Machining method and machining management system

Abstract

To provide a machining method and machining management system that upgrade the machining quality of a workpiece.SOLUTION: A machining method in accordance with an embodiment of the present disclosure includes a step at which a processor causes a machining apparatus to machine a workpiece in a first process, and a step at which the processor predicts a physical property value of the workpiece machined in the first process. The machining method further includes a step at which the processor determines a requirement for machining of the workpiece in a second process on the basis of the predicted physical property value.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a processing method and a processing management system. [Background technology]

[0002] Patent Document 1 discloses a cutting evaluation system for evaluating the quality of a workpiece manufactured by cutting. This cutting evaluation system includes a learning processing unit that updates an evaluation model that derives an evaluation of the cutting process based on state variables, based on measured cutting results. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-127968 A Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure provides a processing method and processing control system for improving the processing quality of a workpiece. [Means for solving the problem]

[0005] A processing method according to one embodiment of the present disclosure includes: A step of causing a processing machine to process a workpiece in a first step by a processor; A step in which a processor predicts a physical property value of a workpiece after processing by the first process; A step in which a processor determines processing conditions for the workpiece by a second process based on the predicted physical property values; Includes.

[0006] A processing management system according to one aspect of the present disclosure includes: Processing machine, a processor for controlling the processing machine; A processing management system for managing processing of a workpiece, The processor The processing machine processes the workpiece in the first step, Predicting physical properties of the workpiece after processing by the first process; Based on the predicted physical property values, processing conditions for the workpiece in the second step are determined. Effect of the Invention

[0007] According to the present disclosure, the processing quality of the workpiece can be improved. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a configuration example of a process management system according to an embodiment of the present disclosure. [Diagram 2] 2 is a flowchart illustrating an example of a front-end process performed by the machining management system of FIG. 1. [Diagram 3] 3 is a graph showing an example of a load waveform L acquired in the acquisition step of FIG. 2. [Figure 4] 1 is a graph illustrating an example of the relationship between tool wear and burr height after punching. [Diagram 5] FIG. 2 is a schematic plan view of the workpiece after punching. [Figure 6] FIG. 11 is a diagram for explaining an example of a method for calculating a burr height. [Figure 7] FIG. 2 is a schematic diagram showing an example of the shape of a workpiece fed into a bending machine; [Figure 8] 2 is a flowchart illustrating an example of a post-process operation by the machining management system of FIG. 1. [Figure 9] FIG. 13 is a block diagram showing an example of the configuration of a processing management system according to a third modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] (Findings on which this disclosure is based) The present inventors have conducted extensive research to improve the processing quality of a workpiece in plastic processing such as press working, and as a result have come to the following findings.

[0010] In plastic processing, processing steps using multiple processing machines may be combined. To maintain high processing quality, it is necessary to adjust the processing conditions according to fluctuations in the physical properties of the workpieces input into each process, such as the external shape and hardness. One example of a cause of fluctuations in physical properties is wear of the parts of the processing machines, such as the dies, tools, and blades. The physical properties of the workpiece after each process is completed may fluctuate due to wear of the parts.

[0011] Patent Document 1 discloses a cutting evaluation system that monitors the state during cutting and evaluates the processing quality of a workpiece manufactured by cutting.

[0012] However, when multiple processing steps are combined, there may be no space or time between the previous and next steps to measure the physical properties of the workpiece. In such a case, the processing of the next step must be performed without being able to grasp the fluctuations in the physical properties caused by the previous step as numerical values. In the system shown in Patent Document 1, which evaluates the processing quality using information within one process, the processing conditions cannot be optimized according to the fluctuations in the physical properties of the workpiece, and there is a limit to maintaining high processing quality.

[0013] In addition, the cutting evaluation system of Patent Document 1, which evaluates the processing quality of a workpiece manufactured by cutting, only evaluates the processing quality of a workpiece that has already been processed. Therefore, although it is possible to feed back the evaluation results to the next and subsequent processing, it is not possible to improve the processing quality of the workpiece that has already been processed.

[0014] Based on these findings, the present inventors have discovered a way to improve the machining quality of a workpiece compared to the conventional technology.

[0015] Hereinafter, the embodiments will be described with reference to the drawings as appropriate. However, more detailed explanation than necessary may be omitted. For example, detailed explanation of already well-known matters or duplicate explanation of substantially the same configuration may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to facilitate understanding by those skilled in the art.

[0016] The inventor(s) provide the accompanying drawings and the following description so that those skilled in the art can fully understand the present disclosure, and do not intend for them to limit the subject matter described in the claims.

[0017] (Embodiment) 1. Configuration 1 is a block diagram showing an example of the configuration of a processing management system 100 according to an embodiment of the present disclosure. The processing management system 100 manages the manufacturing process of processed products. The processing management system 100 includes a punching control device (first processing control device) 10 that controls a punching machine 1, and a bending control device (second processing control device) 20 that controls a bending machine 2. The punching machine 1 is an example of the first processing machine of the present disclosure, and the bending machine 2 is an example of the second processing machine.

[0018] The punching machine 1 performs punching on a workpiece. The punching control device 10 includes a CPU 11, a storage device 12, and an input / output interface 13.

[0019] The CPU 11 performs information processing to realize the functions of the punching control device 10 described below. Such information processing is realized, for example, by the CPU 11 operating according to commands of a program stored in the storage device 12. The CPU 11 is an example of a processor of the present disclosure. The processor is not limited to a CPU as long as it includes an arithmetic circuit that performs calculations for information processing. For example, the processor may be configured by a circuit such as an MPU or an FPGA.

[0020] The storage device 12 is a recording medium for recording various data, a program required to realize the functions of the punching control device 10, and various information including a wear prediction model 121 and a burr prediction model 122 described below. The storage device 12 is realized by, for example, a semiconductor storage device such as a flash memory, a solid state drive (SSD), a magnetic storage device such as a hard disk drive (HDD), or other recording media, either alone or in combination. The storage device 12 may include a volatile memory such as an SRAM or a DRAM.

[0021] The input / output interface 13 is an interface circuit that connects the punching control device 10 to an external device. The input / output interface 13 may be a communication circuit that performs data communication with the external device according to an existing wired communication standard or wireless communication standard. An example of such an external device is the punching machine 1.

[0022] Another example of an external device is a load sensor 5 attached to a tool portion of a die of the punching machine 1. The input / output interface 13 inputs information such as the detection results by the load sensor 5 to the punching control device 10. The load sensor 5 is installed, for example, between the punch and a punch backing plate of the punching machine 1. The load sensor 5 is, for example, a piezoelectric force sensor or an electric force sensor such as a strain gauge type, and measures the load applied to the punch when the punch punches a workpiece.

[0023] Yet another example of the external device is one or more position sensors attached to the mold. The position sensors detect the inclination, height, etc. of the mold.

[0024] Moreover, the CPU 11 outputs information to an external device via the input / output interface 13. Such an external device is, for example, a bending control device 20, a display, another information processing terminal, or the like.

[0025] The bending control device 20 performs bending on a workpiece. The bending control device 20 includes a CPU 21, a storage device 22, and an input / output interface 23. The CPU 21, the storage device 22, and the input / output interface 23 may have the same configurations as the CPU 11, the storage device 12, and the input / output interface 13 of the punching control device 10, respectively. The input / output interface 23 connects the bending control device 20 to external devices such as the punching control device 10.

[0026] The storage device 22 records various data, programs required to realize the functions of the bending control device 20, and various information including a bending condition derivation model 221 described later.

[0027] 2.Operation 2-1. Pre-process Fig. 2 is a flow chart illustrating an example of a pre-process operation by the machining management system 100 of Fig. 1. In this embodiment, the pre-process operation is an operation related to punching. The process of Fig. 2 is executed by, for example, the CPU 11 of the punching control device 10.

[0028] First, the CPU 11 controls the punching machine 1 to perform punching on a workpiece (S11).

[0029] The CPU 11 acquires from the load sensor 5 a load waveform detected by the load sensor 5 during punching (S12). The load waveform is an example of processing state data showing the state of the punching process. The CPU 11 may acquire position data from one or more position sensors attached to the die, or may acquire other processing state data. The CPU 11 acquires information by monitoring the processing state, and performs processing such as converting the acquired analog signal into a digital signal and expressing it as a waveform, and extracting feature values ​​from the waveform.

[0030] FIG. 3 is a graph showing an example of the load waveform L acquired in acquisition step S12. The load waveform L in FIG. 3 shows the load detected by the load sensor 5 attached to the punch of the punching machine 1. In the graph in FIG. 3, it can be seen that the load applied to the punch increases over time from the point when the punch contacts the workpiece during one punching shot. When punching of the workpiece is completed, the load applied to the punch is no longer applied. The load waveform L represents the load state applied to the tool portion of the die of the punching machine 1.

[0031] 2, the CPU 11 predicts the amount of wear of the tool of the punching machine 1 using the wear prediction model 121 (S13). For example, the CPU 11 inputs the load waveform L acquired in step S12 to the wear prediction model 121, and acquires the predicted amount of wear output from the wear prediction model 121.

[0032] The wear prediction model 121 is a trained model trained to output information indicating the amount of wear of the tool of the punching machine 1 in response to input of processing state data such as the load waveform L. Such a trained model is obtained, for example, by having a model having a Convolutional Neural Network (CNN) structure learn a large number of load waveforms and tool wear amounts.

[0033] It is known that the waveform showing the load applied to a tool (punch or workpiece) over time during press processing such as punching depends on the amount of tool wear. Therefore, it is possible to estimate the amount of tool wear from the load waveform obtained during processing. Thus, there is a correlation between the load waveform and the amount of tool wear.

[0034] An example of a learning method of the wear prediction model 121 is as follows. First, a large number of learning data sets including the tool wear amount due to punching and the load waveform applied to the tool are prepared. The tool wear amount and the load waveform of the learning data set may be prepared using a simulation or the like. Alternatively, the tool wear amount and the load waveform of the learning data set may be data obtained by a systematic experiment on the amount of wear in a laboratory environment. In other words, the tool wear amount and the load waveform of the learning data set may be the tool wear amount measured before punching the workpiece and the load waveform measured when actually punching the workpiece, respectively.

[0035] In the learning data set, the tool wear amount is the correct answer data, and the learning model learns the load waveform labeled with such correct answer data. That is, in the learning data set for the wear prediction model 121, the load waveform indicating the load on the tool is an explanatory variable, and the tool wear amount is a response variable.

[0036] A large number of learning data sets thus prepared are stored in a storage device of an information processing device such as a server, and are used for learning the wear prediction model 121.

[0037] The wear prediction model 121, for example, compares the model waveforms of these load waveforms with the actual waveforms obtained at the production site, and selects a waveform that minimizes the difference between the model waveform and the actual waveform of the load waveform. The wear prediction model 121 calculates the tool wear amount associated with the selected waveform.

[0038] After step S13, the CPU 11 uses the burr prediction model 122 to predict the height of burrs generated on the workpiece by the machining in step S11 (S14).

[0039] The burr prediction model 122 is a trained model trained to output the burr height (or burr size) of a workpiece in response to input of information indicating the amount of wear of a tool of the punching machine 1. Such a trained model is obtained, for example, by having a model having a CNN structure learn a large number of tool wear amounts and burr heights.

[0040] In the training data set for the burr prediction model 122, information indicating the burr height is ground truth data, and the training model learns the tool wear amount labeled with such ground truth data. That is, in the training data set for the burr prediction model 122, the tool wear amount is an explanatory variable, and the burr height is a target variable.

[0041] The learning data set for the burr prediction model 122 may be, for example, data obtained by a systematic experiment on the amount of wear in a laboratory environment. Fig. 4 is a graph illustrating an example of the relationship between the amount of tool wear and the burr height after punching. The burr prediction model 122 can predict the burr height of a workpiece from the predicted amount of tool wear by determining the relationship shown in Fig. 4 based on a data set of the amount of tool wear and the burr height after punching prepared using the results of the above-mentioned experiment.

[0042] Here, an example of a method for calculating the burr height will be described. Fig. 5 is a schematic plan view of a workpiece 84 after punching. In Fig. 5, a design dimension 81 of the punching position is shown by a dashed line, and an actual punching dimension 82 indicating the area that is actually punched is shown by a solid line. For example, the design dimension 81 and the actual punching dimension 82 are measured as the distance from the center of the design dimension 81. The actual punching dimension 82 may deviate from the design dimension 81.

[0043] 6 is a diagram for explaining an example of a method for calculating the burr height. In one example, an actual punching dimension 82 is measured over the entire circumference in the circumferential direction, an average value of a design dimension 81 and a punching dimension 82 of a workpiece over the entire circumference in the circumferential direction is calculated, and the difference between the two average values ​​is calculated as the burr height 8.

[0044] The above-mentioned method of calculating the burr height is merely an example, and the basis for calculating the burr height does not necessarily have to be the design dimension 81. For example, when the tools such as the punch and die of the metal mold are hardly worn, there is almost no difference between the actual punching dimension 82 and the design dimension 81. For this reason, the punching dimension of a workpiece punched in a state where the tools are hardly worn may be used as the basis for calculating the burr height.

[0045] Further, in the above example, the burr height is calculated from the punching dimension 82 of the remaining workpiece 84, but the burr height may be calculated from the dimension of the punched-out workpiece.

[0046] Returning to FIG. 2, the CPU 11 transmits the burr height data obtained in step S14 to the bending control device 20 via the input / output interface 13 (S15).

[0047] 2-2. Post-process The workpiece 84 that has been punched in the previous process is sent to the bending machine 2 and is bent in the subsequent process.

[0048] Fig. 7 is a schematic diagram showing an example of the shape of a workpiece 84 fed into the bending machine 2. Fig. 7 shows an example of a rectangular workpiece 84. The edge of the workpiece 84 has a burr 83. For example, the bending machine 2 bends the workpiece 84 so that the workpiece 84 is bent along the dashed line A-A shown in Fig. 7.

[0049] Fig. 8 is a flow chart illustrating an example of the operation of the post-process by the bending control system 100 of Fig. 1. The processing of Fig. 8 is executed by the CPU 21 of the bending control device 20, for example.

[0050] First, the CPU 21 acquires the burr height data transmitted from the bending control device 20 via the input / output interface 23 (S21). This burr height data is the data transmitted from the bending control device 20 in step S15.

[0051] Next, the CPU 21 derives bending conditions (S22) using the bending condition derivation model 221. The bending conditions are processing conditions for bending by the bending machine 2. The CPU 21 controls the bending machine 2 based on the bending conditions derived in step S22, and performs bending on the workpiece (S23).

[0052] The bending condition derivation model 221 is a model that outputs bending conditions such as bending angle in response to input of burr height data.

[0053] An example of the bending condition derivation model 221 is a trained model trained to output bending conditions such as a bending angle in response to input of burr height data. Such a trained model is obtained, for example, by training a model having a CNN structure on a large number of burr height data and bending conditions. In the training data set for the bending condition derivation model 221, information indicating the bending conditions is correct data, and the training model trains on burr height data to which such correct data is attached as a label.

[0054] Next, another example of the bending condition derivation model 221 will be described. In this example, the bending condition derivation model 221 is expressed by a correlation equation. One example of the correlation equation is expressed by one or more polynomials with the bending quality as the objective variable and the bending conditions and the physical property values ​​of the workpiece as explanatory variables. For example, the correlation equation includes the following equation (1) and equation (2).

[0055] Y1= f(X1)+f(X2)+f(X3)+f(X4)+f(X5)+f(X6)+f(X7)+f(X8) ···(1)

[0056] Y2= F(X1)+F(X2)+F(X3)+F(X4)+F(X5)+F(X6)+F(X7)+F(X8) ···(2)

[0057] For example, the objective variable Y1 is the bend angle, and the objective variable Y2 is the twist angle.

[0058] As for the explanatory variables, for example, X1 indicates the pressing load, X2 indicates the pressing position, X3 indicates the pressing speed, X4 indicates the workpiece gripping load, X5 indicates the burr height of the workpiece, X6 indicates the thickness of the workpiece, X7 indicates the material of the workpiece, and X8 indicates the hardness (hardness) of the workpiece.

[0059] f(X n ) and F(X n ) is the explanatory variable X n For example, f(X1) represents the relationship between the pressing load X1 and the bending angle Y1.

[0060] For example, the correlation equation f(X n ) and F(X n ) is calculated by varying the processing conditions among a number of conditions in a simulation. Alternatively, or in addition, f(X n ) and F(X n ) may be calculated using measurements obtained by varying process conditions between multiple conditions in a laboratory or factory floor environment. Alternatively, f(X n ) and F(X n ) may be calculated based on real and simulated measurements.

[0061] In this embodiment, for example, if the burr height X5 of the workpiece fluctuates, the bending angle Y1 will fluctuate. However, the bending condition derivation model 221 can stabilize Y1 by controlling and fluctuating other explanatory variables so as to offset the fluctuation value of Y1.

[0062] For example, when the burr height X5 of the workpiece fluctuates, the bending condition derivation model 221 calculates the fluctuation value of the indentation load X1 of the bending machine 2 so as to offset the fluctuation value of Y1. This allows the CPU 21 using the bending condition derivation model 221 to appropriately control the indentation load for each bending cycle.

[0063] In this way, the processing management system 100 can stabilize the bending quality. Specifically, the processing management system 100 can reduce the risk of a decrease in the quality of the product bent in a later process even if the burr height, thickness, material, hardness, etc. of the workpiece fluctuate in the previous process. For example, even if the burr height increases due to the previous process, which increases the rigidity of the workpiece in the bending direction, the bending control device 20 can control the bending process to increase the pressing load in accordance with the increase in rigidity. In this way, the processing management system 100 can keep the bending angle of the workpiece constant.

[0064] As described above, the machining control system 100 can calculate the machining conditions in the subsequent process based on the physical properties of the workpiece predicted in the previous process, and can stabilize the bending quality.

[0065] 3. Effects etc. As described above, the processing method according to this embodiment includes a step (S11) of causing the punching machine 1 to process the workpiece in a first step, a step (S14) of predicting the physical property values ​​of the workpiece after processing in the first step, and a step (S22) of determining processing conditions for the workpiece in a second step based on the predicted physical property values. The processing method may further include a step (S23) of causing the bending machine 2 to process the workpiece in the second step based on the determined processing conditions. In this embodiment, steps S11 and S14 are executed by the CPU 11, and steps S22 and S23 are executed by the CPU 21.

[0066] According to this method, the machining quality of the workpiece can be improved by using the machining conditions determined based on the predicted physical property values ​​in the second step.

[0067] The step of predicting the physical property value may include predicting the physical property value based on machining state data indicating the state of machining of the workpiece in the first process. For example, the step of predicting the physical property value may include the CPU 11 inputting the machining state data into a trained model, acquiring an output from the trained model, and determining the output as the physical property value. With this configuration, the machining quality of the workpiece can be improved.

[0068] The step of predicting the physical property value may include predicting the physical property value based on load data indicating the load applied to the workpiece in the first process. For example, the step of predicting the physical property value may include estimating a wear amount of a tool of the punching machine 1 based on the load data (S13), and predicting the physical property value based on the estimated wear amount (S14).

[0069] (Other embodiments) As described above, the embodiment has been described as an example of the technology in the present disclosure. However, the technology in the present disclosure is not limited to this, and can be applied to embodiments in which modifications, substitutions, additions, omissions, etc. are appropriately performed. In addition, it is also possible to combine the components described in the above embodiment to form a new embodiment. Therefore, the following describes modified examples as other embodiments.

[0070] (First Modification) In the above embodiment, an example has been described in which tool wear in punching affects the burr height of a workpiece, and the punching control device 10 transmits burr height data to the bending control device 20. However, the physical property value of a workpiece that may be affected by tool wear in punching is not limited to the burr height. For example, punching may affect at least one of the burr height, thickness, hardness, and processing dimension of the workpiece.

[0071] The burr height, thickness, hardness, and machining dimensions of the workpiece described above are examples of the physical property values ​​of the workpiece of the present disclosure. The physical property values ​​of the workpiece are physical quantities that represent the physical properties of the workpiece, including the shape and properties of the workpiece.

[0072] Tool wear in punching can affect the burr height of a workpiece because, for example, tool wear changes the way the workpiece is stretched. Tool wear in punching can affect the thickness of a workpiece because, for example, tool wear changes the way the workpiece is stretched. Tool wear in punching can affect the hardness of a workpiece because, for example, tool wear changes the way the workpiece is work-hardened. Tool wear in punching can affect the thickness of a workpiece because, for example, tool wear changes the way the workpiece is stretched.

[0073] The CPU 11 predicts the physical property value of the workpiece using a physical property value prediction model according to the physical property value of the workpiece that may be affected by tool wear in punching. For example, when tool wear affects the hardness of the workpiece, the CPU 11 predicts the hardness of the workpiece using a hardness prediction model instead of step S14. In this way, the prediction model may be a model that predicts the physical property value of the workpiece that may be affected by the previous process. Such a prediction model is a trained model that has been trained to output the physical property value of the workpiece in response to input of information indicating the amount of wear of the tool of the punching machine 1.

[0074] (Second Modification) In the above embodiment, an example has been described in which the front-end process is a punching process and the back-end process is a bending process, but the front-end process and the back-end process of the present disclosure are not limited to this.

[0075] The previous process may be a plastic process other than punching, such as bending or forging. When the previous process is bending, the first processing machine is a bending machine. When the previous process is forging, the first processing machine is a forging machine.

[0076] The subsequent process may be a plastic process other than bending, such as punching or forging. When the subsequent process is punching, the second processing machine is a punching machine. When the subsequent process is forging, the second processing machine is a forging machine.

[0077] Therefore, both the front-end process and the back-end process can be selected from punching, bending, and forging. In this case, there are nine combinations of the front-end process and the back-end process.

[0078] When the previous process is bending, the physical property value of the workpiece that may be affected by the previous process is, for example, at least one of the thickness, hardness, and processing dimensions of the workpiece.

[0079] Bending can affect the thickness of a workpiece because the plastic deformation during bending varies, causing the thickness of the processed area to change. Bending can affect the hardness of a workpiece because, for example, the plastic deformation during bending varies, causing the hardness of the processed area to change. Bending can affect the processed dimensions of a workpiece because, for example, the tension during bending changes due to wear of tools such as the workpiece holder, causing the bending angle to change.

[0080] When the previous process is a forging process, the physical property value of the workpiece that may be affected by the previous process is, for example, at least one of the thickness, hardness, processing dimension, and burr height of the workpiece.

[0081] The forging process may affect the thickness of the workpiece because, for example, the amount of crushing of the workpiece during the forging process varies due to the influence of the workpiece material hardness, die wear, etc. The forging process may affect the hardness of the workpiece because, for example, the amount of crushing of the workpiece during the forging process varies and the amount of work hardening of the workpiece varies due to the influence of the workpiece material hardness, die wear, etc. The forging process may affect the processing dimensions of the workpiece because, for example, the amount of crushing of the workpiece during the forging process varies and the dimensions of the workpiece vary due to the influence of the workpiece material hardness, die wear, etc. The forging process may affect the burr height of the workpiece because, for example, the burr height varies without being controlled due to die designs that allow burrs.

[0082] When the subsequent process is punching, the CPU 21 changes the processing conditions according to the physical properties of the workpiece as follows. For example, when the hardness of the workpiece varies depending on the previous process, the CPU 21 increases the impact force by changing the punching speed. When the burr height varies depending on the previous process, the CPU 21 can reduce the risk of clogging due to the burrs by changing the punch pressing amount during punching.

[0083] When the subsequent process is bending, the CPU 21 changes the processing conditions according to the physical properties of the workpiece as follows. For example, when the thickness of the workpiece varies depending on the previous process, the CPU 21 changes the pressing load, pressing position, or die height during bending. When the hardness of the workpiece varies depending on the previous process, the CPU 21 changes the pressing load, pressing position, or die height during bending. When the processing dimensions of the workpiece vary depending on the previous process, the CPU 21 shifts the bending center or bending position. When the burr height of the workpiece varies depending on the previous process, the CPU 21 changes the pressing load, pressing position, or die height during bending.

[0084] When the subsequent process is forging, the CPU 21 changes the processing conditions according to the physical properties of the workpiece as follows. For example, when the thickness of the workpiece varies depending on the previous process, the CPU 21 changes the pressing position, pressing amount, or die height during the forging process. For example, when the hardness of the workpiece varies depending on the previous process, the CPU 21 changes the pressing load during the forging process. For example, when the processing dimensions of the workpiece vary depending on the previous process, the CPU 21 changes the pressing position, pressing amount, or die height during the forging process.

[0085] In this modification, instead of the bending condition derivation model 221, a machining condition derivation model for deriving the above-mentioned machining conditions may be used.

[0086] (Third Modification) In the above embodiment, an example has been described in which the punching control device 10, which is an example of the first processing control device, and the bending control device 20, which is an example of the second processing control device, are provided as separate devices. However, the first processing control device and the second processing control device may be integrated. For example, the first processing control device and the second processing control device may be integrated in one housing.

[0087] 9 is a block diagram showing a configuration example of a processing management system 300 according to a third modified example. The processing management system 300 includes a punching machine 1, a bending machine 2, a load sensor 5, and a processing control device 310. The processing control device 310 is connected to the punching machine 1, the bending machine 2, and the load sensor 5. This connection may be via a network. Therefore, the processing control device 310 may be a server device on the network.

[0088] The processing control device 310 includes a CPU 311, a storage device 312, and an input / output interface 313. The CPU 311, the storage device 312, and the input / output interface 313 may have the same configurations as the CPU 11, the storage device 12, and the input / output interface 13 of the punching processing control device 10, respectively.

[0089] Furthermore, the functions (components) of the first processing control device and the second processing control device may be distributed among three or more housings. At least some of the functions of the first processing control device and the second processing control device may be realized by cloud (cloud computing) or the like.

[0090] (Fourth Modification) In the above embodiment, an example has been described in which a processing step by a punching machine 1, which is an example of a first processing machine, and a processing step by a bending machine 2, which is an example of a second processing machine, are combined. However, the first processing machine and the second processing machine may be integrated. The integrated processing machine is equipped with, for example, a die generally called a composite die. In the composite die, the area inside the die is divided into multiple parts. By using the composite die, multiple processing steps (for example, the above-mentioned pre-process and post-process) can be applied to the workpiece in succession without removing the workpiece from the die.

[0091] (Fifth Modification) The first processing control device and / or the second processing control device may include an alarm device that notifies a user of information. Such an alarm device may include a light source such as an LED, a visual alarm device such as a display or an indicator, and / or a sound source such as a speaker, a buzzer, or a horn (auditory alarm device).

[0092] For example, the CPU 11 of the punching processing control device 10, which is an example of a first processing control device, may determine whether the punching condition in the punching machine 1 is normal or not based on the predicted tool wear amount or the burr height value of the workpiece, and may issue an alarm by an alarm device when the condition is not normal.

[0093] In addition, the bending process control device 20, which is an example of a second process control device, may determine whether the process state is normal or not based on the burr height data acquired in step S21 or the bending process state, and may issue an alarm by an alarm device when the process state is not normal.

[0094] (Sixth Modification) In the above embodiment, an example in which the previous process is punching and the next process is bending has been described. However, the number of processes in the present disclosure is not limited to two, and may be three or more. In this case, the bending control device 20, which is an example of the second processing control device, may transmit predicted physical property data of the workpiece after bending to a third processing control device that manages the subsequent process.

[0095] (Seventh Modification) In the above embodiment, the load waveform is given as an example of the processing state data indicating the state of the punching process, but the processing state data is not limited to this. For example, the processing state data may be data other than the load waveform that indicates the state of the punching process. Also, for example, the processing state data may be data that indicates the state of processing by a first processing machine other than the punching machine 1.

[0096] (Example) The following provides examples of aspects of the present disclosure.

[0097] <Aspect 1> A step of causing a processing machine to process a workpiece in a first step by a processor; A step in which the processor predicts a physical property value of the workpiece after processing by the first process; determining processing conditions for the workpiece in a second step based on the predicted physical property value by the processor; A processing method comprising the steps of:

[0098] <Aspect 2> 2. The processing method of claim 1, further comprising the step of: the processor causing the processing machine to process the workpiece in the second step based on the determined processing conditions.

[0099] <Aspect 3> The processing machine includes a first processing machine and a second processing machine, The processor, causing the first processing machine to process the workpiece in the first step; causing the second processing machine to process the workpiece in the second step; The processing method according to aspect 2.

[0100] <Aspect 4> A processing method according to any one of aspects 1 to 3, wherein the physical property value is a physical quantity that changes due to processing in the first step.

[0101] <Aspect 5> A processing method according to aspect 4, wherein the physical property value is at least one of burr height, thickness, hardness, and processing dimension of the workpiece.

[0102] <Aspect 6> A processing method according to any one of aspects 1 to 5, wherein the step of predicting the physical property value includes the processor predicting the physical property value based on processing state data indicating the processing state of the workpiece in the first process.

[0103] <Aspect 7> The step of predicting the physical property value includes, by the processor, inputting the processing state data into a trained model to obtain an output from the trained model, and determining the output as the physical property value; The trained model is a model trained to output the physical property value in response to the input of the processing state data. 7. The processing method according to embodiment 6.

[0104] <Aspect 8> The first process is a punching process, a bending process, or a forging process, The second process is a punching process, a bending process, or a forging process. The processing method according to any one of aspects 1 to 7.

[0105] <Aspect 9> The first step is a punching step, The step of predicting the physical property value includes predicting the physical property value based on load data indicating a load applied to the workpiece in the punching process by the processor. The processing method according to embodiment 8.

[0106] <Aspect 10> The step of predicting the physical property value includes the processor estimating a wear amount of a tool of the processing machine based on the load data, and predicting the physical property value based on the estimated wear amount. A processing method according to aspect 9.

[0107] <Aspect 11> Processing machine, A processor for controlling the processing machine, A processing management system for managing processing of a workpiece, The processor, causing the processing machine to process the workpiece in a first step; Predicting physical properties of the workpiece after processing in the first step; determining processing conditions for the workpiece in a second step based on the predicted physical property values; Processing management system. [Industrial Applicability]

[0108] The present disclosure is applicable to processing methods and processing management systems, and in particular to methods and management systems for plastic processing. [Explanation of symbols]

[0109] 1. Punching machine (first processing machine) 2 Bending machine (second processing machine) 5 Load Sensor 10 Punching process control device (first process control device) 11 CPU (Processor) 12 Storage device 13 Input / Output Interface 20 Bending process control device (second process control device) 21 CPU (Processor) 22 Storage device 23 Input / Output Interface 81 Design dimensions 82 Dimensions 83 Bali 84 Remaining Work 84 Work 100 Processing Management System 121 Wear prediction model 122 Bali Prediction Model 221 Condition Derivation Model 300 Processing Management System 310 Processing control device 311 CPU (processor) 312 Storage device 313 Input / Output Interface

Claims

1. A processor causes a processing machine to process a workpiece in a first step; A step in which the processor predicts a physical property value of the workpiece after processing by the first process; determining processing conditions for the workpiece in a second process based on the predicted physical property value by the processor; A processing method comprising the steps of:

2. The machining method according to claim 1 , further comprising the step of: the processor causing the machine to machine the workpiece in the second step based on the determined machining conditions.

3. The processing machine includes a first processing machine and a second processing machine, The processor, causing the first processing machine to process the workpiece in the first step; causing the second processing machine to process the workpiece in the second step; The processing method according to claim 2.

4. The processing method according to claim 1 , wherein the physical property value is a physical quantity that changes due to processing in the first step.

5. The processing method according to claim 4 , wherein the physical property value is at least one of a burr height, a thickness, a hardness, and a processing dimension of the workpiece.

6. The processing method according to claim 1 , wherein the step of predicting the physical property value includes the processor predicting the physical property value based on processing state data indicating a processing state of the workpiece in the first process.

7. The step of predicting the physical property value includes, by the processor, inputting the processing state data into a trained model to obtain an output from the trained model, and determining the output as the physical property value; The trained model is a model trained to output the physical property value in response to the input of the processing state data. The processing method according to claim 6.

8. The first process is a punching process, a bending process, or a forging process, The second process is a punching process, a bending process, or a forging process. The processing method according to any one of claims 1 to 7.

9. The first step is a punching step, The step of predicting the physical property value includes predicting the physical property value based on load data indicating a load applied to the workpiece in the punching process by the processor. The processing method according to claim 8.

10. The step of predicting the physical property value includes the processor estimating a wear amount of a tool of the processing machine based on the load data, and predicting the physical property value based on the estimated wear amount. The processing method according to claim 9.

11. Processing machine, A processor for controlling the processing machine, A processing management system for managing processing of a workpiece, The processor, causing the processing machine to process the workpiece in a first step; Predicting physical property values ​​of the workpiece after processing in the first step; determining processing conditions for the workpiece in a second step based on the predicted physical property values; Processing management system.

Images