machining, a voltage pulse is applied between the tool electrodes arranged so as to face each other and the workpiece to repeatedly generate electric discharge in the machining gap, and the arc heat of the generated discharge causes the tool electrodes to be applied to the workpiece. It is a process to form a machined hole that is a transfer of the shape of.

In order to obtain a desired machining shape in die-sinking electric discharge machining, it is necessary to perform multiple machining from rough machining to remove a large amount of workpiece to finish machining to finish the machined surface with desired accuracy. be. Therefore, the operator needs to set the optimum processing conditions for obtaining a desired processing result for a plurality of times of processing. A series of machining conditions in this multiple times of machining is hereinafter referred to as a multi-stage machining condition sequence.

In die-sinking electric discharge machining, the machining results such as machining shape and machining speed are determined by the discharge current pulse for one discharge. Therefore, in order to obtain a desired machining result, the operator needs to set the machining conditions that affect the machining results among the machining conditions of the die-sinking electric discharge machining for the multi-stage machining condition sequence. be. For example, the machining conditions that affect the machining result include a combination of the discharge current peak value and the discharge current pulse width. Since the appropriate setting value of this combination differs not only depending on the machining result but also on the material, shape, size, etc. of the tool electrode and the workpiece, there are innumerable combinations. For this reason, it is difficult for an operator to manually set an appropriate combination from an infinite number of combinations, and a technique for automatically setting a combination is being developed.

The machining condition setting device described in Patent Document 1 automatically performs a multi-stage machining condition sequence corresponding to a machining requirement based on basic data showing a theoretical correlation between a plurality of types of parameters including the machining condition and the machining requirement. It is set in.

Japanese Unexamined Patent Publication No. 2009-279735

However, in the technique of Patent Document 1, the discharge current pulse width with respect to the discharge current peak value is uniquely determined by the processing request such as the electrode consumption rate input by the operator, so that the processed surface quality is also uniquely determined. It ends up. The machined surface quality referred to here is a machined surface quality defined differently from the machined surface roughness represented by Ra or Rz of the ISO (International Organization for Standardization) standard, and is the area of a certain area in the machined surface. It is a machined surface quality defined by the ratio of the surface area of the actual machined surface to.

As the work surface roughness is larger, the surface area of the uneven portion of the work surface is larger, so that the surface area ratio is larger. However, even if the work surface roughness is the same, the surface area ratio may be different. For example, as the discharge current pulse width becomes longer, the surface area for the same surface roughness becomes smaller, so that the surface area ratio also becomes smaller. On the other hand, the shorter the discharge current pulse width, the larger the electrode consumption rate. That is, if the surface area ratio is to be increased, the discharge current pulse width needs to be shortened, so that the electrode consumption rate becomes large. Therefore, in the technique of Patent Document 1, the discharge current pulse width with respect to the discharge current peak value is determined by the electrode consumption rate input by the operator, and the discharge current peak value and the discharge current pulse width in consideration of the surface area ratio. There was a problem that it was not possible to determine the appropriate combination with.

The present disclosure has been made in view of the above, and an object of the present invention is to obtain a processing condition setting device capable of generating appropriate processing conditions according to the input surface area ratio of the workpiece.

In order to solve the above-mentioned problems and achieve the object, the machining condition setting device of the present disclosure has an electrode wear rate and a machining speed, which are the wear rates of tool electrodes, as a desired machining result for die-sinking electric discharge machining. It is provided with an input device that accepts the machined surface roughness, which is the roughness of the machined surface of the workpiece, and the surface area ratio of the actual machined surface to the area of a specific region in the machined surface of the workpiece. Further, the machining condition setting device of the present disclosure is a storage device that stores basic data showing a theoretical correlation of machining results with respect to a combination of a pre-acquired discharge current peak value of die-sinking electric discharge machining and a discharge current pulse width. To be equipped. Further, the processing condition setting device of the present disclosure sets the processing conditions from the first stage to the final stage when the processing processes from the first stage to the final stage are performed as the die-cutting discharge processing over a plurality of times. , It is provided with an arithmetic unit that calculates based on the processing result received by the input device and the basic data stored in the storage device.

The processing condition setting device according to the present disclosure has an effect that it is possible to generate appropriate processing conditions according to the input surface area ratio of the workpiece.

The block diagram which shows the structure of the electric discharge machining apparatus provided with the machining condition setting apparatus which concerns on Embodiment 1. Diagram for explaining the relationship between the discharge current waveform and the roughness of the machined surface The figure for demonstrating the relationship between the discharge current pulse width, the machined surface roughness Ra and the surface area ratio Sdr. The figure which shows the example of the surface roughness data used by the machining condition setting apparatus which concerns on Embodiment 1. The figure which shows the example of the electrode wear data used by the processing condition setting apparatus which concerns on Embodiment 1. A flowchart showing a processing procedure for calculating a processing condition sequence by the processing condition setting device according to the first embodiment. The figure for demonstrating the calculation process of the discharge current peak value and the discharge current pulse width by the processing condition setting apparatus which concerns on Embodiment 1. A flowchart showing a processing procedure for calculating a processing condition sequence by the processing condition setting device according to the second embodiment. A block diagram showing a configuration example of the learning device according to the fourth embodiment. A flowchart showing a processing procedure of a learning process by the learning device according to the fourth embodiment. A block diagram showing a configuration example of the inference device according to the fourth embodiment. A flowchart showing a processing procedure of inference processing by the inference device according to the fourth embodiment. The figure which shows the hardware configuration example which realizes the learning apparatus which concerns on Embodiment 4.

Hereinafter, the machining condition setting device, the machining condition setting method, and the electric discharge machining device according to the embodiment of the present disclosure will be described in detail with reference to the drawings.

Embodiment 1.
FIG. 1 is a block diagram showing a configuration of an electric discharge machining apparatus including the machining condition setting device according to the first embodiment. The electric discharge machining device 1 is a die-sinking electric discharge machining device that performs die-sinking electric discharge machining on an object to be machined. The electric discharge machine 1 executes a plurality of times from rough machining to finish machining in order to obtain a machining result required by an operator.

The electric discharge machining device 1 includes a machining condition setting device 30, a machining power supply 34, and a machining mechanism 35. The machining condition setting device 30 is a computer that sets a multi-stage machining condition sequence for obtaining a machining result input by an operator. That is, the machining condition setting device 30 is a computer that sets a combination of machining conditions from rough machining to finish machining in order to obtain a machining result desired by the operator. In the first embodiment, the processing result desired by the operator includes the processing surface area of the work piece. Therefore, the machining condition setting device 30 sets an appropriate multi-stage machining condition sequence according to the machining surface area of the workpiece. The multi-stage processing condition sequence is a combination of processing conditions from the first stage to the final stage. Of the multi-stage machining condition sequence, the first-step machining condition is the coarsest machining condition, and the final-step machining is the smoothest machining condition.

The processing condition setting device 30 includes an input device 31, a storage device 32, and an arithmetic unit 33. The input device 31 receives the machining result when the machining result desired by the operator is input to the machining condition setting device 30. The machining result input by the operator to the input device 31 is, for example, a physical quantity representing the machining result. By inputting a specific physical quantity, it becomes possible to set machining conditions in detail for a desired machining result. Since the processing result received by the input device 31 is the processing result desired by the operator, it can be said that the processing result is a processing request by the operator.

The processing results in the first embodiment are the processing surface roughness, the processing surface area, the electrode consumption rate, the processing speed, the reduction allowance, the processing area, and the like. Examples of machined surface roughness are Ra and Rz specified by ISO, machined surface quality specified by the German Engineers Association standard (VDI: Verein Deutscher Ingenieure), etc., and are indicated by numerical values in the standard desired by the operator. Is done. Ra is the arithmetic mean roughness of the machined surface, and Rz is the maximum height between the peak line and the valley bottom line of the machined surface. Hereinafter, a case where the machined surface roughness Ra is applied as the machined surface roughness will be described.

The processed surface area is the actual surface area ratio of the processed surface to the area of a specific region in the processed surface (hereinafter, referred to as surface area ratio Sdr). The area of the specific area in the machined surface is the area when the specific area is viewed from the upper surface. The surface area of the actual machined surface is an area that takes into consideration the shape of the specific region in the height direction. The surface area of the actual machined surface corresponds to the area when the actual machined surface is developed in a plane parallel to a specific region. As the surface area ratio Sdr, it is possible to input a value within the range corresponding to the machined surface roughness input to the input device 31. That is, the machined surface roughness and the surface area ratio Sdr within the range corresponding to the machined surface roughness are input to the input device 31.

The electrode wear rate is the wear rate of the electrodes used for die-sinking electric discharge machining. Specifically, the electrode wear rate is the ratio of the amount of work to be processed to the amount of electrode wear. The machining speed is the speed of die-sinking electric discharge machining, and corresponds to the formation speed of machined holes formed in the workpiece.

Since the electrode consumption rate and the processing speed cannot be compatible with each other, the operator selects the desired degree of emphasis from the information indicating the degree of emphasis in a stepwise manner and inputs it to the input device 31. do. The machining condition setting device 30 of the first embodiment describes a case where the machining conditions are set so as to achieve the maximum machining speed while satisfying the electrode consumption rate desired by the operator.

The storage device 32 stores the basic data acquired in advance in the experiment. The basic data is a data table showing the processing results for the combination of the discharge current peak value and the discharge current pulse width. Specifically, the basic data includes the machined surface roughness, the machined surface area, the electrode wear rate, and the machined for all or part of the combination of the discharge current peak value and the discharge current pulse width that can be output by the electric discharge machine 1. It is a data table composed of a combination of each experimental data of speed and reduction allowance. That is, the basic data includes data tables such as processed surface roughness, processed surface area, electrode consumption rate, processing speed, and reduction allowance. Details of the basic data will be described later.

The arithmetic unit 33 calculates a multi-stage processing condition sequence based on the processing result input by the operator to the input device 31 and the basic data stored in the storage device 32. The arithmetic unit 33 outputs a command corresponding to each machining condition included in the multi-stage machining condition sequence to the machining power supply 34.

Machining conditions in die-sinking electric discharge machining include electrical conditions of tool electrodes and workpieces, movement conditions of tool electrodes, control conditions related to specific control, environmental conditions, and the like. The electrical conditions include discharge current peak value, discharge current pulse width, pause time, servo reference voltage, polarity, no-load voltage, and the like. The moving conditions are the jump speed, jump time, swing amount, and the like of the tool electrode. The control conditions include a condition for controlling the waveform of the discharge current pulse, a condition for controlling the waveform of the voltage pulse applied to the machining gap, and the like. Environmental conditions include the presence or absence of a machining fluid jet, the method of jetting the machining fluid, and the like.

Among the machining conditions of die-sinking electric discharge machining, the discharge current peak value and the discharge current pulse width have a great influence on the machining results such as the machining surface roughness, the machining surface quality, the electrode consumption rate, and the machining speed. Therefore, the machining condition setting device 30 sets the combination of the optimum discharge current peak value and the discharge current pulse width for the multi-stage machining condition sequence in order to obtain the optimum machining result according to the desired machining result. do.

The processing power supply 34 applies a voltage pulse to the processing mechanism 35 according to a multi-stage processing condition sequence calculated by the arithmetic unit 33. A tool electrode and a work piece are arranged in the processing mechanism 35. In the machining mechanism 35, a voltage pulse from the machining power supply 34 is applied to the machining electrode 36 between the tool electrode and the workpiece. As a result, electric discharge is repeatedly generated in the machining gap between the tool electrode and the workpiece, and the die-sinking electric discharge machining is performed on the workpiece.

Here, the relationship between the combination of the discharge current peak value and the discharge current pulse width and the machined surface roughness which is the machined surface shape of the workpiece will be described. FIG. 2 is a diagram for explaining the relationship between the discharge current waveform and the roughness of the machined surface. The processing power supply 34 can output various discharge current pulses.

FIG. 2 shows a discharge current waveform 11 which is a first waveform example and a discharge current waveform 12 which is a second waveform example. Further, in FIG. 2, the machined surface 13 of the workpiece when the electric discharge machining is executed with the discharge current waveform 11 and the machined surface 14 of the workpiece when the electric discharge machining is executed with the discharge current waveform 12 It shows that.

In FIG. 2, the machined surfaces 13 and 14 obtained by the two types of discharge current pulses such as the discharge current waveforms 11 and 12 have the same surface roughness such as the machined surface roughness Ra or the maximum height Rz. The case where the surface area ratio Sdr is different is shown. Comparing the discharge current waveforms 11 and 12, the discharge current waveform 11 has a lower discharge current peak value and a longer discharge current pulse width than the discharge current waveform 12, but all have the same machined surface roughness. However, since the machined surface 13 has a larger radial size of the discharge marks formed by one discharge than the machined surface 14, the surface area ratio Sdr of the machined surface 13 is smaller than that of the machined surface 14.

FIG. 3 is a diagram for explaining the relationship between the discharge current pulse width, the machined surface roughness Ra, and the surface area ratio Sdr. The horizontal axis of the graph shown in FIG. 3 is the machined surface roughness Ra, and the vertical axis is the surface area ratio Sdr. FIG. 3 shows the machined surface quality (surface texture) 21 when processed with the discharge current pulse width of the first example and the machined surface quality 22 when processed with the discharge current pulse width of the second example. There is. The discharge current pulse width of the first example corresponding to the machined surface quality 21 is smaller than the discharge current pulse width of the second example corresponding to the machined surface quality 22.

As shown in FIGS. 2 and 3, as the discharge current pulse width becomes longer, the surface area with respect to the same machined surface roughness Ra becomes smaller, so that the surface area ratio Sdr also becomes smaller. Since the surface area of the machined surface may have a great influence on the performance of the workpiece after the die-sinking electric discharge machining, the surface area ratio Sdr needs to be appropriately controlled. The machining condition setting device 30 generates appropriate machining conditions (discharge current peak value and discharge current pulse width) according to the input machining surface area (surface area ratio Sdr) of the workpiece.

Next, an example of basic data will be described. FIG. 4 is a diagram showing an example of surface roughness data used by the machining condition setting device according to the first embodiment. FIG. 5 is a diagram showing an example of electrode wear data used by the processing condition setting device according to the first embodiment. The surface roughness data 41, which is the data of the machined surface roughness Ra, and the electrode wear data 42, which is the data of the electrode wear rate, are both examples of basic data. The basic data is data showing the theoretical correlation of the machining results with respect to the combination of the discharge current peak value and the discharge current pulse width of the die-sinking electric discharge machining acquired in advance.

As shown in FIG. 4, the surface roughness data 41 is data indicating the machined surface roughness corresponding to the combination of the discharge current peak value IP and the discharge current pulse width ON. As shown in FIG. 5, the electrode wear data 42 is data indicating the electrode wear rate corresponding to the combination of the discharge current peak value IP and the discharge current pulse width ON.

4 and 5 show the smallest discharge current peak value IP in I _1, m (m is a natural number) indicates a smaller discharge current peak value IP in th at I _m. Further, the smallest discharge current pulse width ON is _{indicated by t 1} , and the n (n is a natural number) thinst discharge current pulse width ON is indicated _{by t n. Im} is the maximum discharge current peak value IP, and t _n is the maximum discharge current pulse width ON.

For example, when electric discharge machining is performed with a combination of the m-th smallest discharge current peak value _Im and the n-th smallest discharge current pulse width t _{n , the machined surface roughness Ra is Ramn} and electrode consumption. The rate is EW _mn .

The machined surface roughness Ra becomes rougher as the discharge current peak value IP is larger. Further, the machined surface roughness Ra becomes coarser as the discharge current pulse width ON becomes smaller. Further, the electrode consumption rate increases as the discharge current peak value IP increases. Further, the electrode consumption rate increases as the discharge current pulse width ON becomes smaller.

The storage device 32 also stores data tables such as the processed surface area (surface area ratio Sdr), the processing speed, and the reduction allowance as basic data. These data tables such as the processed surface area, the processing speed, and the reduction allowance also have the same configurations as the surface roughness data 41 and the electrode wear data 42. For example, in the processing surface area data table, the surface area ratio Sdr corresponding to the combination of the discharge current peak value IP and the discharge current pulse width ON is stored. Similarly, in the processing speed data table, the processing speed corresponding to the combination of the discharge current peak value IP and the discharge current pulse width ON is stored. In the reduction allowance data table, the reduction allowance corresponding to the combination of the discharge current peak value IP and the discharge current pulse width ON is stored.

FIG. 6 is a flowchart showing a processing procedure for calculating a processing condition sequence by the processing condition setting device according to the first embodiment. The input device 31 of the machining condition setting device 30 receives the desired machining result when the machining result desired by the operator such as the reduction allowance and the machining area is input to the machining condition setting device 30 (step S10). The input device 31 sends this processing result to the arithmetic unit 33. The machining result input to the input device 31 includes, for example, desired data such as reduction allowance, machining area, electrode consumption rate, machining surface roughness, machining surface area, and machining speed.

The arithmetic unit 33 calculates the maximum allowable discharge current peak value IP_1 based on the input reduction allowance, processing area, and the like (step S20). IP_1 is the maximum discharge current peak value used in the first-stage electric discharge machining.

Next, the arithmetic unit 33 calculates the discharge current pulse width ON_1 corresponding to the maximum discharge current peak value IP_1 based on the electrode wear data 42, which is the basic data (step S30). In this case, the arithmetic unit 33 determines the maximum pulse width from the selectable discharge current pulse width ON in the electrode wear data 42 to be the discharge current pulse width ON_1. ON_1 is the discharge current pulse width used in the first-stage electric discharge machining.

The arithmetic unit 33 calculates the machined surface roughness Ra_1, which is the maximum machined surface roughness, based on the maximum discharge current peak value IP_1, the discharge current pulse width ON_1, and the surface roughness data 41 (step). S40). Ra_1 is the machined surface roughness when the first-stage electric discharge machining is completed.

Based on the determined maximum machined surface roughness Ra_1 and the machined surface roughness input as the machining result, the arithmetic unit 33 has the number of steps N (N is a natural number) in the machining condition column and the machining in the machining condition sequence in the middle. The surface roughness Ra_k (k is a natural number from 2 to (N-1)) is calculated (step S50). That is, the arithmetic unit 33 determines the number of stages N and the machined surface roughness Ra_1 to Ra_ (N-1) from the second stage to the (N-1) stage.

The arithmetic unit 33 determines the machined surface roughness of each stage to be smaller than the machined surface roughness of the processing conditions of the previous stage. The arithmetic unit 33 calculates the machined surface roughness according to the logic incorporated in the arithmetic unit 33 in advance.

For example, when the arithmetic unit 33 reduces the machined surface roughness to the same size from the first-stage machined surface roughness Ra_1 to the final-stage machined surface roughness Ra_N according to a predetermined ratio p, N = Log _p (Ra_N / Ra_1) and Ra_k = p × Ra_ (k-1) are used to calculate the number of steps N in the machining condition sequence and the machined surface roughness Ra_k of the kth step.

The arithmetic unit 33 is based on the machined surface roughness Ra_k of each machining condition, the surface roughness data 41, and the electrode wear data 42 calculated by the above procedure, and the discharge current peak of each machining condition other than the final stage. The value IP_k and the discharge current pulse width ON_k are calculated (step S60).

FIG. 7 is a diagram for explaining the calculation processing of the discharge current peak value and the discharge current pulse width by the processing condition setting device according to the first embodiment. In the graph shown in the upper part of FIG. 7, the horizontal axis is the discharge current pulse width and the vertical axis is the machined surface roughness. In the graph shown in the lower part of FIG. 7, the horizontal axis is the discharge current pulse width, and the vertical axis is the electrode consumption rate.

The graph shown in the upper part of FIG. 7 shows the correspondence relationships 23 and 24 between the discharge current pulse width and the machined surface roughness. Correspondence relationship 23 is a correspondence relationship between the discharge current pulse width and the machined surface roughness when the discharge current peak value is the discharge current peak value IP_k of the kth stage. Correspondence relationship 24 is a correspondence relationship between the discharge current pulse width and the machined surface roughness when the discharge current peak value is the discharge current peak value IP_ (k + 1) of the (k + 1) stage.

Further, the graph shown in the lower part of FIG. 7 shows the correspondence relationships 25 and 26 between the discharge current pulse width and the electrode consumption rate. Correspondence relationship 25 is a correspondence relationship between the discharge current pulse width and the electrode consumption rate when the discharge current peak value is the discharge current peak value IP_k of the kth stage. Correspondence relationship 26 is a correspondence relationship between the discharge current pulse width and the electrode consumption rate when the discharge current peak value is the discharge current peak value IP_ (k + 1) in the (k + 1) stage. Here, k is, for example, k = 1.

Under the k-th machining condition, the machined surface roughness is Ra_k, the discharge current peak value is IP_k, and the discharge current pulse width is ON_k. _{The arithmetic unit 33 searches the surface roughness data 41 for the discharge current pulse width t k + 1,} which is the machined surface roughness Ra_ (k + 1) of the (k + 1) stage when IP_k is applied to the (k + 1) stage. do.

Next, the arithmetic unit 33 refers to the electrode wear data 42 and calculates the electrode wear rate EW1 corresponding to the combination of _{the searched discharge current pulse width t k + 1 and the discharge current peak value IP_k.} The arithmetic unit 33 determines whether or not the electrode consumption rate EW1 is equal to or less than the desired electrode consumption rate EW.

If EW ≧ EW1, the arithmetic unit 33 sets IP_k and t _{k + 1} as the discharge current peak value IP_ (k + 1) and the discharge current pulse width ON_ (k + 1) in the (k + 1) stage, respectively. That is, if EW ≧ EW1, the arithmetic unit 33 applies the discharge current peak value IP_k of the k-th stage to the discharge current peak value IP_ (k + 1) of the (k + 1) stage.

On the other hand, if EW1> EW, the arithmetic unit 33 has a discharge current of Ra_ (k + 1) with respect to a discharge current peak value IP_ (k-1) that is one step lower than the IP_k that can be set in the electric discharge machine 1. The pulse width t _{k + 1} is searched from the surface roughness data 41.

The computing device 33 refers to the electrode wear data 42 _{and calculates the electrode wear rate EW2 corresponding to the combination of the searched discharge current pulse width t k + 1} and the discharge current peak value IP_ (k-1). The arithmetic unit 33 determines whether or not the electrode consumption rate EW2 is equal to or less than the desired electrode consumption rate EW.

The computing device 33 performs a process of reducing the discharge current pulse width and reducing the discharge current peak value until it finds a combination of the discharge current pulse width and the discharge current peak value that satisfy the desired electrode consumption rate EW. The process is repeated. Thereby, the arithmetic unit 33 can determine the combination of the discharge current peak value and the discharge current pulse width under all the processing conditions except the final stage.

The arithmetic unit 33 uses the combination of the discharge current peak value and the discharge current pulse width in the processing conditions of the final stage as the processing surface roughness Ra_N of the final stage, the surface roughness data 41, and the basic data of the processing surface area. Calculated based on.

As described above, the basic data of the processed surface area is the data indicating the processed surface area corresponding to the combination of the discharge current peak value and the discharge current pulse width. Hereinafter, the basic data of the processed surface area is referred to as surface area data.

The arithmetic unit 33 searches the surface roughness data 41 for a combination of the discharge current peak value and the discharge current pulse width that can obtain the machined surface roughness Ra_N of the final stage. Next, the arithmetic unit 33 reads out the processed surface area corresponding to the combination of the searched discharge current peak value and the discharge current pulse width based on the surface area data. The arithmetic unit 33 determines whether or not the read processed surface area satisfies a desired processed surface area.

When the read processed surface area does not satisfy the desired processed surface area, the arithmetic unit 33 searches for a combination of the discharge current peak value and the discharge current pulse width that satisfy the desired processed surface area. That is, the arithmetic unit 33 searches for a combination of the discharge current peak value and the discharge current pulse width that can obtain the machined surface roughness Ra_N of the final stage and satisfy the desired machined surface area. The arithmetic unit 33 can obtain the machined surface roughness Ra_N of the final stage based on the surface area data 41 and the surface area data, and has a discharge current peak value and a discharge current pulse width that satisfy the desired machined surface area. Extract all combinations.

The arithmetic unit 33 determines, among the combinations satisfying the desired machined surface roughness Ra_N and the machined surface area, the combination having the longest discharge current pulse width as the processing condition of the final stage. That is, the arithmetic unit 33 calculates the discharge current peak value IP_N and the discharge current pulse width ON_N as the processing conditions of the final stage (step S70). As described above, the arithmetic unit 33 can reduce the electrode consumption rate as much as possible by selecting the processing conditions in which the discharge current pulse width is long.

The arithmetic unit 33 determines the machining conditions obtained in the processes of steps S20 to S70 in a multi-stage machining condition sequence (step S80). The processing power supply 34 generates a discharge current pulse according to a multi-stage processing condition sequence calculated by the arithmetic unit 33, and applies discharge energy to the processing poles 36.

By the procedure described with reference to FIG. 6 above, the arithmetic unit 33 can automatically calculate the machining conditions that satisfy the machined surface roughness, the electrode wear rate, and the machined surface area desired by the operator. Further, the arithmetic unit 33 sets the fastest machining condition sequence that satisfies the desired machining surface roughness, electrode consumption rate, and machining surface area by selecting the combination of the discharge current peak value and the discharge current pulse width according to the above procedure. be able to. That is, the arithmetic unit 33 can set the machining speed to the maximum speed by selecting the combination of the largest discharge current peak value and the smallest discharge current pulse width within the range satisfying the machining result desired by the operator. can.

The arithmetic unit 33 may set the machining conditions so as to obtain the minimum electrode consumption rate while satisfying the machining speed desired by the operator. In this case, the arithmetic unit 33 uses the basic data of the processing speed for the combination of the discharge current peak value and the discharge current pulse width instead of the electrode wear data 42. Then, the arithmetic unit 33 uses the processing speed information instead of the electrode consumption rate information in the processing of calculating the processing condition sequence, and executes the same processing as the above-described processing.

As described above, in the first embodiment, the input device 31 accepts the processing conditions when the arithmetic unit 33 performs the processing from the first stage to the final stage as the die-sinking electric discharge machining over a plurality of times. It is calculated based on the machining result and the basic data stored in the storage device 32. The processing result received by the input device 31 includes the processing surface area. Therefore, the machining condition setting device 30 can generate appropriate machining conditions according to the input machining surface area of the workpiece.

Embodiment 2.
Next, the second embodiment will be described with reference to FIG. Since the machining condition setting device 30 according to the second embodiment has the same configuration as the machining condition setting device 30 according to the first embodiment, the description thereof will be omitted.

The machining condition setting device 30 of the first embodiment determines the combination of the discharge current peak value and the discharge current pulse width based on the machining surface area only for the machining conditions of the final stage. Therefore, the electrode consumption rate does not satisfy the desired processing conditions only under the processing conditions of the final stage, but the electrode consumption under the finishing processing conditions is small and can be ignored in some cases.

However, when the operator desires extremely high-precision machining, it is necessary to suppress the electrode consumption rate. Therefore, the machining condition setting device 30 of the second embodiment enables the desired machining result to be set for each of the multi-stage machining condition sequences. As a result, the machining condition setting device 30 sets the electrode wear rate of the machining condition row of the other stage to be smaller than the desired value even when the machining condition of the final stage exceeds the desired electrode wear rate. By doing so, the total amount of electrode consumption under the entire processing conditions can be set to a desired value. More preferably, the machining condition setting device 30 calculates the machining conditions according to the procedure described in the first embodiment, instead of the operator manually setting all the electrode consumption rates of the multi-stage machining condition sequence. After that, the machining condition setting device 30 calculates the electrode consumption amount again. The arithmetic unit 33 of the machining condition setting device 30 incorporates a logic for calculating the electrode wear amount again after calculating the machining conditions according to the procedure described in the first embodiment. Further, in the second embodiment, the desired processing result includes the total amount of electrode consumption under the entire processing conditions.

FIG. 8 is a flowchart showing a processing procedure for calculating a processing condition sequence by the processing condition setting device according to the second embodiment. Of the processes shown in FIG. 8, the same processes as those executed by the processing condition setting device 30 of the first embodiment described with reference to FIG. 6 will be omitted.

The processes of steps S10 to S70 executed by the machining condition setting device 30 of the second embodiment are the same as the processes of steps S10 to S70 executed by the machining condition setting device 30 of the first embodiment.

The arithmetic unit 33 of the second embodiment calculates the total amount of electrode wear in the entire electric discharge machining based on the electrode wear data 42 after calculating the machining conditions of all stages, that is, after the process of step S70.

The arithmetic unit 33 determines whether or not the total electrode wear amount is larger than the desired electrode wear amount (step S75). When the total electrode consumption is less than or equal to the desired electrode consumption (step S75, No), the arithmetic unit 33 sets the calculated processing conditions as they are in the multi-stage processing condition sequence (step S80).

On the other hand, when the total electrode wear amount is larger than the desired electrode wear amount (step S75, Yes), the arithmetic unit 33 determines the reference value of the electrode wear rate under all processing conditions except the final stage (described above). The desired electrode consumption rate EW) is reset to a value smaller than the desired value. That is, the arithmetic unit 33 sets a value lower than the input electrode wear rate to the desired electrode wear rate (step S76). The arithmetic unit 33 uses the reset electrode consumption rate to determine the discharge current pulse width with respect to the maximum discharge current peak value described in the first embodiment. Then, the arithmetic unit 33 re-executes the processes of steps S30 to S70 in FIG. 8 to determine the combination of the discharge current peak value and the discharge current pulse width under all the processing conditions. By repeating the processes of steps S30 to S70, the arithmetic unit 33 can satisfy the desired electrode consumption amount in the entire processing even if the desired electrode consumption rate is not satisfied only in the final stage processing conditions. It is possible to automatically calculate the processing condition sequence of.

As described above, according to the second embodiment, since the machining condition setting device 30 can set the desired machining result for each of the multi-stage machining condition sequences, the multi-stage that satisfies the desired machining result as a whole. It is possible to calculate the processing condition sequence of.

Embodiment 3.
Next, the third embodiment will be described. In the third embodiment, the basic data generated based on the measured values actually measured by the specific electric discharge machine 1 is stored in all the electric discharge machines 1, and the basic data different for each electric discharge machine 1 is stored. I won't let you.

When there are a plurality of electric discharge machines 1, it is desirable that the basic data to be stored in the storage device 32 of each electric discharge machine 1 is the basic data acquired by one specific electric discharge machine 1. That is, the data of each machining result corresponding to the measured value of the discharge current peak value and the discharge current pulse width acquired by one electric discharge machining device 1 is the basic data used by all the electric discharge machining devices 1. Is desirable. In other words, it is desirable that the basic data used by each electric discharge machine 1 is the basic data obtained from a specific electric discharge machine 1.

When the basic data based on the set value set in the specific electric discharge machine 1 is stored in another electric discharge machine 1, the electric discharge machine 1 may vary due to the variation in the electric circuit of each electric discharge machine 1. , The discharge current peak value and the discharge current pulse width will be different. That is, the discharge current peak value and the discharge current pulse width have an error for each electric discharge machine 1. Therefore, in each electric discharge machining apparatus 1, there is a possibility that an error may occur between the basic data and the actual machining result.

In the third embodiment, the basic data generated based on the actually measured values actually measured by the specific electric discharge machine 1 is stored in each electric discharge machine 1. As a result, even if there is a difference in the discharge current peak value and the discharge current pulse width for each electric discharge machining device 1, the basic data can be numerically corrected to be converted into the basic data suitable for each electric discharge machining device 1. It becomes possible to do.

As described above, according to the third embodiment, the basic data generated based on the measured value actually measured by the specific electric discharge machine 1 is stored in the other electric discharge machine 1, so that the other electric discharge machine 1 can be used. By applying numerical correction to the basic data, it becomes possible to convert it into basic data suitable for the own device.

Embodiment 4.
Next, the fourth embodiment will be described with reference to FIGS. 9 to 13. The electric discharge machining apparatus 1 of the first to third embodiments can automatically set the machining conditions at which the machining speed is theoretically the fastest to satisfy the desired machining result based on the basic data. However, in reality, it is conceivable that the fastest machining conditions may change depending on the material used for the tool electrode, the state of the machining chips at the machining electrode 36, and the like. Therefore, in the fourth embodiment, the machining speed during machining with respect to the machining conditions is learned. That is, the electric discharge machining apparatus 1 has the fastest machining speed from the combination of the discharge current peak value and the discharge current pulse width whose machining result satisfies the value desired by the operator based on the input basic data. Search for things. In the fourth embodiment, the processing results desired by the operator are the electrode wear rate, the processed surface roughness, and the processed surface area.

<Learning phase>
FIG. 9 is a block diagram showing a configuration example of the learning device according to the fourth embodiment. The learning device 50 is a computer that learns a trained model 71 for providing the electric discharge machining device 1 with a discharge current peak value and a discharge current pulse width that maximize the machining speed within a range satisfying a desired machining result. Is. The learning device 50 includes a data acquisition unit 51 and a model generation unit 52.

The data acquisition unit 51 acquires the position of the machining shaft, the discharge current peak value corresponding to the position of the machining shaft, and the discharge current pulse width corresponding to the position of the machining shaft as learning data. The position of the machining axis is information corresponding to the machining speed. The data acquired by the data acquisition unit 51 as learning data is the data actually used during processing. The data acquisition unit 51 is the first data acquisition unit.

The model generation unit 52 learns the processing speed with respect to the discharge current peak value and the discharge current pulse width set for learning based on the learning data including the discharge current peak value, the discharge current pulse width, and the position of the processing axis. do. That is, the model generation unit 52 generates a learned model 71 that infers the machining speed with respect to the discharge current peak value and the discharge current pulse width set for learning from the position of the machining axis of the discharge machining device 1.

The model generation unit 52 can use known learning algorithms such as supervised learning, unsupervised learning, and reinforcement learning. As an example, a case where reinforcement learning is applied to the model generation unit 52 will be described. In reinforcement learning, an agent (behavior) in a certain environment observes the current state (environmental parameters) and decides the action to be taken. The environment changes dynamically depending on the behavior of the agent, and the agent is rewarded according to the change in the environment. The agent repeats this process and learns the action policy that gives the most reward through a series of actions. Q-learning and TD-learning are known as typical methods of reinforcement learning. For example, in the case of Q-learning, the general update formula of the action value function Q (s, a) is expressed by the following formula (1).

In the formula (1), s _t represents the state of the environment at time t, a _t represents the behavior in time t. By the action a _t, the state is changed to s _{t + 1.} rt _{+ 1} represents the reward received by the change of the state, γ represents the discount rate, and α represents the learning coefficient. Note that γ is in the range of 0 <γ ≦ 1 and α is in the range of 0 <α ≦ 1. Discharge current peak value and the discharge current pulse width action a _t becomes corresponding to the position of the machining axis, position next state s _t of the machining axis, the model generating unit 52, the best behavior in state s _t at time t a _t To learn.

In the update formula represented by the equation (1), if the action value Q of the action a having the highest Q value at time t + 1 is larger than the action value Q of the action a executed at time t, the action value Q is increased. However, in the opposite case, the action value Q is reduced. In other words, the action value function Q (s, a) is updated so that the action value Q of the action a at time t approaches the best action value at time t + 1. As a result, the best behavioral value in a certain environment is sequentially propagated to the behavioral value in the previous environment.

As described above, when the trained model 71 is generated by reinforcement learning, the model generation unit 52 includes a reward calculation unit 53 and a function update unit 54.

The reward calculation unit 53 calculates the reward based on the discharge current peak value, the discharge current pulse width, and the position of the machining axis. The reward calculation unit 53 calculates the reward r based on the processing speed standard. The reward calculation unit 53 increases the reward r (for example, gives a reward of "1") when the machining speed increases (when the reward increase criterion is satisfied), while the reward calculation unit 53 decreases the machining speed (reward). If the reduction criteria are met), the reward r is reduced (for example, a reward of "-1" is given).

The function update unit 54 updates the function for determining the machining speed according to the reward calculated by the reward calculation unit 53, and outputs the function to the trained model storage unit 70 arranged outside the learning device 50. For example, in the case of Q learning function updating unit 54, the processing action value represented by the formula (1) function Q (s _t, a _t) and, for the set discharge current peak value and the discharge current pulse width for learning It is used as a function to calculate the speed.

The function update unit 54 repeatedly executes the above learning. Learned model storage unit 70, action value is updated by the function updating unit 54 function _{Q (s t, a t)} , i.e., storing the learned model 71.

Next, the processing procedure of the learning process by the learning device 50 will be described with reference to FIG. FIG. 10 is a flowchart showing a processing procedure of the learning process by the learning device according to the fourth embodiment. The data acquisition unit 51 acquires the discharge current peak value, the discharge current pulse width, and the position of the machining axis as learning data (step S110).

The model generation unit 52 calculates the reward based on the discharge current peak value, the discharge current pulse width, and the position of the machining axis (step S120). Specifically, the reward calculation unit 53 acquires the discharge current peak value, the discharge current pulse width, and the position of the machining axis, calculates the machining speed from the position of the machining axis, and determines a predetermined machining speed reference (reward). Determine whether to increase or decrease the reward based on the criteria). The machining speed standard is whether the machining speed has increased or decreased.

The reward calculation unit 53 increases the reward when it determines to increase the reward, and decreases the reward when it determines to decrease the reward. That is, the reward calculation unit 53 increases the reward when the machining speed calculated from the position of the machining axis increases (step S130). On the other hand, the reward calculation unit 53 reduces the reward when the machining speed calculated from the position of the machining axis decreases (step S140). The reward calculation unit 53 does not have to increase or decrease the reward if there is no change in the processing speed.

Function update unit 54 based on the compensation calculated by compensation calculation unit 53 updates the action value learned model storage unit 70 is represented by the formula (1) for storing function Q (s _t, a _t) (Step S150). As a result, the machining speed corresponding to the discharge current peak value and the discharge current pulse width is learned.

Learning apparatus 50 repeatedly executes the steps from S150 from step S110 described above, the generated action-value function Q (s _t, a _t) a, and stores the learned model storage unit 70 as the learned model 71.

The learning device 50 according to the fourth embodiment stores the learned model 71 in the learned model storage unit 70 provided outside the learning device 50, but the learned model storage unit 70 is stored in the learning device 50. It may be provided inside.

<Utilization phase>
FIG. 11 is a block diagram showing a configuration example of the inference device according to the fourth embodiment. The inference device 60 includes a data acquisition unit 61 and an inference unit 62. The data acquisition unit 61 is the second data acquisition unit.

The data acquisition unit 61 acquires the discharge current peak value and the discharge current pulse width as data for inferring the processing speed. The data acquisition unit 61 acquires, for example, the discharge current peak value and the discharge current pulse width described with reference to FIGS. 4 and 5.

The inference unit 62 infers the processing speed corresponding to the discharge current peak value and the discharge current pulse width by using the learned model 71 stored in the learned model storage unit 70. That is, the inference unit 62 inputs the discharge current peak value and the discharge current pulse width acquired by the data acquisition unit 61 into the trained model 71, so that the processing speed suitable for the discharge current peak value and the discharge current pulse width is suitable. Can be inferred. The inference unit 62 outputs the obtained machining speed to the storage device 32 of the electric discharge machining device 1.

In the present embodiment, the case where the inference device 60 uses the learned model 71 learned by the model generation unit 52 of the learning device 50 has been described, but the inference device 60 has learned from another learning device 50. The finished model 71 may be used. In this case as well, the inference device 60 outputs the machining speed based on the learned model 71 acquired from the other learning device 50. The other learning device 50 is a device that learns the learned model 71 from another electric discharge machining device different from the electric discharge machining device 1. That is, the inference device 60 may infer the machining speed suitable for the discharge current peak value and the discharge current pulse width by using the learned model 71 learned by another electric discharge machining device.

Next, the processing procedure of the inference processing by the inference device 60 will be described with reference to FIG. FIG. 12 is a flowchart showing a processing procedure of inference processing by the inference device according to the fourth embodiment. The data acquisition unit 61 acquires inference data, which is data for inferring the processing speed (step S210). Specifically, the data acquisition unit 61 acquires the discharge current peak value and the discharge current pulse width set for inference.

The inference unit 62 inputs the discharge current peak value and the discharge current pulse width into the trained model 71 stored in the trained model storage unit 70 (step S220), and processes the discharge current peak value and the discharge current pulse width. Get speed.

The inference unit 62 outputs the obtained data, that is, the processing speed with respect to the discharge current peak value and the discharge current pulse width, to the electric discharge machining apparatus 1 (step S230).

The storage device 32 of the electric discharge machining device 1 stores the output machining speed with respect to the discharge current peak value and the discharge current pulse width as a learning result (step S240). The inference unit 62 determines whether or not there is a combination of the discharge current peak value and the discharge current pulse width that satisfies the desired machining result and the data has not been acquired (step S250).

When there is a combination for which data has not been acquired (step S250, Yes), the inference unit 62 changes the combination of the discharge current peak value and the discharge current pulse width used for inferring the machining speed (step S260). That is, the inference unit 62 changes the combination of the discharge current peak value and the discharge current pulse width to the combination of the discharge current peak value for which data has not been acquired and the discharge current pulse width, and sets the data for inference.

The inference device 60 repeats the processes from steps S210 to S260 until there are no combinations for which data has not been acquired. When there are no combinations for which data has not been acquired (step S250, No), the inference unit 62 determines the discharge current peak value and the discharge current pulse width that maximize the processing speed from the processing speeds stored as learning results. Search for a combination of. That is, the inference unit 62 searches for a combination of the discharge current peak value and the discharge current pulse width that maximize the processing speed within a range that satisfies the desired processing result.

The inference unit 62 sends the combination of the searched discharge current peak value and the discharge current pulse width to the storage device 32 of the electric discharge machining device 1. The storage device 32 stores the combination of the discharge current peak value and the discharge current pulse width that maximize the processing speed within a range that satisfies the desired processing result. As a result, the electric discharge machine 1 sets the combination of the electric discharge current peak value and the electric discharge current pulse width sent from the inference device 60 as the parameters used in the electric discharge machining (step S270). The electric discharge machine 1 executes electric discharge machining by using the combination of the electric discharge current peak value and the electric discharge current pulse width, which are stored in the storage device 32 and have the maximum processing speed.

In the fourth embodiment, the case where reinforcement learning is applied to the learning algorithm used by the inference unit 62 has been described, but the present invention is not limited to this. As for the learning algorithm, in addition to reinforcement learning, supervised learning, unsupervised learning, semi-supervised learning, and the like can also be applied.

Further, as the learning algorithm used in the model generation unit 52, deep learning, which learns the extraction of the feature amount itself, can also be used. The model generator 52 may perform machine learning according to other known methods such as neural networks, genetic programming, functional logic programming, support vector machines, and the like.

The learning device 50 and the inference device 60 may be, for example, devices connected to the electric discharge machine 1 via a network and separate from the electric discharge machine 1. Further, at least one of the learning device 50 and the inference device 60 may be built in the electric discharge machining device 1. Further, the learning device 50 and the inference device 60 may exist on the cloud server.

Further, the model generation unit 52 may learn the machining speed for the combination of the discharge current peak value and the discharge current pulse width by using the learning data acquired from the plurality of discharge machining devices 1. The model generation unit 52 may acquire learning data from a plurality of electric discharge machines 1 used in the same area, or may be collected from a plurality of electric discharge machines 1 that operate independently in different areas. The machining speed with respect to the discharge current peak value and the discharge current pulse width may be learned by using the learning data. It is also possible to add or remove the electric discharge machine 1 for collecting learning data to the target on the way. Further, with respect to a certain discharge processing device, a learning device that has learned the processing speed with respect to the discharge current peak value and the discharge current pulse width is applied to another discharge processing device 1, and the discharge current is applied to the other discharge processing device 1. The machining speed with respect to the peak value and the discharge current pulse width may be relearned and updated.

Here, the hardware configuration of the learning device 50 and the inference device 60 will be described. Since the learning device 50 and the inference device 60 have the same hardware configuration, the hardware configuration of the learning device 50 will be described below.

FIG. 13 is a diagram showing an example of a hardware configuration that realizes the learning device according to the fourth embodiment. The learning device 50 can be realized by the processor 91, the memory 92, the output device 93, and the input device 94.

An example of the processor 91 is a CPU (Central Processing Unit, a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a DSP (Digital Signal Processor)) or a system LSI (Large Scale Integration). Examples of the memory 92 are RAM (Random Access Memory) and ROM (Read Only Memory).

The learning device 50 is realized by the processor 91 reading and executing a computer-executable learning program for executing the operation of the learning device 50 stored in the memory 92. It can be said that the learning program for executing the operation of the learning device 50 causes the computer to execute the procedure or method of the learning device 50. The learning program for executing the operation of the learning device 50 includes a program for learning the machining speed and the like.

The learning program executed by the learning device 50 has a modular configuration including a model generation unit 52, these are loaded on the main storage device, and these are generated on the main storage device.

The input device 94 receives the discharge current peak value, the discharge current pulse width, the position of the processing shaft, and the like and sends them to the processor 91. The memory 92 stores basic data such as surface roughness data 41 and electrode wear data 42. Further, the memory 92 is used as a temporary memory when the processor 91 executes various processes. The output device 93 outputs the machining speed generated by the processor 91 to the machining power supply 34.

The learning program may be a file in an installable or executable format, stored in a computer-readable storage medium, and provided as a computer program product. Further, the learning program may be provided to the learning device 50 via a network such as the Internet. It should be noted that some of the functions of the learning device 50 may be realized by dedicated hardware such as a dedicated circuit, and some may be realized by software or firmware. The machining condition setting device 30 also has the same hardware configuration as the learning device 50.

As described above, in the fourth embodiment, the learning device 50 uses the trained model 71 that infers the machining speed based on the position of the machining axis of the electric discharge machining device 1, the set discharge current peak value, and the discharge current pulse width. It is generating. Further, the inference device 60 infers the machining speed from the position of the machining axis of the electric discharge machining device 1, the set discharge current peak value, and the discharge current pulse width by using the trained model 71. This makes it possible to acquire the set discharge current peak value and the processing speed with respect to the discharge current pulse width according to the position of the processing shaft of the electric discharge machining apparatus 1. The inference device 60 can execute electric discharge machining in a short time while satisfying the desired machining result by applying the combination of the discharge current peak value that maximizes the machining speed and the discharge current pulse width to the electric discharge machining. It will be possible.

The configuration shown in the above embodiments is an example, and can be combined with another known technique, can be combined with each other, and does not deviate from the gist. It is also possible to omit or change a part of the configuration.

1 Discharge processing device, 11,12 Discharge current waveform, 13,14 Machined surface, 21,22 Machined surface quality, 23-26 correspondence, 30 Processing condition setting device, 31,94 Input device, 32 Storage device, 33 Computing device , 34 Machining power supply, 35 Machining mechanism, 36 Machining poles, 41 Surface roughness data, 42 Electrode wear data, 50 Learning device, 51, 61 Data acquisition unit, 52 Model generation unit, 53 Reward calculation unit, 54 Function update unit , 60 inference device, 62 inference unit, 70 trained model storage unit, 71 trained model, 91 processor, 92 memory, 93 output device.

Claims (10)

The desired machining results for die-sinking electric discharge machining include the electrode wear rate, which is the wear rate of the tool electrode, the machining speed, the machined surface roughness, which is the roughness of the machined surface of the work piece, and the work piece. An input device that accepts the surface area ratio of the actual machined surface to the area of a specific area in the machined surface,
A storage device that stores basic data showing the theoretical correlation of the machining results with respect to the combination of the discharge current peak value and the discharge current pulse width of the die-sinking electric discharge machining acquired in advance.
The input device accepts the processing conditions from the first stage to the final stage when the processing from the first stage to the final stage is performed as the engraving discharge processing over a plurality of times. An arithmetic unit that calculates based on the processing result and the basic data stored in the storage device, and
To prepare
A processing condition setting device characterized by this. The processing result is information indicating a physical quantity.
The processing condition setting device according to claim 1. The input device receives the machining result for each die-sinking electric discharge machining.
The processing condition setting device according to claim 1 or 2. When the electrode consumption rate does not satisfy the processing result in the processing of the final stage of the die-sinking electric discharge machining over a plurality of times, the arithmetic unit performs the processing of the stage prior to the final stage. After lowering the electrode consumption rate, the machining conditions from the first stage to the final stage are calculated.
The processing condition setting device according to claim 3, wherein the processing condition setting device is characterized. The machining result includes the total amount of consumption of the tool electrode in the die-sinking electric discharge machining.
The arithmetic unit calculates the processing conditions in which the total amount of electrode consumption from the first stage to the final stage satisfies the processing result.
The processing condition setting device according to claim 4, wherein the processing condition setting device is characterized. The basic data is data generated based on actual measurement values when the die-sinking electric discharge machining is executed in advance.
The processing condition setting device according to any one of claims 1 to 5, wherein the processing condition setting device is characterized. Further equipped with a learning device for learning the processing speed of the die-sinking electric discharge machining,
The learning device is
Acquire learning data including the position of the machining axis of the die-sinking electric discharge machine that performs the die-sinking electric discharge machining using the machining conditions, the discharge current peak value at the position of the machining shaft, and the discharge current pulse width. The first data acquisition unit and
Model generation to generate a trained model for inferring the processing speed corresponding to the discharge current peak value and the discharge current pulse width from the discharge current peak value and the discharge current pulse width using the training data. Department and
Have,
The processing condition setting device according to any one of claims 1 to 6, wherein the processing condition setting device is characterized. An inference device for inferring the processing speed using the trained model is further provided.
The inference device
A second data acquisition unit that acquires the discharge current peak value and the discharge current pulse width at the position of the processing shaft, and
Using the trained model, an inference unit that infers and outputs the processing speed corresponding to the discharge current peak value and the discharge current pulse width acquired by the second data acquisition unit, and
Have,
The processing condition setting device according to claim 7. A storage step in which the storage device stores basic data showing the theoretical correlation of the machining result of the die-sinking electric discharge machining with respect to the combination of the discharge current peak value and the discharge current pulse width of the die-sinking electric discharge machining acquired in advance. ,
The input device has the desired machining results for die-sinking electrical discharge machining, such as electrode wear rate, which is the tool electrode wear rate, machining speed, machined surface roughness, which is the roughness of the machined surface of the workpiece, and the above. An input step that accepts the surface area ratio of the actual machined surface to the area of a specific area in the machined surface of the workpiece,
The input device sets the processing conditions from the first stage to the final stage when the arithmetic unit performs the processing from the first stage to the final stage as the engraving discharge processing over a plurality of times. A calculation step calculated based on the processing result received by the storage device and the basic data stored in the storage device.
including,
A processing condition setting method characterized by this. A processing mechanism that performs die-sinking electric discharge machining and
A machining condition setting device that sets the machining conditions used by the machining mechanism, and
Have,
The processing condition setting device is
The desired machining results for die-sinking electric discharge machining include the electrode wear rate, which is the wear rate of the tool electrode, the machining speed, the machined surface roughness, which is the roughness of the machined surface of the work piece, and the work piece. An input device that accepts the surface area ratio of the actual machined surface to the area of a specific area in the machined surface,
A storage device that stores basic data showing the theoretical correlation of the machining results with respect to the combination of the discharge current peak value and the discharge current pulse width of the die-sinking electric discharge machining acquired in advance.
The input device accepts the processing conditions from the first stage to the final stage when the processing from the first stage to the final stage is performed as the die-sinking electric discharge machining over a plurality of times. An arithmetic unit that calculates based on the machining result and the basic data stored in the storage device.
To prepare
An electric discharge machine characterized by this.

Images