Control device for wire electrical discharge machining machine, control program product for wire electrical discharge machining machine, and control method for wire electrical discharge machining machine
By detecting the short circuit in the online discharge machining machine, the problem of reducing machining accuracy caused by short circuit is solved, and higher machining accuracy and fewer short circuits occur.
Patent Information
- Application Number
- CN202280096793.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-15
- Publication Date
- 2025-06-27
- Estimated Expiration
- 2042-11-15
AI Technical Summary
When a short circuit occurs in existing wire discharge machining machines, the processing accuracy is easily reduced, and the processing chips cannot be fully removed, resulting in repeated short circuits.
When a short circuit is detected, power supply to the line electrode is stopped, processing is interrupted, and the line electrode is moved at a predetermined retreat distance. After the predetermined retreat time has passed, close to the processing surface and resume processing.
It effectively suppresses the reduction of machining accuracy in the case of short circuit, ensuring the integrity and accuracy of the machining surface.
Smart Images

Figure CN119317502B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a control device for a wire electric discharge machine that generates electric discharge between a wire electrode and a processing object to thereby remove the processing object, a control program product for the wire electric discharge machine, and a control method for the wire electric discharge machine. Background Art
[0002] The wire EDM generates discharge by generating a voltage between the wire electrode and the object to be processed, thereby removing the object to be processed. In the wire EDM, in order to obtain the desired shape and the desired surface roughness, the wire electrode is usually rotated multiple times along the trajectory of the desired shape to repeat the processing. During the repeated processing, the offset of the wire electrode relative to the desired shape is continuously changed at each repetition, and the input processing energy is continuously reduced, thereby gradually reducing the surface roughness. At this time, as the input processing energy decreases, the processing amount decreases, and therefore the distance between the wire electrode and the object to be processed is continuously shortened. If the wire electrode vibrates and processing chips are generated in this state, the wire electrode and the object to be processed may contact and cause a short circuit.
[0003] Patent Document 1 discloses a technique for retracting a wire electrode in order to release the short circuit when a short circuit is detected. In the technique disclosed in Patent Document 1, when a short circuit is detected, the wire electrode is retracted by a predetermined distance, and if the short circuit is not released at this time, it is determined that a short circuit has occurred in a direction perpendicular to the traveling direction of the wire electrode, and the retracting direction and retracting distance of the wire electrode are determined.
[0004] Patent Document 1: Japanese Patent Application Laid-Open No. 5-77109 Summary of the invention
[0005] However, according to the above-mentioned conventional technology, the time from when the wire electrode is retracted until the machining is resumed, i.e., the machining reset time, is not taken into consideration. If the machining reset time is short, machining chips may not be sufficiently removed, or the vibration of the wire electrode may not converge, resulting in repeated short circuits, thereby causing a problem of reduced machining accuracy.
[0006] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a control device for a wire electrical discharge machine that can suppress a decrease in machining accuracy even when a short circuit occurs.
[0007] In order to solve the above problems and achieve the object, a control device for a wire electrical discharge machining machine according to the present invention controls the wire electrical discharge machining machine, which applies a voltage between a wire electrode and a workpiece to generate a discharge, thereby performing a removal machining on the workpiece. The control device for the wire electrical discharge machining machine is characterized in that when the wire electrical discharge machining machine detects a short circuit based on at least one of the machining current and the machining voltage during machining, which is a monitored object value, the power supply to the wire electrode is stopped to interrupt the machining, and the wire electrode is moved by a predetermined retraction distance in the offset direction in which the distance between the wire electrode and the machining surface of the workpiece increases, thereby retracting the wire electrode. After a predetermined retraction time has elapsed after retracting the wire electrode, the wire electrode is again brought close to the machining surface to resume machining.
[0008] Effect of the Invention
[0009] The control device for the wire electrical discharge machining machine according to the present invention has the following effect, that is, even when a short circuit occurs, a reduction in machining accuracy can be suppressed. Description of the Drawings
[0010] Figure 1 It is a diagram showing the structure of a wire electrical discharge machining system according to Embodiment 1.
[0011] Figure 2 It is an explanatory diagram of the cause of machining streaks during finish machining.
[0012] Figure 3 It is an explanatory diagram of the retraction process and the re-machining process when a short circuit occurs in Embodiment 1.
[0013] Figure 4 It is a flowchart for explaining the machining control operation according to Embodiment 2.
[0014] Figure 5 It is a diagram showing the structure of a wire electrical discharge machining system according to Embodiment 3.
[0015] Figure 6 It is an explanatory diagram of the retraction process and the re-machining process when a short circuit occurs in Embodiment 3.
[0016] Figure 7 It is a flowchart for explaining the machining control operation according to Embodiment 3.
[0017] Figure 8 It is a diagram showing the first example of detecting a short circuit of a detection object in Embodiment 4.
[0018] Figure 9 It is a diagram showing the second example of detecting a short circuit of a detection object in Embodiment 4.
[0019] Figure 10 This is a diagram showing the retraction direction and re-approaching direction of the wire electrode in Embodiment 5.
[0020] Figure 11 This is a structural diagram of a learning device related to a wire electrical discharge machining machine.
[0021] Figure 12 This is a diagram showing an example of a neural network.
[0022] Figure 13 This is a flowchart related to the learning process of the learning device.
[0023] Figure 14 This is a structural diagram of an inference device related to a wire electrical discharge machining machine.
[0024] Figure 15 This is a flowchart related to the inference process of the inference device.
[0025] Figure 16 This is a diagram showing the first example of the hardware structure of a wire electrical discharge machining system.
[0026] Figure 17 This is a diagram showing the second example of the hardware structure of a wire electrical discharge machining system. Detailed Embodiments
[0027] Next, based on the accompanying drawings, a control device for a wire electrical discharge machining machine, a control program for a wire electrical discharge machining machine, and a control method for a wire electrical discharge machining machine according to the embodiments of the present invention will be described in detail.
[0028] Embodiment 1.
[0029] Figure 1 This is a diagram showing the structure of a wire electrical discharge machining system 100 according to Embodiment 1. The wire electrical discharge machining system 100 includes: a wire electrical discharge machining machine 1 that performs removal machining on a machining object 4; a control device 2 that controls the wire electrical discharge machining machine 1; a power supply device 3 that supplies power to the wire electrical discharge machining machine 1; and a detection unit 5 that detects the state of the wire electrical discharge machining machine 1.
[0030] The wire electrical discharge machining machine 1 has a function of performing removal machining on the machining object 4. The wire electrical discharge machining machine 1 includes: a wire electrode 11; a position adjustment unit 12 that adjusts the relative position between the wire electrode 11 and the machining object 4; and a power supply unit 13 that supplies machining energy by supplying power from the power supply device 3 to the wire electrode 11.
[0031] The wire discharge machining machine 1 repeatedly performs multiple processes until a processed shape corresponding to a desired shape is obtained. The first process performed among the multiple processes is called rough processing, which is a process that places more emphasis on processing speed than on shape accuracy. The second and subsequent processes among the multiple processes are called finishing processing. Finishing processing is performed more than or equal to one time, and when it is repeated multiple times, finishing processing is performed in a manner that gradually improves the shape accuracy.
[0032] The control device 2 gives a position command indicating the relative position between the wire electrode 11 and the object 4 to the position adjustment unit 12 of the wire electrical discharge machine 1 and gives a power supply command to the power supply device 3 , thereby controlling the removal process of the object 4 by the wire electrical discharge machine 1 .
[0033] The detection unit 5 detects at least one of the machining current and the machining voltage, that is, a monitoring target value, and outputs the detected monitoring target value to the control device 2 .
[0034] The control device 2 can detect a short circuit between the wire electrode 11 and the object 4 based on the monitoring target value output by the detection unit 5. When a short circuit is detected, the control device 2 sequentially performs a retraction process of the wire electrode 11 and a re-machining process of resuming machining after the retraction process.
[0035] Here, a description will be given of machining defects that occur when a short circuit occurs during finish machining. Figure 2 This is an explanatory diagram of the cause of machining streaks in finishing. Finishing requires less machining energy than roughing to make the surface roughness finer, so the wire electrode 11 is closer to the machining surface than roughing. Figure 2 In the example shown, finishing machining is performed while the wire electrode 11 is moved in the machining direction D1. In this state, when machining chips 41 are generated and contact the wire electrode 11, the machining chips 41 may contact the machining object 4 because the distance between the wire electrode 11 and the machining surface of the machining object 4 is close. In this case, a short circuit occurs between the wire electrode 11 and the machining object 4. If machining is continued in this state and the machining chips 41 are removed, the short circuit can be resolved, but since no discharge occurs in the short-circuited portion, the machining object 4 is not cut, and the cut remaining portion of the machining object 4 becomes convex compared to the normally machined surface, and is recognized as a machining streak 42.
[0036] Figure 3 1 is an explanatory diagram of the avoidance process and reprocessing process when a short circuit occurs in the first embodiment. Figure 3In the example shown, machining is performed while moving the wire electrode 11 in the machining direction D1. Here, if the machining chips 41 come into contact with the wire electrode 11 and a short circuit occurs, the control device 2 detects the short circuit based on at least one of the machining current and the machining voltage, i.e., the monitored value. If a short circuit is detected, the control device 2 stops supplying power to the wire electrode 11 to interrupt machining, and moves the wire electrode 11 in the retraction direction D2. The retraction direction D2 is a deviation direction in which the distance of the wire electrode 11 from the machining surface of the workpiece 4 increases, and is preferably set to a direction obliquely rearward with respect to the machining direction D1 so that the distance from the inclined surface 45 connecting the previous machining surface 43 and the machining surface 44 being machined also increases. In addition, the retraction path is set so as not to contact the machining surface during retraction. The control device 2 moves the wire electrode 11 by a predetermined retraction distance. Thereby, the short circuit can be eliminated.
[0037] If the short circuit can be eliminated, the control device 2 resumes machining of the workpiece 4. At this time, the control device 2 retracts the wire electrode 11, and then after a predetermined retraction time has elapsed, brings the wire electrode 11 close to the machining surface. The re-approaching direction D3 at this time is, for example, the opposite direction of the retraction direction D2. The control device 2 resumes supplying power to the wire electrode 11 until the wire electrode 11 approaches the workpiece 4 to a distance at which discharge occurs for the applied machining voltage and machining current. Thereby, it is possible to suppress re-discharge on the already machined machining surface 44 and suppress over-machining.
[0038] Embodiment 2.
[0039] Figure 4 It is a flowchart for explaining the machining control operation according to Embodiment 2. After machining starts, the control device 2 acquires at least one of the machining current and the machining voltage as the monitored value (step S101). The control device 2 determines whether a short circuit is detected based on the acquired monitored value (step S102). Here, the control device 2 can determine that a short circuit has occurred when the change amount of the acquired monitored value exceeds a predetermined threshold value.
[0040] When no short circuit is detected (step S102: No), the control device 2 returns to the process of step S101. When a short circuit is detected (step S102: Yes), the control device 2 stops supplying power to the wire electrode 11 (step S103) and interrupts machining.
[0041] The control device 2 determines the retraction direction D2 based on the state of the wire electrical discharge machining machine 1 at the moment when machining is interrupted (step S104). The control device 2 moves the wire electrode 11 by a predetermined retraction distance in the retraction direction D2 (step S105).
[0042] Here, the control device 2 estimates the length in the machining direction D1 of the machining stripe 42 generated by the short circuit, taking into account the control delay from the occurrence of the short circuit until the wire electrode 11 is retracted, and estimates the start point of the machining stripe 42 (step S106). The control device 2 uses the estimated start point of the machining stripe 42 as the start point of re-machining, and determines a re-approach trajectory for approaching from the position of the retracted wire electrode 11 to the start point of the machining stripe 42 (step S107).
[0043] After starting the retraction of the wire electrode 11, the control device 2 determines whether a retraction time of a pre-determined length has elapsed (step S108). If the retraction time has not elapsed (step S108: No), the control device 2 returns to the process of step S108. If the retraction time has elapsed (step S108: Yes), the control device 2 moves the wire electrode 11 along the re-approach trajectory (step S109). After moving the wire electrode 11, the control device 2 resumes power supply (step S110) and resumes machining.
[0044] Embodiment 3.
[0045] Figure 5 FIG. is a diagram showing the structure of the wire electrical discharge machining system 100-1 according to Embodiment 3. The wire electrical discharge machining system 100-1 has a wire electrical discharge machining machine 1-1 instead of the wire electrical discharge machining machine 1 of the wire electrical discharge machining system 100, and has a control device 2-1 instead of the control device 2 of the wire electrical discharge machining system 100.
[0046] The wire electrical discharge machining machine 1-1 has a wire electrode 11, a position adjustment unit 12, a power supply unit 13, a machining fluid supply unit 14 that supplies machining fluid between the wire electrode 11 and the workpiece 4, and a wire feed speed adjustment unit 15 that adjusts the feed speed of the wire electrode 11. The machining fluid is a liquid supplied between the wire electrode 11 and the workpiece 4 during machining by the wire electrical discharge machining machine 1-1, and is supplied for the purpose of discharging the machining chips 41 or cooling the heated wire electrode 11 during machining.
[0047] On the basis of the functions of the control device 2, the control device 2-1 further has the following function: when a short circuit of the wire discharge machining machine 1-1 is detected, in order to improve the discharge capacity of the machining chips 41, the flow rate of the machining fluid and the feeding speed of the wire electrode 11 are adjusted. Specifically, when the control device 2-1 detects a short circuit, when retracting the wire electrode 11, it instructs the machining fluid supply unit 14 to increase the flow rate of the machining fluid compared with that before adjustment, and instructs the wire feeding speed adjustment unit 15 to increase the feeding speed of the wire electrode 11 compared with that before adjustment. In addition, in the third embodiment, when the control device 2-1 detects a short circuit, it adjusts both the flow rate of the machining fluid and the feeding speed of the wire electrode 11, but the control device 2-1 may also control at least one of the flow rate of the machining fluid and the feeding speed of the wire electrode 11. In addition, when retracting the wire electrode 11, the control device 2-1 may discharge the machining chips 41 after increasing the flow rate of the machining fluid and the feeding speed of the wire electrode 11, and then restore the flow rate of the machining fluid and the feeding speed of the wire electrode 11 to the original state, or may still use the adjusted flow rate and feeding speed after resuming machining.
[0048] Figure 6 FIG. is an explanatory diagram of the retraction process and the reprocessing process when a short circuit occurs in the third embodiment. If discharge occurs between the wire electrode 11 and the workpiece 4 while moving the wire electrode 11 in the machining direction D1, and thus the workpiece 4 is removed, the removed workpiece 4 may become machining chips 41 and adhere to the machining surface. In finish machining, since the distance between the wire electrode 11 and the machining surface is short, it may sometimes contact the machining chips 41 and a short circuit may occur.
[0049] Here, sometimes a control delay DW occurs until the wire electrode 11 is retracted after a short circuit occurs. In addition, when the machining chips 41 are continuously generated, sometimes the machining streak 42 has a certain length l. Here, the height of the machining streak 42 is set to h.
[0050] If the control device 2-1 detects a short circuit, it estimates the amount of the control delay DW and the length l of the machining streak 42, and estimates the start point Ps of the machining streak 42 based on the control delay DW and the length l of the machining streak 42. After the control device 2-1 interrupts the machining by retracting the wire electrode 11, when resuming machining, it brings the wire electrode 11 close to the estimated start point Ps.
[0051] In addition, in the case where a control delay DW has occurred, since the section of the control delay DW has already been machined, if machining is performed again, overcutting will occur. Therefore, the control device 2-1 estimates a reprocessing section 51, which is a section where machining streaks 42 have occurred and reprocessing is required, and a non-reprocessing section 52, which is a section that has already been machined and does not require reprocessing, based on the control delay DW, the length l of the machining streaks 42, the starting point Ps of the machining streaks 42, and the diameter of the wire electrode 11. After resuming machining, power is supplied to the wire electrode 11 between the starting point Ps of the machining streaks 42 and the reprocessing section 51 for machining. When the reprocessing section 51 has been passed through, the power supply to the wire electrode 11 is stopped to cut off the machining current, and the machining current is kept cut off until the wire electrode 11 has moved through the non-reprocessing section 52. Then, if the wire electrode 11 has passed through the non-reprocessing section 52, the control device 2-1 resumes power supply to the wire electrode 11 to resume machining.
[0052] Figure 7 is a flowchart for explaining the machining control operation according to Embodiment 3. In addition, for Figure 7 the same parts as those in the Figure 4 shown processing are labeled with the same reference numerals, and the description thereof is omitted.
[0053] Regarding the processing of steps S101 to S104, it is the same as Figure 4 . Before starting the retraction, the control device 2-1 changes the flow rate of the machining fluid and the feeding speed of the wire electrode 11 (step S120). At this time, the control device 2-1 increases the flow rate of the machining fluid and raises the feeding speed of the wire electrode 11 compared with before the change, thereby improving the discharge capacity of the machining chips 41.
[0054] After the control device 2-1 performs the same processing as Figure 4 in step S105, instead of Figure 4 step S106, it estimates the starting point Ps of the machining streaks 42 and the length l of the machining streaks 42 (step S121). The control device 2-1 determines a re-approaching trajectory in such a way that the wire electrode 11 approaches the starting point Ps of the machining streaks 42 based on the estimated starting point Ps and the length l of the machining streaks 42 (step S122).
[0055] Regarding the processing of steps S108 to S110, it is the same as Figure 4 . In addition, regarding the operation after power supply is resumed through step S110, as Figure 6 explained, the control device 2-1 controls to perform machining in the reprocessing section 51 and not to perform machining in the non-reprocessing section 52.
[0056] Embodiment 4.
[0057] Figure 8 This is a diagram showing the first example of a short circuit in the object to be detected in Embodiment 4. Figure 9 This is a diagram showing the second example of a short circuit in the object to be detected in Embodiment 4. In Figure 8 and Figure 9 , the horizontal axis is time or a position on the workpiece 4, and the vertical axis is at least one of the machining current and the machining voltage. Figure 8 It includes a normal machining waveform 61 and an instantaneous short-circuit waveform 62-1. Figure 9 It includes a normal machining waveform 61 and a continuous short-circuit waveform 62-2.
[0058] As Figure 8 shown, when a short circuit occurs instantaneously, the instantaneous short-circuit waveform 62-1 exhibits a behavior of at least one of an instantaneous decrease and an instantaneous increase in at least one of the machining current and the machining voltage with respect to the normal machining waveform 61.
[0059] As Figure 9 shown, when a short circuit occurs continuously, the continuous short-circuit waveform 62-2 exhibits a behavior of at least one of a continuous decrease and a continuous increase in at least one of the machining current and the machining voltage with respect to the normal machining waveform 61. In Embodiment 4, both an instantaneous short circuit and a continuous short circuit are determined as short circuits. Therefore, when the change amount of the monitored value is greater than or equal to the threshold value, the control device 2 determines that it is a short circuit regardless of whether the change is an instantaneous change as Figure 8 shown, or a continuous change as Figure 9 shown.
[0060] In addition, the control device 2 can estimate the height h of the machining stripe 42 based on the behavior of at least one of the machining current and the machining voltage during a short circuit, and change the machining conditions during reprocessing based on the estimated height h of the machining stripe 42.
[0061] Embodiment 5.
[0062] Figure 10This is a diagram showing the retraction direction D2 and the re-approaching direction D3 of the wire electrode 11 in Embodiment 5. The retraction direction D2 of the wire electrode 11 is an offset direction in which the distance from the workpiece 4 increases. The retraction trajectory of the wire electrode 11 is determined based on the machining trajectory so that the wire electrode 11 does not contact the machining surface. The re-approaching direction D3 of the wire electrode 11 has a vector component parallel to the machining direction D1, and the wire electrode 11 re-approaches the machining surface obliquely from the rear of the machining direction D1. When the wire electrode 11 is short-circuited, the wire electrode 11 moves in the machining direction D1. Therefore, behind the machining direction D1, the wire electrode 11 forms a bent shape. When resuming machining, the wire electrode 11 is closest to the machining surface in the bent shape at the time of short circuit. Therefore, the wire electrode 11 approaches the machining surface closest from the rear of the machining direction D1, thereby being able to suppress the situation where the wire electrode 11 contacts the machining surface.
[0063] Embodiment 6.
[0064] In Embodiment 6, an example of learning information used to control the wire electrical discharge machining machine 1 using machine learning is described.
[0065] <Learning stage>
[0066] Figure 11 This is a structural diagram of the learning device 70 related to the wire electrical discharge machining machine 1. Hereinafter, in Embodiment 6, it is assumed that learning related to the wire electrical discharge machining machine 1 is performed for explanation, but learning related to the wire electrical discharge machining machine 1-1 may also be performed.
[0067] The learning device 70 includes a learning data acquisition unit 71 and a model generation unit 72. The learning data acquisition unit 71 acquires learning data, which includes condition information indicating the conditions of machining performed by the wire electrical discharge machining machine 1, and information used to re-machine the cutting residue portion, i.e., the machining streak 42, of the workpiece 4 that occurs when a short circuit occurs in the wire electrical discharge machining machine 1.
[0068] Here, the condition information is the specifications of the wire electrical discharge machining machine 1, the elements of the workpiece 4, the settings during machining of the wire electrical discharge machining machine 1, etc., and includes, for example, at least one of the plate thickness of the workpiece 4, the electrode diameter of the wire electrode 11, the machining trajectory, the machining conditions, the machining current, and the machining voltage. The "information used to re-machine the machining streak 42" acquired by the learning data acquisition unit 71 includes, for example, at least one of the control delay DW of the wire electrical discharge machining machine 1, the length l of the machining streak 42, the retraction trajectory, the re-approaching trajectory, and the height h of the machining streak 42.
[0069] The model generation unit 72 generates a trained model that infers "information used for reprocessing the machining stripe 42" based on the combination of the condition information output from the learning data acquisition unit 71 and the information used for reprocessing the machining stripe 42. Here, the learning data is data that correlates the condition information and "information used for reprocessing the machining stripe 42" with each other.
[0070] The information used for reprocessing the machining stripe 42 inferred by the trained model generated by the model generation unit 72 includes, for example, at least one of the start point Ps of the machining stripe 42, the reprocessing interval 51, the non-reprocessing interval 52, the machining conditions during reprocessing, the retraction trajectory, and the re-approach trajectory.
[0071] The learning algorithm used by the model generation unit 72 can use known algorithms such as supervised learning, unsupervised learning, and reinforcement learning. As an example, the case of applying a neural network will be described.
[0072] The model generation unit 72 learns the information used for reprocessing the machining stripe 42, for example, according to a neural network model, through so-called supervised learning. Here, supervised learning is called the following method, that is, by giving a data group of input and the result as a label to the learning device 70, the features existing in these learning data are learned, and the result is inferred based on the input.
[0073] A neural network is composed of an input layer composed of multiple neurons, an intermediate layer composed of multiple neurons, and an output layer composed of multiple neurons. The intermediate layer is also called a hidden layer. The intermediate layer can be one layer or two or more layers.
[0074] For example, if it is Figure 12 the three-layer neural network shown, multiple inputs are input to the input layer (X1 - X3), their values are multiplied by the weights W1 (w11 - w16) and input to the intermediate layer (Y1 - Y2), and the result is further multiplied by the weights W2 (w21 - w26) and output from the output layer (Z1 - Z3). This output result changes according to the values of the weights W1 and the weights W2. Figure 12 is a diagram showing an example of a neural network.
[0075] In the present application, the neural network learns the information used for reprocessing the machining stripe 42 through so-called supervised learning according to the learning data created based on the combination of the condition information obtained by the learning data acquisition unit 71 and the information used for reprocessing the machining stripe 42.
[0076] That is, the neural network adjusts the weights W1 and W2 to learn so that the result output from the output layer is close to the "information used for reprocessing the machining stripes 42" when the input layer inputs the conditional information.
[0077] By performing the above learning, the model generation unit 72 generates a trained model and outputs it.
[0078] The trained model storage unit 73 stores the trained model output from the model generation unit 72.
[0079] Next, use Figure 13 to explain the process learned by the learning device 70. Figure 13 It is a flowchart related to the learning process of the learning device 70.
[0080] The learning data acquisition unit 71 acquires the conditional information and the information used for reprocessing the machining stripes 42 as learning data (step S201). In addition, the learning data acquisition unit 71 acquires the conditional information and the information used for reprocessing the machining stripes 42 simultaneously, but it is sufficient to input the conditional information and the information used for reprocessing the machining stripes 42 in association with each other, and the conditional information and the information used for reprocessing the machining stripes 42 can be acquired at different timings. For example, in the case where the conditional information includes multiple pieces of information, each piece of information included in the conditional information can be acquired at different timings, and in the case where the information used for reprocessing the machining stripes 42 includes multiple pieces of information, each piece of information included in the information used for reprocessing the machining stripes 42 can be acquired at different timings.
[0081] The model generation unit 72 performs a learning process, that is, according to the learning data created based on the combination of the conditional information acquired by the learning data acquisition unit 71 and the information used for reprocessing the machining stripes 42, through so-called supervised learning, it learns the information used for reprocessing the machining stripes 42 and generates a trained model (step S202).
[0082] The trained model storage unit 73 stores the trained model generated by the model generation unit 72 (step S203).
[0083] <Effective usage stage>
[0084] Figure 14 It is a structural diagram of the inference device 74 related to the wire electrical discharge machining machine 1. The inference device 74 has an inference data acquisition unit 75 and an inference unit 76.
[0085] The inference data acquisition unit 75 acquires condition information indicating the conditions of machining performed by the wire electrical discharge machining machine 1 as inference data.
[0086] Specifically, the condition information acquired by the inference data acquisition unit 75 is the specifications of the wire electrical discharge machining machine 1, the elements of the workpiece 4 to be machined, the settings during machining of the wire electrical discharge machining machine 1, etc., and includes, for example, at least one of the thickness of the workpiece 4, the electrode diameter of the wire electrode 11, the machining trajectory, the machining conditions, the machining current, and the machining voltage.
[0087] The inference unit 76 infers "information used for reprocessing the cutting residue portion of the workpiece 4, i.e., the machining streaks 42, generated when a short circuit occurs in the wire electrical discharge machining machine 1" obtained using the trained model. Specifically, the information inferred by the inference unit 76 includes at least one of the start point Ps of the machining streaks 42, the reprocessing section 51, the non-reprocessing section 52, the machining conditions during reprocessing, the retraction trajectory, and the re-approach trajectory. The inference unit 76 inputs the condition information acquired by the inference data acquisition unit 75 into the trained model, and thereby can output the "information used for reprocessing the machining streaks 42" inferred based on the condition information.
[0088] In addition, in the present embodiment, it has been described that inference is performed using the trained model learned by the model generation unit 72 of the wire electrical discharge machining machine 1, but it is also possible to obtain a trained model from an external source such as another wire electrical discharge machining machine and perform inference based on the trained model.
[0089] Next, Figure 15 , the process for obtaining the "information used for reprocessing the machining streaks 42" using the inference device 74 will be described. Figure 15 is a flowchart related to the inference process of the inference device 74.
[0090] The inference data acquisition unit 75 acquires condition information as inference data (step S211). The inference unit 76 inputs the condition information into the trained model stored in the trained model storage unit 73 (step S212) to obtain the "information used for reprocessing the machining streaks 42".
[0091] The inference unit 76 performs data output processing of outputting the "information used for reprocessing the machining streak 42" obtained by the trained model to the control device 2 of the wire electrical discharge machining machine 1 (step S213). The control device 2 of the wire electrical discharge machining machine 1 uses the output "information used for reprocessing the machining streak 42" to move the wire electrode 11 closer to the machining surface again (step S214). For example, when the "information used for reprocessing the machining streak 42" is the start point Ps of the machining streak 42, the reprocessing section 51, the non-reprocessing section 52, the machining conditions during reprocessing, the retraction trajectory, and the re-approach trajectory, after the control device 2 retracts the wire electrode 11 by the retraction distance according to the retraction trajectory shown in the inference result, after the elapse of the retraction time, the control device 2 moves the wire electrode 11 to the start point Ps shown in the inference result according to the closest approach trajectory, performs machining between the start point Ps and the reprocessing section 51, and stops supplying power to the wire electrode 11 during the period of passing through the non-reprocessing section 52, thereby stopping the machining. Thus, the machining streak 42 can be removed.
[0092] In addition, in the present embodiment, the case where supervised learning is applied to the learning algorithm used in the model generation unit 72 has been described, but it is not limited thereto. Regarding the learning algorithm, in addition to supervised learning, reinforcement learning, unsupervised learning, or semi-supervised learning can also be applied.
[0093] In addition, the learning device 70 and the inference device 74 are used to learn the "information used for reprocessing the machining streak 42" of the wire electrical discharge machining machine 1, but for example, they can also be connected to the wire electrical discharge machining machine 1 via a network and be devices separate from the wire electrical discharge machining machine 1. In addition, the learning device 70 and the inference device 74 can also be built into the wire electrical discharge machining machine 1. Moreover, the learning device 70 and the inference device 74 can also exist on a cloud server.
[0094] In addition, the model generation unit 72 can learn "information used for reprocessing the machining streaks 42" according to the learning data created for multiple wire electrical discharge machines 1. Further, the model generation unit 72 can obtain the learning data from multiple wire electrical discharge machines 1 used in the same area, or can also use the learning data collected from multiple wire electrical discharge machines 1 operating independently in different areas to learn "information used for reprocessing the machining streaks 42". Additionally, it is also possible to add the wire electrical discharge machine 1 that collects the learning data to the object midway or remove it from the object. Moreover, the learning device after a certain wire electrical discharge machine 1 learns "information used for reprocessing the machining streaks 42" can be applied to other wire electrical discharge machines 1 different from it, and relearn "information used for reprocessing the machining streaks 42" about the other wire electrical discharge machines 1 for updating.
[0095] In addition, as the learning algorithm used in the model generation unit 72, deep learning that learns the extraction of the feature quantity itself can also be used, or machine learning can be executed according to other known methods, such as genetic programming, functional logic programming, support vector machines, etc.
[0096] Embodiment 7.
[0097] In Embodiment 7, an example of the hardware structure of the wire electrical discharge machining system is shown. Figure 16 It is a diagram showing the first example of the hardware structure of the wire electrical discharge machining system. The wire electrical discharge machining system includes a control device 2, a power supply device 3, and a second control device 6. The control device 2 is a numerical control device. The power supply device 3 can, for example, have Figure 1 the detection unit 5 shown. The second control device 6 includes a learning device 70, an inference device 74, and a trained model storage unit 73.
[0098] Regarding the wire electrode 11, the wire electrode 11 is supplied from the wire electrode spool 81, and the wire electrode 11 is conveyed to the machining unit 88 through the wire electrode conveying roller 83. Additionally, if power is supplied from the power supply device 3 to the upper power supply member 13a and the lower power supply member 13b, a machining current is supplied to the wire electrode 11. The position of the wire electrode 11 is held by the upper guide member 86 and the lower guide member 87. After the wire electrode 11 machines the workpiece 4 placed on the table T in the machining unit 88, it is recovered into the wire electrode recovery box 82 through the lower roller 84. The conveying speed of the wire electrode 11 is controlled by the wire traveling speed control motor 85. In Figure 16In the structure shown, position commands are input from the control device 2 to the X-axis motor 89 and the Y-axis motor 90 to change the positions on the X-axis and Y-axis of the stage T, thereby controlling the relative position of the wire electrode 11 with respect to the workpiece 4. In this case, the process of retracting the wire electrode 11 corresponds to the following process, that is, moving the stage T so that the relative distance between the wire electrode 11 and the workpiece 4 becomes the retraction distance, thereby moving the wire electrode 11 relative to the workpiece 4 by the retraction distance.
[0099] In addition, the functions of the control device 2 and the second control device 6 are implemented using a processing circuit. The processing circuit can be dedicated hardware or a control circuit having a processor such as a CPU (Central Processing Unit) and a memory. When the processing circuit is dedicated hardware, the processing circuit is, for example, a single circuit, a composite circuit, a programmed processor, a parallel-programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. In addition, when the above processing circuit is implemented by a control circuit using a CPU, the control circuit has a processor and a memory also referred to as a CPU, an arithmetic unit, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), etc. The memory is, for example, a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM), an EEPROM (registered trademark) (Electrically EPROM), a magnetic disk, a floppy disk, an optical disk, a compact disk, a minidisk, a DVD (Digital Versatile Disk), etc.
[0100] In the case where the above processing circuit is implemented by the control circuit, it is achieved by the processor reading and executing a control program stored in the memory corresponding to the processing of each structural element. In addition, the memory is also used as a temporary memory in each process executed by the processor. The control program is, for example, a computer program for causing the wire electrical discharge machining machine 1 to execute the following steps: when the wire electrical discharge machining machine 1 detects a short circuit based on at least one of the machining current and the machining voltage during machining, that is, the monitored value, the step of stopping the power supply to the wire electrode 11 and interrupting the machining; the step of retracting the wire electrode 11 by moving the wire electrode 11 in the offset direction in which the distance between the wire electrode 11 and the machining surface of the workpiece 4 increases by a predetermined retraction distance; and the step of resuming the machining by bringing the wire electrode 11 close to the machining surface again after a predetermined retraction time has elapsed after retracting the wire electrode 11. In addition, the control program can be provided in a state stored in a storage medium or can be provided via a communication path such as the Internet.
[0101] Figure 17 It is a diagram showing a second example of the hardware configuration of the wire electrical discharge machining system. Regarding Figure 17 , mainly the parts different from Figure 16 will be described. In the first example shown in Figure 16 , the relative position between the wire electrode 11 and the workpiece 4 is controlled by moving the table T on which the workpiece 4 is placed, but in the second example shown in Figure 17 , position commands are input from the control device 2 to the upper guide member 86 and the lower guide member 87, and the position of the wire electrode 11 is changed, thereby controlling the relative position between the wire electrode 11 and the workpiece 4.
[0102] In addition, use Figure 16 and Figure 17The two examples shown illustrate the hardware structure of the wire electrical discharge machining system, but the hardware structure of the wire electrical discharge machining system is not limited to the examples shown. For example, in the first and second examples above, the control device 2 is a numerical control device, and the second control device 6 is a device different from the numerical control device. However, the numerical control device may also have the functions of the second control device 6. Alternatively, in the first and second examples above, the second control device 6 has a learning device 70, an inference device 74, and a trained model storage unit 73. However, a part of these functions may also be provided in a device different from the second control device 6 and the control device 2. For example, as described above, the functions of the learning device 70 and the trained model storage unit 73 can be implemented on a cloud server. In addition, a part of the functions of the control device 2 may also be executed by a device different from the numerical control device. In addition, in the first and second examples above, the power supply device 3 has a detection unit 5. However, if the detection unit 5 can detect at least one of the machining current and the machining voltage, it may not be built in the power supply device 3.
[0103] The structures shown in the above embodiments represent an example, and can also be combined with other known technologies, and the embodiments can also be combined with each other. Without departing from the gist, a part of the structure can also be omitted or changed.
[0104] Description of reference numerals
[0105] 1, 1-1 wire electrical discharge machining machine, 2, 2-1 control device, 3 power supply device, 4 workpiece to be machined, 5 detection unit, 6 second control device, 11 wire electrode, 12 position adjustment unit, 13 power supply unit, 13a upper power supply member, 13b lower power supply member, 14 machining fluid supply unit, 15 wire feed speed adjustment unit, 41 machining chips, 42 machining streaks, 43, 44 machined surfaces, 45 inclined surface, 51 reprocessing section, 52 non-reprocessing section, 61 normal machining waveform, 62-1 momentary short-circuit waveform, 62-2 continuous short-circuit waveform, 70 learning device, 71 learning data acquisition unit, 72 model generation unit, 73 trained model storage unit, 74 inference device, 75 inference data acquisition unit, 76 inference section, 81 wire electrode spool, 82 wire electrode recovery box, 83 wire electrode conveying roller, 84 lower roller, 85 wire traveling speed control motor, 86 upper guide member, 87 lower guide member, 88 machining section, 89 X-axis motor, 90 Y-axis motor, 100, 100-1 wire electrical discharge machining system, D1 machining direction, D2 retraction direction, D3 re-approach direction, Ps starting point, T platform.
Claims
1. A control device for a wire electrical discharge machining machine that controls the wire electrical discharge machining machine which applies a voltage between a wire electrode and a workpiece to generate a discharge, thereby performing removal machining on the workpiece. The control device for the wire electrical discharge machining machine is characterized in that when the wire electrical discharge machining machine detects a short circuit based on at least one of the machining current and machining voltage during machining, i.e., a monitored value, the power supply to the wire electrode is stopped to interrupt the machining, the wire electrode is moved in the offset direction in which the distance between the wire electrode and the machining surface of the workpiece increases by a predetermined retraction distance, thereby retracting the wire electrode, after a predetermined retraction time has elapsed after retracting the wire electrode, the wire electrode is again moved closer to the machining surface to resume the machining.
2. The control device for the wire electrical discharge machining machine according to claim 1, characterized in that when resuming the machining after retracting the wire electrode, after the wire electrode approaches the workpiece to a distance at which a discharge will occur for the applied machining voltage and machining current, the power supply to the wire electrode is resumed.
3. The control device for the wire electrical discharge machining machine according to claim 1 or 2, characterized in that the short circuit is detected based on at least one of the change amount of the monitored value and the duration in a state where the change amount of the monitored value exceeds a predetermined threshold.
4. The control device for the wire electrical discharge machining machine according to claim 1, characterized in that re-machining is performed on the portion of the machining streak based on the control delay time from detecting the short circuit to retracting the wire electrode and the length of the machining streak, which is the remaining cut portion of the workpiece caused by the short circuit estimated from the monitored value.
5. The control device for the wire electrical discharge machining machine according to claim 4, characterized in that based on the control delay time and the length of the machining streak, the start point of the machining streak, the re-machining section for performing re-machining on the workpiece, and the non-re-machining section where re-machining is not performed are estimated, and based on the estimation result, the position of the wire electrode and the power supply to the wire electrode during re-machining are controlled.
6. The control device for the wire electrical discharge machining machine according to claim 1, characterized in that when detecting the short circuit, the flow rate of the machining fluid supplied between the workpiece and the wire electrode is increased compared to before detecting the short circuit.
7. The control device for the wire electrical discharge machining machine according to claim 1, characterized in that the machining conditions when resuming the machining are changed based on the change amount of the monitored value when detecting the short circuit.
8. The control device for the wire electrical discharge machining machine according to claim 7, characterized in that Based on the change amount of the monitored object value when the short circuit is detected, the height of the cutting residue part of the workpiece to be machined, that is, the machining streak, caused by the short circuit is estimated, and the machining conditions during the restoration of the machining are changed according to the estimated value of the height of the machining streak.
9. The control device of the wire electrical discharge machining machine according to claim 1, characterized in that: It further has a learning device, and the learning device has: A learning data acquisition unit that acquires learning data, which includes condition information indicating the conditions of the machining performed by the wire electrical discharge machining machine, and information used for reprocessing the cutting residue part of the workpiece to be machined, that is, the machining streak, generated when a short circuit occurs in the wire electrical discharge machining machine; And A model generation unit that uses the learning data to generate a trained model for inferring the information used for reprocessing the machining streak according to the condition information of the wire electrical discharge machining machine.
10. The control device of the wire electrical discharge machining machine according to claim 1, characterized in that: It further has an inference device, and the inference device has: An inference data acquisition unit that acquires condition information indicating the conditions of the machining performed by the wire electrical discharge machining machine as inference data; and An inference unit that uses a trained model for inferring the information used for reprocessing the cutting residue part of the workpiece to be machined, that is, the machining streak, generated when a short circuit occurs in the wire electrical discharge machining machine according to the condition information, and outputs the information used for reprocessing the machining streak according to the condition information input from the inference data acquisition unit, The machining is restored using the inference result of the inference device.
11. The control device of the wire electrical discharge machining machine according to claim 9 or 10, characterized in that: The condition information includes at least one of the plate thickness of the workpiece to be machined, the electrode diameter of the wire electrode, the machining trajectory, the machining conditions, the machining current, and the machining voltage.
12. The control device of the wire electrical discharge machining machine according to claim 9 or 10, characterized in that: The information used for reprocessing the machining streak includes at least one of the start point of the machining streak, the reprocessing interval corresponding to the part where the machining streak is generated, that is, the reprocessing interval, the non - machining interval where machining is not performed after starting the reprocessing, the machining conditions during reprocessing, the retraction trajectory, and the re - approach trajectory.
13. A control program product for a wire electrical discharge machining machine, which controls a wire electrical discharge machining machine that applies a voltage between a wire electrode and a workpiece to be machined to generate a discharge, thereby performing removal machining on the workpiece to be machined, The control program product of the wire electrical discharge machining machine is characterized in that: When the wire electrical discharge machining machine detects a short circuit based on at least one of the machining current and the machining voltage during machining, that is, the monitored object value, The wire electrical discharge machining machine is made to execute the following steps: Stop supplying power to the wire electrode to interrupt the machining; The wire electrode is retracted by moving the wire electrode in a deviation direction in which the distance between the wire electrode and the machining surface of the workpiece increases by a predetermined retraction distance; and After a predetermined retraction time has elapsed after retracting the wire electrode, the wire electrode is again brought close to the machining surface to resume the machining.
14. A control method for a wire electrical discharge machining machine that controls a wire electrical discharge machining machine which applies a voltage between a wire electrode and a workpiece to generate a discharge, thereby performing a removal machining on the workpiece, The control method for the wire electrical discharge machining machine is characterized in that when the wire electrical discharge machining machine detects a short circuit based on at least one of a machining current and a machining voltage during machining, which are monitored values, the method includes the following steps: stopping the power supply to the wire electrode to interrupt the machining; the wire electrode is retracted by moving the wire electrode in a deviation direction in which the distance between the wire electrode and the machining surface of the workpiece increases by a predetermined retraction distance; and after a predetermined retraction time has elapsed after retracting the wire electrode, the wire electrode is again brought close to the machining surface to resume the machining.
Citation Information
Patent Citations
Short-circuit releasing method of wire cut electric discharge machine
JP1993077109A
Wire electrical discharge machine and control method
CN112296462A
Wire cut discharge machining device
JP1994277949A