JP6881700B1 - Machining report creation device and wire electric discharge machine - Google Patents

Abstract

The machining report creation device (100) includes the abnormality information which is the information about the abnormality generated during the machining of the discharge machining, and the machining record information which is the information which shows the result of the machining result and the machining state during the machining of the discharge machining. The machining information acquisition unit (10) and the machining information acquisition unit (10) that acquire the machining information that is the information related to the discharge machining, including the machining state information that is the information to be shown, from the wire discharge machine (200). Based on the processing information storage unit (20) that stores processing information for each processing unit and the processing information stored in the processing information storage unit (20), information indicating the location of the abnormality and the type of abnormality is processed. A machining report creation unit (30) that creates a machining report having machining trajectory information which is information added on the trajectory, machining record information, and abnormality peripheral information which is machining state information before and after the occurrence of an abnormality. Be prepared.

Description

The present disclosure relates to a machining report creating apparatus and a wire electric discharge machine for creating a machining report summarizing information on machining results of electric discharge machining performed by a wire electric discharge machine.

In machining of a wire electric discharge machine, by checking information indicating the machining state during machining such as machining voltage and machining speed (hereinafter referred to as "machining state information"), changes in the machining state can be found and the machining conditions can be obtained. It is important to improve the reproducibility of machining by giving feedback on the above, and to investigate the cause of troubles due to poor accuracy of the machined products at an early stage. However, when a wide variety of machining is performed in the machining of a wire electric discharge machine, the machining shape, machining conditions, and setup information differ for each machining, so it is necessary to judge the quality of the machining conditions and investigate the cause in case of trouble. It is necessary to confirm the measurement data in relation to the machining conditions at the time of machining, the setting information such as the machining program, and the setup information. Patent Document 1 discloses a configuration for displaying a list of information related to processing such as setting information such as processing conditions and processing programs, processing state information, and measurement data of processed products.

WO1999 / 059039

However, in Patent Document 1, by displaying a list of information related to machining, operations such as confirmation of machining state and judgment of quality of machining conditions can be efficiently performed, but wire electrode disconnection (hereinafter, "" Information on abnormalities that occurred during machining such as (hereinafter referred to as "disconnection"), short circuit in which the wire electrode and the workpiece come into contact (hereinafter referred to as "short circuit"), or other alarms (hereinafter referred to as "abnormal information") Since it is not recorded in relation to it, if a machining defect occurs due to an abnormality during machining, it is difficult to grasp which part of the listed information is related to the abnormality during machining. It was. Therefore, in order to estimate the cause of the processing defect from the information displayed in the list, there is a problem that it depends largely on the experience and know-how of the worker, and it takes time to investigate the cause of the processing defect. ..

The present disclosure has been made in view of the above, and an object of the present invention is to obtain a processing report creating apparatus capable of providing information necessary for efficiently investigating the cause of processing defects.

In order to solve the above-mentioned problems and achieve the object, the machining report creating apparatus of the present disclosure includes anomaly information which is information about anomalies generated during machining of discharge machining, and is information showing the actual results of machining. Machining information acquisition unit that acquires machining information that is information about discharge machining, including machining record information and machining state information that is information indicating the machining state during machining of discharge machining, and machining information. Based on the processing information storage unit that stores the processing information acquired by the acquisition unit for each processing unit and the processing information stored in the processing information storage unit, information indicating the location of the abnormality and the type of abnormality is stored in the processing locus. a processing trajectory information is information added to the above, a processing record information, a processing reporting unit to create a machining report with abnormal peripheral information is processed status information of the section until after occurrence before occurrence of an abnormality It is characterized by having.

The machining report creating apparatus according to the present disclosure has the effect that it can support the efficient investigation of the cause of the machining defect by providing the information necessary for efficiently investigating the cause of the machining defect. Play.

The figure which shows the structure of the machining report making apparatus which concerns on Embodiment 1. The figure which shows the schematic structure of the wire electric discharge machine shown in FIG. The figure which shows an example of the processing information stored in the processing information storage part shown in FIG. The figure which shows an example of the 1st machining report created by the machining report creating apparatus which concerns on Embodiment 1. The figure which shows an example of the 2nd machining report created by the machining report creating apparatus which concerns on Embodiment 1. The figure which shows an example of the machining report created by the machining report creating apparatus which concerns on Embodiment 1. Flowchart explaining the process of creating a machining report The figure which shows the example of the hardware composition in the case where the function of the processing report creation apparatus which concerns on Embodiment 1 is realized by using a computer system. The figure which shows the structure of the machining report making apparatus which concerns on Embodiment 2. The figure which shows an example of the 1st machining report of the similar machining report created by the machining report creating apparatus which concerns on Embodiment 2. The figure which shows an example of the 2nd machining report of the similar machining report created by the machining report creation apparatus which concerns on Embodiment 2. A flowchart illustrating a process in which the machining report creating apparatus according to the second embodiment creates a similar machining report. The figure which shows the structure of the machining report making apparatus which concerns on Embodiment 3. The figure which shows the structure of the machine learning apparatus which concerns on Embodiment 3. The figure which shows an example of the three-layer neural network which concerns on Embodiment 3. Flowchart for learning process of machine learning device according to the third embodiment Configuration diagram of the improvement proposal section for the machining report creation device according to the third embodiment The figure which shows the display example of the machining report with the countermeasure plan created by the machining report creation apparatus which concerns on Embodiment 3. Diagram explaining the process for obtaining a machining report with countermeasures using the Kaizen Proposal Department The figure which shows the structure of the wire electric discharge machine which concerns on Embodiment 4.

Hereinafter, the machining report creating apparatus and the wire electric discharge machine according to the embodiment of the present disclosure will be described in detail with reference to the drawings. The present disclosure is not limited to this embodiment.

Embodiment 1.
FIG. 1 is a diagram showing a configuration of a machining report creating apparatus according to the first embodiment of the present disclosure. The machining report creating device 100 creates a machining report summarizing information on the machining results of electric discharge machining performed by the wire electric discharge machine 200. FIG. 1 shows a machining report creating device 100 and a wire electric discharge machine 200 connected to the machining report creating device 100. In the following description, the electric discharge machining performed by the wire electric discharge machine may be simply referred to as "electric discharge machining" or "machining". Further, the machining conditions for electric discharge machining may be simply referred to as "machining conditions".

FIG. 2 is a diagram showing a schematic configuration of the wire electric discharge machine 200 shown in FIG. The wire electric discharge machine 200 processes the workpiece 212 by generating an electric discharge in the gap between the wire electrode 218 and the workpiece 212. The wire electric discharge machine 200 has a processing unit 210 that processes the workpiece 212, and a control device 230 that controls the entire wire electric discharge machine 200. The control device 230 includes a machining control section 231 that controls the machining section 210, a machining program 233, and a storage section 232 that stores the machining conditions 234.

The processing unit 210 includes a table 221 on which the workpiece 212 is placed, a drive device 211 for moving the table 221, a power supply 213, a wire bobbin 214 to which the wire electrode 218 is fed, and a pair of power supply electrons in contact with the wire electrode 218. It has 219 and a pair of dies 220 that support the wire electrode 218. The power supply 213 applies a pulse voltage between the power supply 219 and the table 221. The processing unit 210 includes a feed roller 215 that runs the wire electrode 218 unwound from the wire bobbin 214 toward the work piece 212, a recovery roller 217 that collects the wire electrode 218, and a wire electrode that has passed through the work piece 212. It has a lower roller 216 that causes the 218 to travel toward the recovery roller 217. The machining control unit 231 controls the drive device 211 and the power supply 213 according to the machining program 233 stored in the storage unit 232 and the machining condition 234.

FIG. 1 shows a functional configuration of the machining report creating apparatus 100. The machining report creating device 100 includes a machining information acquisition unit 10 that acquires machining information from the wire discharge processing machine 200, a machining information storage unit 20 that stores the machining information acquired by the machining information acquisition section 10, and a machining report from the machining information. The processing report creation unit 30 for creating the above is provided. Further, the machining report creating device 100 includes a display unit 40 for displaying the machining report created by the machining report creating unit 30. The machining report created by the machining report creation unit 30 may be output to a document output device such as a printer (not shown). Further, the machining report creating device 100 includes a setting unit 50 and a storage unit 60, and stores the information set by the setting unit 50 in the storage unit 60. The information stored in the storage unit 60 includes, for example, extraction range designation information described later. The storage unit 60 includes an extraction range designation information storage unit 61 that stores the extraction range designation information, and the extraction range designation information is set in the extraction range designation information storage unit 61 via the setting unit 50.

The machining information includes abnormality information which is information about an abnormality generated during the machining of electric discharge machining, machining record information which is information which shows the result of machining, and machining which is information which shows the machining state during machining of electric discharge machining. Information about electric discharge machining, including state information. Specific examples of machining record information are information such as machining time, machining distance, average machining speed, number of disconnections, number of short circuits, and amount of wire used from the start of machining to the end of machining. Specific examples of machining state information include machining speed, machining voltage, discharge pulse, tension, number of wire traveling pulses which are feed pulses of the feed roller 215, liquid quality of machining liquid, and machining, which are detected at each time during machining. Information such as liquid temperature, room temperature, processing machine body temperature, machine coordinates, etc. Specific examples of the abnormality information are information such as the position where the abnormality occurs, the type of the abnormality, and the number of times the abnormality occurs.

The machining information acquisition unit 10 may acquire machining information in real time or periodically from the machining control unit 231 of the wire electric discharge machine 200 while the wire electric discharge machine 200 is machining, or the wire electric discharge machine 200 may acquire machining information. The machining information may be temporarily stored in the storage unit 232 of the electric discharge machine 200 during machining, and the machining information stored in the storage section 232 may be acquired after the machining is completed.

The processing information storage unit 20 stores the processing information acquired by the processing information acquisition unit 10 for each processing unit. As the processing unit, the processing from the start to the end of processing on the workpiece 212 may be used as the processing unit, or the processing for one processing condition or the processing for one processing program may be used as the processing unit. In FIG. 3, the processing from the start to the end of processing on the workpiece 212 is set as the processing unit, and the processing No. is set for each executed processing unit. Is an example of processing information stored in a database format with. For example, JOTAI01, JISSEKI01, and SYOGEN01 have processing numbers. It shows the processing state information, processing record information, and processing specification information when the processing of No. 1 is performed. It is managed together with the date and time when the processing of 1 was performed. JOTAI01 is the processing No. A series of machining state information acquired by the machining information acquisition unit 10 from the start of machining to the end of machining of 1 is shown. JISSEKI01 is the processing No. The processing record information about the processing of 1 is shown. SYOGEN01 is the processing No. The processing specification information about the processing of 1 is shown. Specific examples of the processing specification information include information on the wire diameter and material of the wire electrode 218, the material and plate thickness of the work piece 212, and the setting parameter information related to adaptive control. FIG. 3 shows an example in which processing information corresponding to a plurality of processings is stored in the processing information storage unit 20, but at least the processing information corresponding to the processing performed this time is stored in the processing information storage unit 20. It should be done. Further, as shown in FIG. 3, not only the machining information corresponding to the machining performed this time but also the machining information corresponding to a plurality of machining is stored in the machining information storage unit 20, and the designated machining No. The machining report for the above may be created by the machining report creation unit 30 described later.

The machining report creating unit 30 creates a machining report based on the machining information stored in the machining information storage unit 20. The machining report has machining record information including abnormality information, machining locus information, and abnormality peripheral information. Other information may be included in the processing report. For example, by making a machining report that summarizes machining specification information and machining conditions together with machining record information, machining trajectory information, and abnormal peripheral information, information on machining results of electric discharge machining is aggregated, and further machining defects are obtained. It is possible to efficiently investigate the cause of the above. The abnormality peripheral information is processing state information before and after the occurrence of the abnormality. The machining locus information is information obtained by adding information indicating the occurrence position of the abnormality and the type of the abnormality on the machining locus. In the first embodiment, the machining report displays the first machining report having the machining record information, the machining locus information, and the machining specification information, and the abnormality peripheral information in a time-series graph format. An example consisting of the machining report of is shown. Examples of the first machining report and the second machining report are shown in FIGS. 4 and 5.

Generally, in wire electric discharge machining, the target finish machining is not performed in one machining, but the same machining is performed while changing the machining conditions so that the machining accuracy is gradually improved from rough machining to finish machining. Machining on the trajectory multiple times. FIG. 4 shows, as an example, a case where the same machining locus is machined three times under three machining conditions. The first time is machining under rough machining conditions (condition number 1) and disconnection A occurs, the second is machining under semi-finishing machining conditions (condition number 2) and short circuit B occurs, and the third is finish machining. It shows the case where the short circuit C occurs after processing under the condition (condition number 3). In the machining locus information of FIG. 4, a disconnection A, a short circuit B, and a short circuit C are displayed on the same machining locus. By displaying in this way, it is possible to easily confirm at which position on the machining locus the abnormality has occurred. As a result, for example, if an abnormality occurs in a characteristic machined shape part such as a corner part, a straight line part, or an arc part, the factors and countermeasures corresponding to the characteristic machined shape should be examined. Is possible.

Further, in the processing record information of FIG. 4, information indicating the processing record when processing is performed under the three processing conditions of condition numbers 1 to 3 is described. FIG. 5 shows, as an example, the abnormal peripheral information regarding the short circuit B and the processing conditions at that time. The abnormality peripheral information in FIG. 5 displays the states before and after the time t when the short circuit B occurs in a time-series graph format for the plurality of machining state information. The horizontal axis indicates time, and the unit on the vertical axis is not shown, but it is a unit of physical quantity corresponding to each processing state information. In FIG. 5, a plurality of processing state information is displayed side by side as separate graphs with a common time axis, but a plurality of processing state information may be superimposed and displayed on the same graph.

By checking the abnormal peripheral information displayed in this way, for example, if the operator can confirm that the machining speed immediately before the short circuit B in FIG. 5 occurs is slower than usual, this is a machining defect. It is possible to presume that it is the cause. In this way, by displaying the machining states before and after the occurrence of an abnormality in a time-series graph format for a plurality of machining state information, it becomes easy for the operator to visually identify the cause of the machining defect. .. Although not shown, similarly, abnormal peripheral information regarding disconnection A and short circuit C is also created. That is, in the second machining report, information on the periphery of the abnormality is created for each abnormality that occurs. By creating such a machining report, when a machining defect due to an abnormality occurs, the operator can check the position where the abnormality occurred, the machining state before and after the abnormality, and when the abnormality occurs. It is possible to easily confirm the processing specifications and processing conditions used for processing, and the processing results related to processing when the abnormality occurs.

In this embodiment, an example in which the machining report is composed of the first machining report and the second machining report is shown, but the machining report is composed of at least the machining record information, the machining locus information, and the abnormality peripheral information. There is no distinction between the first machining report and the second machining report, and the first machining report and the second machining report may be integrated as shown in FIG. 6, for example. Further, in this embodiment, an example in which the abnormality peripheral information is displayed in a graph format in chronological order is shown, but it is sufficient if the processing state before and after the occurrence of the abnormality can be grasped. For example, numerical information of the processing state information before and after the occurrence of the abnormality. May be displayed in a list format.

Next, the process of creating a machining report by the machining report creating apparatus 100 will be described. FIG. 7 is a flowchart illustrating a process of creating a machining report. When the wire electric discharge machine 200 starts machining, the machining report creating device 100 acquires machining information from the wire electric discharge machine 200 (step 1). Specifically, the machining information acquisition unit 10 acquires machining information such as machining record information including abnormality information, machining state information, and machining specification information from the machining control section 231 of the wire electric discharge machine 200. In this case, while the wire electric discharge machine 200 is machining, machining information may be acquired in real time or periodically from the machining control unit 231 of the wire electric discharge machine 200, or the wire electric discharge machine 200 finishes machining. Then, the machining information acquisition unit 10 may collectively acquire the machining information stored in the storage unit 232 of the wire electric discharge machine 200.

Next, the machining information storage unit 20 stores machining information such as machining record information, machining state information, and machining specification information acquired by the machining information acquisition unit 10 for each machining unit (step 2). Next, the machining report creating unit 30 creates machining locus information from the machining information stored in the machining information storage unit 20 (step 3). To create machining locus information, first, information on the machine coordinates of each axis is extracted from the machining state information, and the machining locus is calculated from the relative positional relationship between the wire electrode and the workpiece. Then, the machining locus information is created by extracting the information on the abnormality occurrence position and the abnormality type from the abnormality information and adding the information indicating the abnormality occurrence position and the abnormality type to the calculated machining locus. To do. In this embodiment, as an example, symbols such as ×, Δ, and 〇 are added to the calculated machining locus as information indicating the occurrence position of the abnormality and the type of the abnormality.

Next, the machining report creation unit 30 extracts abnormal peripheral information from the machining information stored in the machining information storage section 20 based on the extraction range designation information set in the extraction range designation information storage section 61 (step 4). ). The extraction range specification information is information that specifies an extraction range for extracting abnormal peripheral information from the processing information stored in the processing information storage unit. Specifically, the extraction range designation information includes the distance from the abnormality occurrence position, the number of blocks from the block corresponding to the abnormality occurrence position among the blocks constituting the machining program, or the time from the abnormality occurrence time. The machining report creation unit 30 extracts the machining state information corresponding to the extraction range designation information as the abnormality peripheral information.

Next, the machining report creation unit 30 creates a machining report having machining record information, machining trajectory information, and abnormal peripheral information (step 5). In the present embodiment, as an example, the first machining report that summarizes the machining specification information, the machining trajectory information, and the machining record information for each machining unit, and the abnormal peripheral information and the machining conditions are processed. A second machining report is created for each unit. The order of step 3 and step 4 may be changed as appropriate, or may be performed in parallel. Next, the display unit 40 displays the processing report created by the processing report creation unit 30 (step 6).

As described above, in the first embodiment, the machining report creating device 100 creates a machining report having machining record information, machining trajectory information, and abnormal peripheral information based on the machining information. As a result, when a machining defect occurs due to an abnormality during machining, the operator can easily grasp which part of the information displayed in the machining report is related to the abnormality during machining. Therefore, it is possible to efficiently investigate the cause of processing defects.

Next, the hardware configuration of the machining report creating apparatus 100 according to the first embodiment will be described. Each functional unit included in the processing report creating apparatus 100 is realized by using a computer system such as a personal computer or a general-purpose computer. FIG. 8 is a diagram showing an example of a hardware configuration when the function of the machining report creating apparatus 100 according to the first embodiment is realized by using a computer system.

The processing report creating device 100 includes a processor 901 that executes various processes, a memory 902 that is a built-in memory, an external storage device 903 that stores various information, and an input / output interface 904 that handles input and output of various information. It has a display 905 that displays various information.

The processor 901 is a CPU (Central Processing Unit). The processor 901 may be a processing device, an arithmetic unit, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor). The function of the processing report creating unit 30 is realized by the processor 901 and software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the external storage device 903. The processor 901 reads the software or firmware stored in the external storage device 903 into the memory 902 and executes it.

Memory 902 is a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EEPROM (Erasable Programmable Read Only Memory) or EEPROM® (Electrically Erasable). Programmable Read Only Memory). The external storage device 903 is an HDD (Hard Disk Drive) or an SSD (Solid State Drive). Each function of the processing information storage unit 20 and the storage unit 60 is realized by using the external storage device 903.

The input / output interface 904 is responsible for inputting information from the wire electric discharge machine 200 and outputting information to the wire electric discharge machine 200. The input / output interface 904 also includes an input device such as a keyboard, mouse or touch panel. The functions of the processing information acquisition unit 10 and the setting unit 50 are realized by using the input / output interface 904. The function of the display unit 40 is realized by using the display 905.

Embodiment 2.
In the first embodiment, the configuration for creating a machining report summarizing the information on the machining results so that the relation with the abnormality generated during the machining can be understood based on the machining information acquired from the wire electric discharge machine has been described. If there is a machining report similar to the created machining report, it is highly possible that the measures taken for this similar machining report can be applied to the newly created machining report. Moreover, even if it cannot be applied as it is, it is highly likely that it will be helpful when considering countermeasures. In the second embodiment, when a new machining report is created, a machining report similar to the newly created machining report is extracted from the machining reports created in the past and presented to the operator. The same parts as those of the first embodiment will be omitted, and the parts different from the first embodiment will be described.

FIG. 9 is a diagram showing a configuration of a machining report creating apparatus according to the second embodiment. The configuration of the machining report creating device 101 according to the second embodiment includes the machining report accumulating unit 70, the countermeasure information input section 71, the similar machining report extracting section 72, and the configuration of the machining report creating device of the first embodiment. Similar determination condition storage unit 62 and the like are added.

The machining report accumulating unit 70 accumulates the machining report created by the machining report creating section 30 each time electric discharge machining is performed. The countermeasure information input unit 71 inputs, as countermeasure information, the countermeasures taken for the abnormality generated during machining into the machining report stored in the machining report storage unit 70. Specifically, after creating the machining report, the worker investigates the cause of the abnormality that occurred during machining while referring to the machining report, and implements countermeasures. The worker inputs the found factor and the content of the countermeasure from the countermeasure information input unit 71 as the countermeasure information.

When the machining report creation unit 30 creates a new machining report, the similar machining report extraction section 72 inputs a machining report similar to this new machining report, and the countermeasures stored in the machining report storage section 70 are input. Extract as a similar machining report from the machining report with countermeasures, which is a machining report.

Specifically, the storage unit 60 includes a similarity determination condition storage unit 62 that stores similar determination conditions, which are conditions for determining that the new processing report and the processing report with countermeasures are similar, and the similar processing report extraction unit 72. Extracts a similar machining report from the machining reports with countermeasures stored in the machining report storage section 70 based on the similarity determination condition set in the similarity determination condition storage unit 62. The similarity determination condition is designated by the setting unit 50 by the processing specification information, the processing condition, the processing program, the processing locus information, and the like, and is set in the similarity determination condition storage unit 62.

The similar machining report extraction unit 72 compares the new machining report with the machining report with countermeasures, and determines that they are similar if at least one of the plurality of conditions set as the similarity judgment conditions is the same. To do. Then, the similar processing report extraction unit 72 outputs the extracted similar processing report to the display unit 40. Display examples of similar machining reports are shown in FIGS. 10 and 11. FIG. 10 shows an example of the first machining report of the similar machining report, and FIG. 11 shows an example of the second machining report of the similar machining report. Similar machining reports may be displayed side by side with the new machining report for easy comparison with the new machining report. The operator can consider the countermeasure contents for the machining defect caused by the abnormality shown in the new machining report by referring to the countermeasure information described in the displayed similar machining report.

Next, the process of creating a similar machining report by the machining report creation device 101 will be described. FIG. 12 is a flowchart illustrating a process of creating a similar machining report.

When the machining report creation device 101 creates a new machining report, that is, when the machining report creation unit 30 creates a new machining report by the same method as that described in the first embodiment, the similar machining report extraction unit 72 acquires a new machining report from the machining report creation unit 30 (step 7). Next, the similar machining report extraction unit 72 sets the new machining report acquired from the machining report creation section 30 and the machining report with countermeasures stored in the machining report storage section 70 to preset similarity determination conditions. Compare based on (step 8). Machining with countermeasures judged to be similar to the new machining report acquired from the machining report creation unit 30 from the machining reports with countermeasures accumulated in the machining report storage unit 70 based on the preset similarity judgment conditions. The report is extracted as a similar processing report (step 9). The similar machining report extraction unit 72 outputs the extracted similar machining report to the display unit 40 (step 10).

As described above, in the second embodiment, when the machining report creating device 101 creates a new machining report, a machining report similar to the newly created machining report is extracted from the machining reports created in the past. Since it is presented to the worker, the worker can efficiently consider the countermeasure content by referring to the countermeasure content taken for this similar machining report.

Embodiment 3.
In the first and second embodiments, a machining report creating device for creating a machining report capable of confirming the machining state during machining and the actual result of machining in relation to an abnormality generated during machining has been described. In Section 3, a machining report creating device capable of presenting a countermeasure plan for machining defects caused by abnormalities generated during machining will be described using the results learned by the machine learning device. The same parts as those of the first and second embodiments will be omitted, and the parts different from the first and second embodiments will be described.

FIG. 13 is a diagram showing the configuration of the machining report creating device 102 according to the third embodiment. The configuration of the machining report creating device 102 according to the third embodiment is such that the machine learning device 80 and the improvement proposal unit 90 are added to the configuration of the machining report creating device of the second embodiment. The similar machining report extraction unit 72 and the similarity determination condition storage unit 62 provided in the machining report creation device of the second embodiment do not necessarily have to be provided in the machining report creation device 102 of the third embodiment. This is omitted in the machining report creating apparatus according to the third embodiment.

Next, each operation of the machine learning device 80 and the improvement proposal unit 90 will be described separately for a learning phase and a utilization phase.

<Learning phase>
FIG. 14 is a diagram showing the configuration of the machine learning device 80 according to the third embodiment. The machine learning device 80 includes a learning data acquisition unit 81, a model generation unit 82, and a trained model storage unit 83.

The learning data acquisition unit 81 acquires a processing report with countermeasures from the processing report storage unit 70. The model generation unit 82 extracts machining locus information, machining record information, abnormality peripheral information, and countermeasure information from the machining report with countermeasures acquired by the learning data acquisition unit 81, and extracts machining trajectory information, machining record information, and abnormality peripherals. The countermeasure information is learned based on the learning data created based on the combination of the information and the countermeasure information. That is, a learned model that infers the optimum countermeasure information for the machining defect caused by the abnormality generated during machining is generated from the machining locus information, the machining record information, the abnormality peripheral information, and the countermeasure information. Here, the learning data is data in which processing locus information, processing record information, abnormality peripheral information, and countermeasure information are associated with each other. In addition, other information, for example, processing specification information or information such as processing conditions may be extracted from the processing report with countermeasures and added to the learning data.

The machine learning device 80 is used to learn the countermeasure information of the machining report with countermeasures stored in the machining report storage unit 70. For example, the machine learning device 80 is connected to the machining report creation device 102 via a network, and the machine learning device 80 is connected to the machining report creation device 102 via a network. It may be a device separate from the machining report creating device 102. Further, the machine learning device 80 may be built in the machining report creating device 102. Further, the machine learning device 80 may exist on the cloud server.

As the learning algorithm used by the model generation unit 82, known algorithms such as supervised learning, unsupervised learning, and reinforcement learning can be used. As an example, a case where a neural network is applied will be described. The model generation unit 82 learns countermeasure information by, for example, supervised learning according to a neural network model. Here, supervised learning refers to a method of learning a feature in the learning data by giving a set of input and result (label) data to a machine learning device and inferring the result from the input.

A neural network is composed of an input layer composed of a plurality of neurons, an intermediate layer (hidden layer) composed of a plurality of neurons, and an output layer composed of a plurality of neurons. The intermediate layer may be one layer or two or more layers.

For example, in the case of a three-layer neural network as shown in FIG. 15, when a plurality of inputs are input to the input layer (X1-X3), the value is multiplied by the weight W1 (w11-w16) to form an intermediate layer (w11-w16). It is input to Y1-Y2), and the result is further multiplied by the weight W2 (w21-w26) and output from the output layer (Z1-Z3). This output result depends on the values of the weights W1 and W2.

In the present application, the neural network is based on a combination of machining locus information, machining record information, machining specification information, abnormal peripheral information, and countermeasure information extracted from a machining report with countermeasures acquired by the learning data acquisition unit 81. According to the created learning data, countermeasure information is learned by so-called supervised learning.

That is, the neural network inputs the machining locus information, the machining record information, the machining specification information, and the abnormality peripheral information into the input layer, and the result output from the output layer is weighted so as to approach the actually implemented countermeasure information. Learn by adjusting W1 and W2.

The model generation unit 82 generates and outputs a trained model by executing the above learning.

The trained model storage unit 83 stores the trained model output from the model generation unit 82.

Next, the process of learning by the machine learning device 80 will be described with reference to FIG. FIG. 16 is a flowchart relating to the learning process of the machine learning device.

In step b1, the learning data acquisition unit 81 acquires a processing report with countermeasures.

In step b2, the model generation unit 82 extracts the machining locus information, the machining record information, the abnormality peripheral information, and the countermeasure information from the machining report with countermeasures acquired by the learning data acquisition unit 81, and extracts the machining trajectory information and the machining results. According to the learning data created based on the combination of information, abnormal peripheral information, and countermeasures, countermeasure information is learned by so-called supervised learning, and a trained model is generated.

In step b3, the trained model storage unit 83 stores the trained model generated by the model generation unit 82.
<Utilization phase>

FIG. 17 is a configuration diagram of an improvement proposal unit 90 relating to the machining report creating device 102. The improvement proposal unit 90 includes a data acquisition unit 91, an inference unit 92, and a processing report creation unit 93 with a countermeasure plan.

The data acquisition unit 91 acquires a processing report newly created by the processing report creation unit 30.

The inference unit 92 infers the countermeasure information obtained by using the trained model. That is, by inputting the machining locus information, the machining record information, and the abnormality peripheral information extracted from the machining report acquired by the data acquisition unit 91 into this trained model, the countermeasure information inferred from the machining report is output. Can be done. In other words, the inference unit 92 infers countermeasure information from the processing report.

In this embodiment, the countermeasure information is output using the trained model learned by the model generation unit 82 of the machining report creation device 102, but it has been learned from the outside of another machining report creation device or the like. A model may be acquired and countermeasure information may be output based on this learned model.

The processing report creation unit 93 with a countermeasure plan acquires countermeasure information from the inference unit 92, inputs the countermeasure information into the processing report acquired from the data acquisition unit 91 as a countermeasure plan, and creates a processing report with a countermeasure plan. Then, the processing report creating unit 93 with a countermeasure plan outputs the created processing report with a countermeasure plan to the display unit 40. FIG. 18 shows a display example of a machining report with a countermeasure plan. FIG. 18 shows an example in which countermeasures for disconnection A, short-circuit B, and short-circuit C are presented, but not countermeasures for each abnormality, but comprehensive countermeasures that are effective for the entire abnormality. It may be presented. The second machining report corresponding to FIG. 18 is the same as that of FIG. 5, and is not shown. The worker can consider the countermeasure contents for the machining defect caused by the abnormality shown in the new machining report by referring to the countermeasure plan displayed in the machining report with the countermeasure plan. ..

Next, with reference to FIG. 19, a process for obtaining a machining report with a countermeasure plan by using the improvement proposal unit 90 will be described.

In step c1, the data acquisition unit 91 acquires the processing report newly created by the processing report creation unit 30.

In step c2, the inference unit 92 extracts the processing locus information, the processing record information, and the abnormality peripheral information from the processing report, and applies the extracted information to the trained model acquired from the trained model storage unit 83. Input the processing record information and the abnormality peripheral information to obtain the countermeasure information.

In step c3, the inference unit 92 outputs the countermeasure information obtained by the trained model to the processing report creation unit 93 with a countermeasure plan.

In step c4, the processing report creation unit 93 with a countermeasure plan creates a processing report with a countermeasure plan in which the countermeasure information output by the inference unit 92 is input as a countermeasure plan to the processing report acquired from the data acquisition unit 91.

In the present embodiment, the case where supervised learning is applied to the learning algorithm used by the model generation unit 82 has been described, but the present invention is not limited to this. As for the learning algorithm, it is also possible to apply reinforcement learning, unsupervised learning, semi-supervised learning, or the like, in addition to supervised learning.

Further, the model generation unit 82 may learn the countermeasure information according to the learning data created for the plurality of machining report creation devices. The model generation unit 82 may acquire learning data from a plurality of machining report creation devices used in the same area, or may be collected from a plurality of machining report creation devices operating independently in different areas. You may learn the countermeasure information by using the learning data. It is also possible to add a processing report creation device that collects learning data to the target on the way or remove it from the target. Furthermore, the machine learning device that has learned the countermeasure information for one machining report creation device is applied to another machining report creation device, and the countermeasure information is relearned and updated for the other machining report creation device. You may.

Further, as the learning algorithm used in the model generation unit 82, deep learning, which learns the extraction of the feature amount itself, can also be used, and other known methods such as genetic programming, functional logic programming, etc. Machine learning may be performed according to a support vector machine or the like.

According to the third embodiment, a countermeasure plan for a machining defect caused by an abnormality generated during machining is presented by using the result learned by a machine learning device using a machining report with countermeasures accumulated in the past. Therefore, the operator can refer to the countermeasure plan and consider the countermeasure contents for the machining defect caused by the abnormality shown in the new machining report.

Embodiment 4.
FIG. 20 is a diagram showing a configuration of a wire electric discharge machine according to a fourth embodiment. The wire electric discharge machine 201 according to the fourth embodiment includes a processing unit 210 and a control device 240 that controls the entire wire electric discharge machine 201. The control device 240 includes a machining control unit 231, a storage unit 232, and a machining report creating device 103. The machining report creating device 103 is the machining report creating device according to any one of the first to third embodiments. In the fourth embodiment, the same components as those in the first to third embodiments are designated by the same reference numerals, and the configurations different from those in the first embodiment will be mainly described.

According to the fourth embodiment, the wire electric discharge machine 201 is implemented without using an external device of the wire electric discharge machine 201 because the wire electric discharge machine 201 is provided with the processing report creating device 103. A machining report can be created in the same manner as in the first to third embodiments.

The configuration shown in the above-described embodiment shows an example of the contents of the present disclosure, can be combined with another known technique, and has a configuration within a range that does not deviate from the gist of the present disclosure. It is also possible to omit or change a part.

10 Machining information acquisition unit, 20 Machining information storage unit, 30 Machining report creation unit, 40 Display unit, 50 Setting unit, 60 Storage unit, 61 Extraction range designation information storage unit, 62 Similar judgment condition storage unit, 70 Machining report storage unit , 71 Countermeasure information input unit, 72 Similar processing report extraction unit, 80 Machine learning device, 81 Learning data acquisition unit, 82 Model generation unit, 83 Learned model storage unit, 90 Improvement proposal unit, 91 Data acquisition unit, 92 Inference Department, 93 Processing report creation unit with countermeasure plan, 100,101,102,103 Processing report creation device, 200,201 Wire discharge processing machine, 210 Processing unit, 211 Drive device, 212 Work piece, 213 Power supply, 214 Wire bobbin, 215 Feed roller, 216 Lower roller, 217 Recovery roller, 218 Wire electrode, 219 Power supply, 220 die, 221 table, 230,240 controller, 231 machining control unit, 232 storage section, 233 machining program, 234 machining conditions.

Claims (11)

It is a machining report creation device that creates a machining report that summarizes information on the machining results of electric discharge machining performed by a wire electric discharge machine for each machining unit.
Processing record information that includes abnormality information that is information about an abnormality that occurred during the machining of the electric discharge machining and that indicates the results of the machining results, and machining state that is information that indicates the machining state during machining of the electric discharge machining. A machining information acquisition unit that acquires machining information, which is information about the electric discharge machining, including information, from the wire electric discharge machine.
A processing information storage unit that stores the processing information acquired by the processing information acquisition unit for each processing unit, and a processing information storage unit.
Based on the processing information stored in the processing information storage unit, processing locus information which is information obtained by adding information indicating the occurrence position of the abnormality and the type of the abnormality on the processing locus and the processing record information. When the processing reporting unit to create a machining report with, said machining status information and is abnormal peripheral information section up after occurrence before occurrence of the abnormality,
Processing report creation device equipped with. It is a machining report creation device that creates a machining report that summarizes information on the machining results of electric discharge machining performed by a wire electric discharge machine for each machining unit.
Processing record information that includes abnormality information that is information about an abnormality that occurred during the machining of the electric discharge machining and that indicates the results of the machining results, and machining state that is information that indicates the machining state during machining of the electric discharge machining. A machining information acquisition unit that acquires machining information, which is information about the electric discharge machining, including information, from the wire electric discharge machine.
A processing information storage unit that stores the processing information acquired by the processing information acquisition unit for each processing unit, and a processing information storage unit.
Based on the processing information stored in the processing information storage unit, processing locus information which is information obtained by adding information indicating the occurrence position of the abnormality and the type of the abnormality on the processing locus and the processing record information. And a machining report creation unit that creates a machining report having the abnormality peripheral information that is the machining state information before and after the occurrence of the abnormality.
An extraction range designation information storage unit that stores extraction range designation information that specifies an extraction range for extracting the abnormal peripheral information from the processing information stored in the processing information storage unit, and an extraction range designation information storage unit.
With
The processing report creation unit
A machining report creating apparatus characterized in that the abnormal peripheral information is extracted from the machining information stored in the machining information storage unit based on the extraction range designation information and the machining report is created. The extraction range designation information is
The feature is that it is either the distance from the abnormality occurrence position, the number of blocks from the block corresponding to the abnormality occurrence position among the blocks constituting the machining program, or the time from the abnormality occurrence time. The processing report creating apparatus according to claim 2. A machining report storage section that stores the machining report created by the machining report creation section, and a machining report storage section.
A countermeasure information input unit for inputting countermeasure information, which is a countermeasure implemented for the abnormality, into the machining report stored in the machining report storage unit.
When the machining report creation unit creates a new machining report, the new machining report is selected from the machining reports with countermeasures in which the countermeasure information stored in the machining report storage section is input. Similar machining report extraction unit that extracts similar machining reports as similar machining reports,
The machining report creating apparatus according to any one of claims 1 to 3, further comprising. A similarity judgment condition storage unit that stores similar judgment conditions, which are conditions for judging that the new machining report and the machining report with countermeasures are similar,
Further prepare
The similar processing report extraction unit
The machining report creating apparatus according to claim 4 , wherein the similar machining report is extracted from the machining report with countermeasures based on the similarity determination condition. The similar judgment condition is
It is a condition specified by the processing specification information, the processing condition, the processing program, or the processing locus information.
The similar processing report extraction unit
Comparing the new machining report and the machining report with countermeasures, if at least one of the machining specification information, the machining conditions, the machining program, or the machining locus information is the same, they are similar. The processing report creating apparatus according to claim 5, wherein the processing report is determined. An improvement proposal unit that outputs a machining report with a countermeasure plan in which a countermeasure plan for the abnormality is input to the machining report, and
Further prepare
The improvement proposal department
A data acquisition unit that acquires the processing report from the processing report creation unit,
The machining input from the data acquisition unit using a trained model for inferring the countermeasure information generated by machine learning based on the machining report with countermeasures which is a machining report in which the countermeasure information is input. The inference unit that infers the countermeasure information from the report,
A processing report creation unit with a countermeasure plan for creating a processing report with a countermeasure plan, in which the countermeasure information inferred by the inference unit is input as the countermeasure plan in the processing report input from the data acquisition unit, and a processing report creation unit with a countermeasure plan.
The machining report creating apparatus according to any one of claims 4 to 6, wherein the machining report is provided. A machining report storage section that stores the machining report created by the machining report creation section, and a machining report storage section.
A countermeasure information input unit for inputting countermeasure information, which is a countermeasure implemented for the abnormality, into the machining report stored in the machining report storage unit.
An improvement proposal unit that outputs a machining report with a countermeasure plan in which a countermeasure plan for the abnormality is input to the machining report, and
Further prepare
The improvement proposal department
A data acquisition unit that acquires the processing report from the processing report creation unit,
The machining input from the data acquisition unit using a trained model for inferring the countermeasure information generated by machine learning based on the machining report with countermeasures which is a machining report in which the countermeasure information is input. The inference unit that infers the countermeasure information from the report,
A processing report creation unit with a countermeasure plan for creating a processing report with a countermeasure plan, in which the countermeasure information inferred by the inference unit is input as the countermeasure plan in the processing report input from the data acquisition unit, and a processing report creation unit with a countermeasure plan.
The machining report creating apparatus according to any one of claims 1 to 3, wherein the machining report is provided. A learning data acquisition unit that acquires the processing report with countermeasures from the processing report storage unit, and
From the machining report with countermeasures acquired by the learning data acquisition unit, the machining locus information, the machining record information, the abnormality peripheral information, and the countermeasure information are extracted, and the machining locus information, the machining record information, and the above. A model generation unit that generates the trained model by machine learning according to the training data created based on the combination of the anomaly peripheral information and the countermeasure information.
Machining reporting device according to claim 7 or claim 8, further comprising a. The processing report creation unit
The machining report creating apparatus according to any one of claims 1 to 9 , wherein the machining report is created by displaying the abnormality peripheral information in a graph format in a time series. A wire electric discharge machine comprising the machining report creating apparatus according to any one of claims 1 to 10.

Images