Apparatus for estimating processing state and method for estimating processing state

Abstract

This machining state estimating device comprises a storage device and a processor. The storage device stores a plurality of parameters defining a machining state of a machining machine, and a plurality of items of reference data corresponding respectively to all combinations of the plurality of parameters. The processor acquires measurement data representing measurement results of a machining load for the machining machine, determines a degree of similarity, being an indicator of a degree of similarity between the plurality of items of reference data and the measurement data, and estimates the machining state on the basis of the determined degree of similarity.

Description

Technical Field

[0001] The present disclosure relates to a processing state estimation device and a processing state estimation method.

Background Art

[0002] Patent Document 1 discloses a technique for obtaining a determination value by synthesizing state quantities in normal equipment and state quantities in abnormal equipment in an apparatus that repeats the same operation, such as a press machine, in a relatively short cycle. The determination device of Patent Document 1 generates an alarm when the state quantity of the target device exceeds the determination value or falls below the determination value.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

[0004] However, in the prior art, if the determination value is set leniently, abnormalities of the device such as tool wear cannot be detected, and if the determination value is set strictly, there is a problem that the device is determined to be abnormal even though it is normal.

[0005] An object of the present disclosure is to provide a processing state estimation device and a processing state estimation method that can estimate the processing state by a processing machine with higher accuracy than the prior art.

[0006] A processing state estimation device according to an aspect of the present disclosure includes a storage device and a processor, wherein the storage device stores a plurality of parameters that define the processing state of the processing machine, and a plurality of reference data respectively corresponding to all combinations of the plurality of parameters, and the processor acquires measurement data indicating a measurement result of a processing load by the processing machine, The degree of similarity between the aforementioned multiple reference data and the aforementioned measurement data is determined, Based on the determined similarity, the processing state is estimated.

[0007] A method for estimating the processing state according to one aspect of this disclosure is: The processor acquires measurement data showing the measurement results of the machining load by the machining machine, The processor performs the step of determining a similarity, which is an indicator of the degree of similarity between a plurality of reference data corresponding to all combinations of a plurality of parameters defining the machining state of a machining tool and the measurement data, The processor estimates the processing state based on the determined similarity, Includes.

[0008] According to the processing state estimation device and processing state estimation method described herein, the processing state by a processing machine can be estimated with greater accuracy than in the prior art. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing an example configuration of a processing state estimation device according to the first embodiment. [Figure 2] Figure 1 is a schematic cross-sectional view showing a press machine to which the load sensor is attached. [Figure 3] Figure 1 is a flowchart illustrating the procedure for estimating the machining state, which is performed by the CPU of the machining state estimation device. [Figure 4] Figure 1 is a schematic graph showing an example of a measurement waveform from the load sensor. [Figure 5] Figure 1 shows an example of the status data in this table. [Figure 6] Figure 3 is a flowchart illustrating the processing flow of the normal state estimation process S5. [Figure 7] Figure 6 is a flowchart illustrating the punch wear amount estimation process S51. [Figure 8]It is a flowchart exemplifying the die wear amount estimation process S52 in FIG. 6. [Figure 9] It is a flowchart exemplifying the process flow of the post-polishing state estimation process S6 shown in FIG. 3. [Figure 10] It is a flowchart exemplifying the process flow of the post-polishing clearance estimation process S63 shown in FIG. 9. [Figure 11] It is a flowchart exemplifying the process flow of the post-exchange state estimation process S7 shown in FIG. 3. [Figure 12] It is a table showing an example of state data in a modification of the first embodiment. [Figure 13] It is a flowchart exemplifying the process flow of the normal state estimation process S5a in a modification of the first embodiment. [Figure 14] It is a flowchart exemplifying the process flow of the workpiece thickness estimation process S50 shown in FIG. 13. [Figure 15] It is a block diagram showing a configuration example of a machining state estimation device according to the second embodiment. [Figure 16] It is a schematic diagram for explaining the outline of the complementation process executed by the CPU of the machining state estimation device in FIG. 15. [Figure 17] It is a flowchart exemplifying the procedure of the complementation process. [Figure 18] It is a schematic diagram for explaining an example of the complementation process. [Figure 19] It is a block diagram showing a configuration example of a machining state estimation device according to the third embodiment. [Figure 20] It is a schematic graph showing an example of the time-slide position relationship data in FIG. 19. [Figure 21] It is a flowchart exemplifying the punch wear amount estimation process S51a in the third embodiment. [Figure 22] It is a schematic diagram for explaining the unit conversion process. [Figure 23] It is a schematic diagram for explaining the machining state in each cycle of the cycle machining. [[Embodiments for Carrying Out the Invention]]

[0010] (Knowledge that forms the basis of this disclosure) The inventors have conducted extensive research to accurately estimate the processing state of a processing machine in press working, particularly punching, and have obtained the following findings. Here, "processing state" refers to a state related to at least one of the parameters that define the processing state in this disclosure. In this embodiment, the processing state refers to at least one of the tool wear amount, clearance, or workpiece thickness.

[0011] The load applied to the punch or workpiece during punching depends on values ​​such as punch wear, die wear, clearance, and workpiece thickness.

[0012] Punch wear and die wear are examples of punch wear parameters, which are indicators of the degree of punch wear, and die wear parameters, which are indicators of the degree of die wear, respectively. Tool wear, such as punch wear and die wear, is expressed, for example, by the change in tool dimensions from when the tool was newly manufactured or regrinded. Tool wear may also be expressed as a change in shape, volume, mass, etc. Alternatively, tool wear may be expressed as the radius of an arc when the wear is approximated as a circular arc.

[0013] Clearance is the gap between the die and the punch. For example, clearance is the gap between the die and the punch when a punched hole is made in a workpiece. Clearance may also be expressed as the ratio of the gap between the die and the punch to the thickness of the workpiece.

[0014] Since the load depends on these parameters, it is conceivable to estimate these parameters from the load waveform obtained during machining. For example, if the amount of tool wear, such as punch wear and die wear, can be estimated, the optimal timing for sharpening or regrinding (hereinafter simply referred to as "sharpening") the tool can be determined in a machining center that performs cyclic machining. By sharpening the tool at the optimal timing, it is possible to prevent situations such as machining workpieces with worn tools and producing a large number of defective products, thereby increasing productivity.

[0015] In machining equipment that performs cyclic processing, there is an advantage to using the estimated processing state from the previous punching operation to estimate the processing state. One reason for this is that, as shown in Figure 23, values ​​such as clearance, punch wear, and die wear usually do not change significantly from the values ​​in the previous punching operation. By performing estimation under the condition that the results do not change significantly from the previous estimation, the accuracy of the estimation can be improved.

[0016] Furthermore, it has been found that tool wear, such as punch wear and die wear, increases unless the tool is sharpened or replaced, clearance remains almost constant unless the tool is replaced, and punch wear progresses faster than die wear. Based on these findings, the inventors discovered that the accuracy of estimation can be improved by performing estimations under conditions where, for example, punch wear and die wear increase slightly from the previous estimation result, leading to the present invention.

[0017] According to a first aspect of this disclosure, a device comprising a storage device and a processor, The aforementioned storage device is Multiple parameters that define the machining state of a machining machine, Each stores multiple reference data corresponding to each combination of the aforementioned multiple parameters, The aforementioned processor, Measurement data showing the measurement results of the processing load by the aforementioned processing machine is obtained, The degree of similarity between the aforementioned multiple reference data and the aforementioned measurement data is determined, Based on the determined similarity, the processing state is estimated. The present invention provides a device for estimating the processing state.

[0018] According to a second aspect of this disclosure, the processor is From the aforementioned multiple reference data, search for the reference data with the highest similarity. The present invention provides a processing state estimation device according to a first embodiment, which determines the combination of parameters corresponding to the searched reference data as an estimated parameter set representing the processing state at the time of measurement of the measurement data.

[0019] According to a third aspect of this disclosure, the processor provides a processing state estimation device according to the second aspect, wherein in the process of searching for the reference data, the processor sequentially changes the plurality of parameters within a predetermined range based on the estimated parameter set already determined by the processor, and searches for a portion of the plurality of reference data that has the greatest similarity to the measurement data.

[0020] According to a fourth aspect of this disclosure, the plurality of parameters include wear parameters that define the degree of tool wear of the processing machine and clearance parameters that define the clearance of the processing machine. The estimated parameter set includes an estimated wear parameter estimated as the wear parameter at the time of measurement of the measurement data, and an estimated clearance parameter estimated as the clearance parameter at the time of measurement of the measurement data. The present invention provides a processing state estimation device according to the second or third embodiment.

[0021] According to a fifth aspect of this disclosure, the processor, in the process of searching for the reference data, sequentially changes the wear parameter within a range of values ​​greater than or equal to the estimated wear parameter to search for the reference data that has the greatest similarity to the measurement data. The present invention provides a processing state estimation device according to the fourth embodiment.

[0022] According to a sixth aspect of this disclosure, the wear parameter includes a punch wear parameter that defines the degree of wear of the punch of the processing machine, and a die wear parameter that defines the degree of wear of the die of the processing machine, The estimated wear parameter includes the estimated punch wear parameter estimated as the punch wear parameter at the time of measurement of the measurement data, and the estimated die wear parameter estimated as the die wear parameter at the time of measurement of the measurement data. In the process of searching for the reference data, the processor sequentially changes the punch wear parameter within a value greater than or equal to the estimated punch wear parameter, and sequentially changes the die wear parameter within a value greater than or equal to the estimated die wear parameter, in order to search for reference data that has the greatest similarity to the measurement data. The present invention provides a processing state estimation device according to the fourth or fifth embodiment.

[0023] According to a seventh aspect of this disclosure, the processor, in the process of searching for the reference data, The die wear parameter is fixed to the estimated die wear parameter, and the punch wear parameter is sequentially varied within a value greater than or equal to the estimated punch wear parameter to search for reference data that has the maximum first similarity to the measured data. The punch wear parameter corresponding to the reference data with the highest first similarity is determined as the estimated punch wear parameter. The punch wear parameter is fixed to the determined estimated punch wear parameter, and the die wear parameter is sequentially varied within a value greater than or equal to the estimated die wear parameter to search for reference data that maximizes the second similarity with the measurement data. The die wear parameter corresponding to the reference data with the highest second similarity is determined as the estimated die wear parameter. The present invention provides a processing state estimation device according to the sixth embodiment.

[0024] According to an eighth aspect of this disclosure, the plurality of parameters further include a work thickness parameter that defines the thickness of the workpiece to be machined by the machining machine, In the process of searching for the reference data, the processor sequentially changes the workpiece thickness parameter to search for the reference data that has the greatest similarity to the measurement data. The present invention provides a processing state estimation device according to any of the fourth to seventh embodiments.

[0025] According to a ninth aspect of this disclosure, the wear parameter includes a punch wear parameter that defines the degree of wear of the punch of the processing machine, and a die wear parameter that defines the degree of wear of the die of the processing machine, The estimated wear parameter includes the estimated punch wear parameter estimated as the punch wear parameter at the time of measurement of the measurement data, and the estimated die wear parameter estimated as the die wear parameter at the time of measurement of the measurement data. In the process of searching for the reference data, the processor sequentially changes the punch wear parameter within a value greater than or equal to the estimated punch wear parameter, and sequentially changes the die wear parameter within a value greater than or equal to the estimated die wear parameter, in order to search for reference data that has the greatest similarity to the measurement data. The present invention provides a processing state estimation device according to the eighth aspect.

[0026] According to a tenth aspect of this disclosure, the estimated parameter set includes an estimated work thickness parameter estimated as the work thickness parameter at the time of measurement of the measurement data, In the process of searching for the reference data, the processor The die wear parameter and the punch wear parameter are fixed to the estimated die wear parameter and the estimated punch wear parameter, respectively, and the workpiece thickness parameter is sequentially changed to search for reference data that has the maximum first similarity to the measured data. The workpiece thickness parameter corresponding to the reference data with the highest first similarity is determined as the estimated workpiece thickness parameter. The workpiece thickness parameter and the die wear parameter are fixed to the determined estimated workpiece thickness parameter and the estimated die wear parameter, respectively, and the punch wear parameter is sequentially changed within a value greater than or equal to the estimated punch wear parameter to search for reference data that maximizes the second similarity with the measurement data. The punch wear parameter corresponding to the reference data with the highest second similarity is determined as the estimated punch wear parameter. The punch wear parameter is fixed to the determined estimated punch wear parameter, and the die wear parameter is sequentially varied within a value greater than or equal to the estimated die wear parameter to search for reference data that maximizes the third similarity with the measurement data. The die wear parameter corresponding to the reference data with the highest similarity is determined as the estimated die wear parameter. The present invention provides a processing state estimation device according to the ninth aspect.

[0027] According to an eleventh aspect of the present disclosure, the processing state estimation device according to any of the fourth to tenth aspects is provided, wherein the processor sets the estimated wear parameter to an initial value when it receives a signal indicating that the tool has been replaced or when it receives a signal indicating that the tool has been polished.

[0028] According to a twelfth aspect of this disclosure, the processor sets the estimated wear parameter to an initial value and then, until a predetermined period of time has elapsed, The estimated wear parameters are kept at their initial values. The present invention provides a processing state estimation apparatus according to an eleventh embodiment, wherein, in the process of searching for the aforementioned reference data, the clearance parameter is sequentially changed to search for the reference data from among the plurality of reference data that has the greatest similarity to the measurement data.

[0029] According to a thirteenth aspect of this disclosure, each of the plurality of parameters is a variable selected from a plurality of discrete values, The processor generates complementary reference data corresponding to the parameters between the plurality of discrete values ​​based on the plurality of reference data, In the search process, the processor searches for the reference data or complementary reference data from among the plurality of reference data and complementary reference data that has the greatest similarity to the measurement data. The present invention provides a processing state estimation device according to any of the first to twelfth embodiments.

[0030] According to a fourteenth aspect of this disclosure, each of the plurality of reference data shows the relationship between time and machining load during machining by the machining machine, The aforementioned measurement data shows the relationship between time and processing load during processing by the processing machine as measured. The present invention provides a processing state estimation device according to any of the first to thirteen embodiments.

[0031] According to a 15th aspect of this disclosure, each of the plurality of reference data indicates the relationship between the sliding position of the processing machine and the processing load during processing by the processing machine, The aforementioned measurement data shows the relationship between time and processing load during processing by the processing machine as measured. The storage device further stores time-slide position relationship data that defines the relationship between time and the slide position, In the process of searching for the reference data, the processor Based on the aforementioned time-slide position relationship data, the plurality of reference data are converted into a plurality of secondary reference data each showing the relationship between time and machining load during machining by the machining machine, The system searches for the secondary reference data that has the greatest similarity to the measurement data from among the multiple secondary reference data. The present invention provides a processing state estimation device according to any of the first to thirteen embodiments.

[0032] According to a sixteenth aspect of this disclosure, each of the plurality of reference data indicates the relationship between the sliding position of the processing machine and the processing load during processing by the processing machine, The aforementioned measurement data shows the relationship between time and processing load during processing by the processing machine as measured. The storage device further stores time-slide position relationship data that defines the relationship between time and the slide position, In the process of searching for the reference data, the processor Based on the aforementioned time-slide position relationship data, the measurement data is converted into transformed measurement data that shows the relationship between the slide position and the processing load during processing by the processing machine. By searching for the reference data with the greatest similarity to the converted measurement data from among the multiple reference data, the system searches for the reference data with the greatest similarity to the measurement data from among the multiple reference data. The present invention provides a processing state estimation device according to any of the first to thirteen embodiments.

[0033] According to one aspect of this disclosure, the processor includes the steps of acquiring measurement data showing the measurement result of the machining load by the machining machine, The processor performs the step of determining a similarity, which is an indicator of the degree of similarity between a plurality of reference data corresponding to each combination of a plurality of parameters defining the machining state of a machining tool and the measurement data, The processor estimates the processing state based on the determined similarity, The present invention provides a method for estimating the processing state, including [specific details omitted].

[0034] Embodiments of this disclosure will be described in detail below, with reference to the drawings as appropriate. However, unnecessary details may be omitted. For example, detailed explanations of already well-known matters or redundant explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily verbose and to facilitate understanding by those skilled in the art. The inventors provide the accompanying drawings and the following explanation so that those skilled in the art can fully understand this disclosure, and do not intend to limit the subject matter described in the claims by means of these.

[0035] (First Embodiment) [1. Structure] Figure 1 is a block diagram showing an example configuration of a machining state estimation device 100 according to the first embodiment of the present disclosure. The machining state estimation device 100 comprises a CPU 1, a storage device 2, an input interface (I / F) 3, and an output interface (I / F) 4.

[0036] CPU1 performs information processing to realize the functions of the processing state estimation device 100, which will be described later. Such information processing is realized, for example, by CPU1 operating according to the instructions of program 21 stored in memory device 2. CPU1 is an example of a processor of this disclosure. The processor does not need to include an arithmetic circuit that performs calculations for information processing, and is not limited to a CPU. For example, the processor may be composed of circuits such as an MPU or FPGA.

[0037] The storage device 2 is a recording medium that stores various information, including data such as the waveform library 23 and state data 22 described later, and the program 21 necessary to realize the functions of the processing state estimation device 100. The storage device 2 can be implemented, for example, as a semiconductor storage device such as flash memory or 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 thereof. The storage device 2 may also include volatile memory such as SRAM or DRAM.

[0038] The input interface 3 is an interface circuit that connects the machining state estimation device 100 to an external device in order to input information such as the detection results from the load sensor 11 to the machining state estimation device 100. Such an external device may be, for example, the load sensor 11 or other information processing terminals. The input interface 3 may also be a communication circuit that performs data communication according to an existing wired communication standard or wireless communication standard.

[0039] The output interface 4 is an interface circuit that connects the machining state estimation device 100 to an external output device in order to output information from the machining state estimation device 100. Such an output device may be, for example, a display or another information processing terminal. The output interface 4 may also be a communication circuit that performs data communication according to an existing wired communication standard or wireless communication standard. The input interface 3 and the output interface 4 may be implemented by similar hardware.

[0040] Figure 2 is a schematic cross-sectional view showing a press machine 50 to which the load sensor 11 shown in Figure 1 is attached. The press machine 50 is an example of a processing machine that performs cyclic processing, repeating the same processing. The press machine 50 comprises a bolster 51 and a slide 52 that repeatedly performs an up-and-down cyclic motion from top dead center to bottom dead center relative to the bolster 51. A die backing plate 61 is mounted on the bolster 51, and a die plate 62 is mounted on the die backing plate 61. The die plate 62 grips the die 63.

[0041] A punch backing plate 71 is attached to the lower part of the slide 52, and a punch plate 72 is attached to the lower part of the punch backing plate 71. The punch plate 72 grips the punch 73. The press machine 50 further includes a stripper plate 74. The stripper plate 74 is attached to the punch plate 72 or punch backing plate 71 and fasteners such as bolts via positioning guides such as posts (not shown). The stripper plate 74 is biased downward, for example by a compression spring, and has the function of guiding the punch 73 to stay in a constant position, as well as the function of removing material adhering to the punch 73 after punching the workpiece 80, and / or the function of fixing the workpiece 80 when punching the workpiece 80.

[0042] The load sensor 11 is installed, for example, between the punch 73 and the punch backing plate 71. The load sensor 11 is an electrical force sensor, such as a piezoelectric force sensor or a strain gauge type, and measures the load applied to the punch 73 when the punch 73 punches out the workpiece 80.

[0043] [2. Operation] [2-1. Overall Flow] Figure 3 is a flowchart illustrating the procedure for estimating the machining state, which is performed by the CPU 1 of the machining state estimation device 100 shown in Figure 1.

[0044] First, the CPU 1 acquires a measurement waveform from the load sensor 11 that shows the measurement result of the load applied to the load sensor 11 during press processing by the press machine 50 (S1).

[0045] Figure 4 is a schematic graph showing an example of a measurement waveform from the load sensor 11. In the graph of Figure 4, the horizontal axis represents time, and the vertical axis represents load. The graph of Figure 4 shows a bell-shaped waveform in which, during punching, a load is applied to the workpiece 80, and therefore to the punch 73 and load sensor 11, from the moment the punch 73 moves down and contacts the workpiece 80, and then the load rapidly decreases to almost zero after the workpiece 80 is punched out. The punching period of the punching process can be measured, for example, as the period from the time when the load exceeds the rising threshold to the time when it falls below the falling threshold in the measurement waveform. Such rising and falling thresholds may be defined as absolute values ​​or as a percentage of the peak value of the load.

[0046] Returning to Figure 3, after step S1, CPU1 acquires state data 22, which represents the estimated parameter set, which is the result of the previous estimation (S2).

[0047] Figure 5 is a table showing an example of state data 22. State data 22 includes parameters that define the state of machining. In this embodiment, as shown in Figure 5, such parameters define the state of the tool, and more specifically, punch wear, die wear, and clearance.

[0048] Returning to Figure 3, after step S2, the CPU1 determines whether a predetermined period has elapsed since the tool change was performed (S3). For example, the CPU1 determines whether a predetermined period has elapsed since receiving a tool change signal indicating that a tool change has been performed. The CPU1 may also determine that a predetermined period has elapsed if press work has been performed a predetermined number of times or more since receiving the tool change signal. Such a tool change signal is transmitted to the CPU1, for example, when the user presses a tool change completion button provided on the user interface of the press machine 50, the processing state estimation device 100, etc.

[0049] If it is determined that a predetermined period has elapsed since the tool was replaced (Yes in S3), the CPU1 determines whether a predetermined period has elapsed since the tool was sharpened (S4). For example, the CPU1 determines whether a predetermined period has elapsed since receiving a die sharpening signal indicating that the die has been sharpened and / or a punch sharpening signal indicating that the punch has been sharpened. The CPU1 may also determine that a predetermined period has elapsed if press operations have been performed a predetermined number of times or more since receiving the die sharpening signal and / or punch sharpening signal. Such signals are transmitted to the CPU1, for example, when the user presses a die sharpening completion button and / or punch sharpening completion button provided on the user interface of the press machine 50, the processing state estimation device 100, etc.

[0050] If it is determined that a predetermined period has elapsed since the tool was sharpened (Yes in S4), CPU1 executes the first state estimation process (hereinafter referred to as the "normal state estimation process") S5. Details of the normal state estimation process S5 will be described later.

[0051] In step S4, if it is determined that a predetermined period has not elapsed since the tool was sharpened (No in S4), CPU1 executes the second state estimation process (hereinafter referred to as the "post-sharpening state estimation process") S6. Details of the post-sharpening state estimation process S6 will be described later.

[0052] In step S3, if it is determined that a predetermined period has not elapsed since the tool was replaced (No in S3), CPU1 executes a third state estimation process (hereinafter referred to as the "post-replacement state estimation process") S7. Details of the post-replacement state estimation process S7 will be described later.

[0053] [2-2. Normal state estimation process S5] Figure 6 is a flowchart illustrating the processing flow of the normal state estimation process S5 shown in Figure 3.

[0054] In the normal state estimation process S5, CPU1 executes the punch wear estimation process S51, and then the die wear estimation process S52. This order of execution is used because, generally, punch wear progresses faster than die wear, so the punch wear amount is estimated preferentially over the die wear amount.

[0055] In the normal state estimation process S5, the clearance value of the state data 22 is fixed to the value estimated in the previous machining state estimation process. The reason for fixing the clearance is that in the normal state estimation process S5, where a predetermined period has elapsed since tool replacement or grinding, the clearance does not change at all or hardly at all even if press working is repeated.

[0056] Figure 7 is a flowchart illustrating the punch wear amount estimation process S51 in Figure 6. In the punch wear amount estimation process S51, the CPU 1 first calculates the degree of agreement between the reference waveform (reference data) corresponding to the value of each parameter in the state data 22 and the measured waveform acquired in step S1 (S511).

[0057] Here, the degree of agreement is an indicator that shows the degree of agreement between two waveforms. The degree of agreement can be, for example, the cosine similarity, Euclidean distance, or Manhattan distance between two waveforms during the punching period. CPU1 may calculate a loss instead of the degree of agreement, which is an indicator that shows the degree of mismatch between the two waveforms. Both the degree of agreement and the degree of mismatch are examples of "similarity," which is an indicator that shows the degree of similarity between two waveforms.

[0058] The reference waveforms corresponding to the values ​​of each parameter in the state data 22 in step S511 are pre-registered in the waveform library 23. The waveform library 23 has pre-registered reference waveforms corresponding to all combinations of candidate values ​​for clearance, die wear amount, and punch wear amount. In this embodiment, the reference waveform is a waveform that represents the relationship between time and load, similar to the measurement waveform in Figure 4.

[0059] The clearance can be set to any of the following candidate values, for example: 3 μm, 4 μm, 5 μm, 6 μm, or 7 μm. The die wear amount can be set to any of the following candidate values, for example: 0 μm, 2 μm, 4 μm, 6 μm, 8 μm, 10 μm, or 12 μm. The punch wear amount can be set to any of the following candidate values, for example: 0 μm, 2 μm, 4 μm, 6 μm, 8 μm, 10 μm, or 12 μm. Note that the candidate values ​​for clearance, die wear amount, and punch wear amount are not limited to these, and the number of candidate values ​​is not limited to the above numbers.

[0060] As in the example above, if there are 5 candidate values ​​for clearance, 7 candidate values ​​for die wear, and 7 candidate values ​​for punch wear, then 245 reference waveforms are pre-registered in the waveform library 23. In this way, the waveform library 23 is a three-dimensional table in which reference waveforms corresponding to arrays of clearance, die wear, and punch wear are registered.

[0061] In step S512 of Figure 7, CPU1 determines whether the loop processing in punch wear amount estimation process S51 has converged (completed) (S512). Convergence means that all candidate values ​​that can be selected based on predetermined selection rules have been set in the provisional state data. In step S512, CPU1 makes a convergence determination by determining whether all candidate values ​​for punch wear amount that are greater than the punch wear amount estimated in the previous machining state estimation process have been set in the provisional state data.

[0062] If CPU1 determines in step S512 that the loop processing in punch wear amount estimation process S51 has not converged (No in S512), it executes step S513. If it determines that it has converged (Yes in S512), it terminates punch wear amount estimation process S51.

[0063] In step S513, CPU1 prepares temporary state data by modifying state data 22 to set the punch wear amount to a value greater than the punch wear amount estimated previously (S513). In the above example, if the punch wear amount estimated previously was 4 μm, CPU1 sets the punch wear amount in the temporary state data to one of 6 μm, 8 μm, 10 μm, or 12 μm. In step S513, the other parameters of the temporary state data, clearance and die wear amount, are fixed to the clearance and die wear amounts estimated previously, respectively.

[0064] Next, CPU1 calculates the degree of agreement between the reference waveform in the waveform library 23 corresponding to the value of each parameter in the provisional state data and the measured waveform acquired in step S1 (S514).

[0065] Next, CPU1 determines whether the degree of match calculated in step S514 has increased compared to the degree of match calculated in the most recent step S511 (S515). If CPU1 determines that the degree of match has increased (Yes in S515), it proceeds to step S516. If it determines that the degree of match has not increased (No in S515), it returns to step S512.

[0066] In step S516, CPU1 updates the state data 22 so that the temporary state data prepared in step S513 becomes the state data 22 (S516). After completing step S516, CPU1 returns to step S511.

[0067] As described above, if CPU1 determines in step S512 that the loop processing in punch wear amount estimation process S51 has converged (Yes in S512), it finishes punch wear amount estimation process S51 and executes die wear amount estimation process S52 (see Figure 6). In the above example, CPU1 finishes punch wear amount estimation process S51 when it has completed all loops in which the punch wear amount of the provisional state data is set to 6 μm, 8 μm, 10 μm, and 12 μm.

[0068] Figure 8 is a flowchart illustrating the die wear amount estimation process S52 in Figure 6. In the die wear amount estimation process S52, the CPU 1 first calculates the degree of agreement between the reference waveform corresponding to the value of each parameter in the state data 22 and the measured waveform acquired in step S1 (S521). Step S521 is the same process as step S511 in Figure 7.

[0069] Next, CPU1 determines whether the loop processing in die wear amount estimation process S52 has converged (S522). That is, CPU1 determines whether all candidate values ​​for die wear amount that are greater than the die wear amount estimated in the previous machining state estimation process have been set in the provisional state data. If CPU1 determines that the loop processing in die wear amount estimation process S52 has not converged (No in S522), it executes step S523. If it determines that it has converged (Yes in S522), it terminates die wear amount estimation process S52.

[0070] In step S523, CPU1 prepares provisional state data by modifying state data 22 to set the die wear amount to a value greater than the die wear amount estimated previously (S523). In step S523, the clearance, which is another parameter of the provisional state data, is fixed to the clearance estimated previously, and the punch wear amount is fixed to the punch wear amount determined in step S516 in Figure 7.

[0071] Next, CPU1 calculates the degree of agreement between the reference waveform in the waveform library 23 corresponding to the value of each parameter in the provisional state data and the measured waveform acquired in step S1 (S524).

[0072] Next, CPU1 determines whether the degree of match calculated in step S524 has increased compared to the degree of match calculated in the most recent step S521 (S525). If CPU1 determines that the degree of match has increased (Yes in S525), it proceeds to step S526. If it determines that the degree of match has not increased (No in S525), it returns to step S522.

[0073] In step S526, CPU1 updates the state data 22 so that the temporary state data prepared in step S523 becomes the state data 22 (S526). After completing step S526, CPU1 returns to step S521.

[0074] As described above, in the normal state estimation process S5, the CPU1 estimates the state data 22 by fixing the clearance and prioritizing the punch wear amount estimation process S51 over the die wear amount estimation process S52.

[0075] [2-3. Post-polishing state estimation process S6] Figure 9 is a flowchart illustrating the processing flow of the post-polishing state estimation process S6 shown in Figure 3. In the post-polishing state estimation process S6, the processing performed differs depending on whether the polished tool is a punch, a die, or both.

[0076] For example, CPU1 determines whether both the punch and the die have been polished (S61). In the example above, CPU1 determines whether it has received both a die polishing signal indicating that the die has been polished and a punch polishing signal indicating that the punch has been polished. If CPU1 determines that both the punch and the die have been polished (Yes in S61), it proceeds to step S62; otherwise (No in S61), it proceeds to step S64.

[0077] In step S64, CPU1 determines whether the punch has been polished or not. If CPU1 determines that the punch has been polished (Yes in S64), it proceeds to step S65; otherwise (No in S64), it proceeds to step S66. In other words, if both the punch and the die have been polished, step S62 is executed; if only the punch has been polished, step S65 is executed; and if only the die has been polished, step S66 is executed.

[0078] In step S62, CPU1 sets the punch wear amount and die wear amount to the initial value of 0 μm. With the punch wear amount and die wear amount fixed in this way, CPU1 executes a clearance estimation process (hereinafter referred to as the "clearance estimation process after polishing") S63. Details of the clearance estimation process after polishing S63 will be described later.

[0079] If it is determined in step S64 that the punch has been polished (Yes in S64), the CPU1 sets the punch wear amount to the initial value of 0 μm (S65), and then executes the clearance estimation process after polishing S63 and the die wear amount estimation process S52.

[0080] If it is determined in step S64 that the punch has not been polished (No in S64), the CPU1 sets the die wear amount to the initial value of 0 μm (S66), and then executes the clearance estimation process after polishing S63 and the punch wear amount estimation process S51.

[0081] Figure 10 is a flowchart illustrating the processing flow of the clearance estimation process S63 after polishing, as shown in Figure 9.

[0082] In the clearance estimation process S63 after polishing, the CPU 1 first calculates the degree of agreement between the reference waveform corresponding to the value of each parameter in the state data 22 and the measured waveform acquired in step S1 (S631). Step S631 is the same process as step S511 in Figure 7 and step S521 in Figure 8.

[0083] Next, CPU1 determines whether the loop processing in the clearance estimation process S63 after polishing has converged (S632). That is, CPU1 determines whether all candidate clearance values ​​that are within a predetermined range from the clearance estimated in the previous machining state estimation process have been set in the provisional state data. If CPU1 determines that the loop processing in the clearance estimation process S63 after polishing has not converged (No in S632), it executes step S633. If it determines that it has converged (Yes in S632), it terminates the clearance estimation process S63 after polishing.

[0084] In step S633, CPU1 prepares temporary state data by modifying state data 22 to set the clearance to a value within a predetermined range from the previously estimated clearance (S633). For example, if the previously estimated clearance was 5 μm, CPU1 sets the clearance in the temporary state data to a value within ±1 μm of 5 μm, i.e., 4 μm or 6 μm. The reason for limiting the range of clearance change to a predetermined range is that, unlike when the tool is replaced, it has been found that the clearance hardly changes even when the tool is ground.

[0085] Next, CPU1 calculates the degree of agreement between the reference waveform in the waveform library 23 corresponding to the value of each parameter in the provisional state data and the measured waveform acquired in step S1 (S634).

[0086] Next, CPU1 determines whether the degree of match calculated in step S634 has increased compared to the degree of match calculated in the most recent step S631 (S635). If CPU1 determines that the degree of match has increased (Yes in S635), it proceeds to step S636. If it determines that the degree of match has not increased (No in S635), it returns to step S632.

[0087] In step S636, CPU1 updates the state data 22 so that the temporary state data prepared in step S633 becomes the state data 22 (S636). After completing step S636, CPU1 returns to step S631.

[0088] [2-4. State estimation process after exchange S7] Figure 11 is a flowchart illustrating the processing flow of the state estimation process S7 after the exchange shown in Figure 3.

[0089] In the post-replacement state estimation process S7, the CPU1 first sets the punch wear amount and die wear amount to the initial value of 0 μm (S62).

[0090] Next, CPU1 calculates the degree of agreement between the reference waveform corresponding to the value of each parameter in the state data 22 and the measured waveform acquired in step S1 (S71).

[0091] Next, CPU1 determines whether the loop processing in the post-exchange state estimation process S7 has converged (S72). That is, CPU1 determines whether all candidate clearance values ​​have been set in the provisional state data. If CPU1 determines that the loop processing in the post-exchange state estimation process S7 has not converged (No in S72), it executes step S73. If it determines that it has converged (Yes in S72), it terminates the post-exchange state estimation process S7.

[0092] In step S73, CPU1 prepares temporary state data by modifying state data 22 to set the clearance to one of the candidate clearance values ​​(S73).

[0093] Next, CPU1 calculates the degree of agreement between the reference waveform in the waveform library 23 corresponding to the value of each parameter in the provisional state data and the measured waveform acquired in step S1 (S74).

[0094] Next, CPU1 determines whether the degree of match calculated in step S74 has increased compared to the degree of match calculated in the most recent step S71 (S75). If CPU1 determines that the degree of match has increased (Yes in S75), it proceeds to step S76; if it determines that the degree of match has not increased (No in S75), it returns to step S72.

[0095] In step S76, CPU1 updates state data 22 so that the temporary state data prepared in step S73 becomes state data 22 (S76). After completing step S76, CPU1 returns to step S71.

[0096] The machining state estimation device 100 may notify the user if, in the estimated state data 22, the punch wear amount or die wear amount is above a predetermined threshold, and / or the clearance is not within a predetermined range. This allows the user to perform maintenance such as changing tools. Such notification may be performed by means of, for example, lighting or flashing an LED in red, generating a warning sound from a speaker, or displaying the state data 22 on a display.

[0097] [3. Effects, etc.] As described above, the processing state estimation device 100 according to this embodiment comprises a storage device 2 and a CPU 1, which is an example of a processor. The storage device 2 stores state data 22, which are a plurality of parameters that define the processing state of the press machine 50, and a plurality of reference waveforms, each corresponding to a combination of the plurality of parameters. The CPU 1 acquires a measurement waveform, which is an example of measurement data showing the measurement result of the processing load by the press machine 50 (S1). The CPU 1 searches for a reference waveform from among the plurality of reference waveforms that has the maximum degree of agreement, which is an indicator of the degree of agreement with the measurement waveform, and determines the combination of parameters corresponding to the searched reference waveform as an estimation parameter set that represents the processing state at the time of measurement of the measurement waveform (S5).

[0098] Conventionally, there is a known technique that determines a measurement waveform to be normal if it falls within a predetermined range between a set upper and lower limit, and abnormal if it falls outside that range. However, with conventional techniques, if the predetermined range is set too wide, it cannot detect abnormalities in the device such as tool wear, and if it is set too narrow, it may determine that the device is abnormal even though it is functioning normally. In contrast, the machining state estimation device 100 according to this embodiment, which searches for a reference waveform, can estimate the machining state with greater accuracy than conventional techniques.

[0099] In the process of searching for a reference waveform, CPU1 may sequentially change multiple parameters within a predetermined range based on an estimated parameter set already determined by CPU1, and search for a reference waveform from among a portion of multiple reference waveforms that has the greatest degree of agreement with the measured waveform.

[0100] This configuration allows for a more accurate estimation of the processing state by performing a search based on the above criteria. Furthermore, it reduces the computational cost of estimation compared to a configuration without these criteria.

[0101] Multiple parameters may include wear parameters that define the degree of tool wear of the press machine 50 and clearance parameters that define the clearance of the press machine 50. The estimated parameter set may include estimated wear parameters estimated as wear parameters at the time of measurement of the measurement waveform and estimated clearance parameters estimated as clearance parameters at the time of measurement of the measurement waveform.

[0102] This configuration allows for more accurate estimation of tool wear and clearance than conventional techniques.

[0103] In the process of searching for a reference waveform, CPU1 sequentially changes the wear parameter within a range of values ​​greater than or equal to the estimated wear parameter to search for a reference waveform that has the greatest degree of agreement with the measured waveform.

[0104] With this configuration, the amount of tool wear and clearance can be estimated with greater accuracy by exploring a reference waveform while varying the wear parameter under the condition that it is greater than or equal to the estimated wear parameter.

[0105] The wear parameters may include a punch wear parameter that defines the degree of wear of the punch of the press machine 50, and a die wear parameter that defines the degree of wear of the die of the press machine 50. The estimated wear parameters may include an estimated punch wear parameter estimated as the punch wear parameter at the time of measurement of the measurement waveform, and an estimated die wear parameter estimated as the die wear parameter at the time of measurement of the measurement waveform. In the process of searching for a reference waveform, the CPU 1 sequentially changes the punch wear parameter within a value greater than or equal to the estimated punch wear parameter, and sequentially changes the die wear parameter within a value greater than or equal to the estimated die wear parameter, in order to search for a reference waveform that has the greatest degree of agreement with the measurement waveform.

[0106] With this configuration, the amount of tool wear and clearance can be estimated with greater accuracy by searching for a reference waveform while varying the punch wear parameter and die wear parameter under the condition that they are greater than or equal to already estimated values.

[0107] CPU1 may perform the following processing in the process of searching for a reference waveform. Specifically, CPU1 fixes the die wear parameter to the estimated die wear parameter and sequentially changes the punch wear parameter within a value greater than or equal to the estimated punch wear parameter to search for a reference waveform that has the maximum first degree of agreement with the measured waveform, and determines the punch wear parameter corresponding to the waveform with the maximum first degree of agreement as the estimated punch wear parameter (S51). Next, CPU1 fixes the punch wear parameter to the determined estimated punch wear parameter and sequentially changes the die wear parameter within a value greater than or equal to the estimated die wear parameter to search for a reference waveform that has the maximum second degree of agreement with the measured waveform, and determines the die wear parameter corresponding to the waveform with the maximum second degree of agreement as the estimated die wear parameter (S52).

[0108] Generally, punch wear progresses faster than die wear. Therefore, by prioritizing the estimation of punch wear over die wear, tool wear and clearance can be estimated with greater accuracy.

[0109] CPU1 may set the estimated wear parameter to an initial value when it receives a signal indicating that the tool has been replaced or when it receives a signal indicating that the tool has been sharpened (S62, S65, S66).

[0110] This configuration allows for more accurate estimation of the machining state. Furthermore, since the estimated wear parameters are set to initial values, the computational load for search and estimation can be reduced.

[0111] After setting the estimated wear parameters to their initial values, CPU1 may keep the estimated wear parameters at their initial values ​​until a predetermined period has elapsed, and in the process of searching for a reference waveform, sequentially change the clearance parameters to search for a reference waveform from among multiple reference waveforms that has the greatest degree of agreement with the measured waveform.

[0112] For a predetermined period after a tool is replaced or sharpened, wear can be approximated to its initial value. Therefore, this configuration allows for accurate estimation of the machining state. Furthermore, since the estimated wear parameter is set to an initial value, the computational load for search and estimation can be reduced.

[0113] (Modification of the first embodiment) In the first embodiment, as shown in Figure 5, an example was described in which the state data 22 consists of punch wear amount, die wear amount, and clearance, which represent the state of the tool. However, the parameters included in the state data 22 are not limited to these. For example, as shown in Figure 12, the state data 22 may also include workpiece state data that indicates the state of the workpiece, such as the workpiece thickness.

[0114] The workpiece thickness can be set to any of the following candidate values, for example, 46 μm, 48 μm, 50 μm, 52 μm, or 54 μm, but is not limited to these, and the number of candidate workpiece thickness values ​​is not limited to seven.

[0115] In this way, when the state data 22 includes workpiece state data in addition to tool state data, the waveform library 23 becomes a four-dimensional table in which waveforms corresponding to the arrangement of clearance, die wear amount, punch wear amount, and workpiece thickness are registered.

[0116] Figure 13 is a flowchart illustrating the processing flow of a normal state estimation process S5a in a modified version of the first embodiment. Compared with the normal state estimation process S5 shown in Figure 6, in the normal state estimation process S5a, the workpiece thickness estimation process S50 is executed before the punch wear amount estimation process S51. The reason why the workpiece thickness estimation process S50 is executed in priority over the punch wear amount estimation process S51 and the die wear amount estimation process S52 is that it has been found that the workpiece thickness is a parameter that usually changes each time the workpiece is replaced.

[0117] Figure 14 is a flowchart illustrating the processing flow of the workpiece thickness estimation process S50 shown in Figure 13.

[0118] In the workpiece thickness estimation process S50, the CPU 1 first calculates the degree of agreement between the waveform corresponding to the value of each parameter in the state data 22 and the measurement waveform acquired in step S1 (S501).

[0119] Next, CPU1 determines whether the loop processing in work thickness estimation process S50 has converged (S502). That is, CPU1 determines whether all candidate values ​​for work thickness have been set in the provisional state data. If CPU1 determines that the loop processing in work thickness estimation process S50 has not converged (No in S502), it executes step S503. If it determines that it has converged (Yes in S502), it terminates work thickness estimation process S50.

[0120] In step S503, CPU1 modifies the state data 22 to set the workpiece thickness to one of the candidate values ​​for workpiece thickness, thereby preparing temporary state data (S503).

[0121] Next, CPU1 calculates the degree of agreement between the waveforms in the waveform library 23 corresponding to the values ​​of each parameter in the provisional state data and the measured waveforms acquired in step S1 (S504).

[0122] Next, CPU1 determines whether the degree of match calculated in step S504 has increased compared to the degree of match calculated in the most recent step S501 (S505). If CPU1 determines that the degree of match has increased (Yes in S505), it proceeds to step S506. If it determines that the degree of match has not increased (No in S505), it returns to step S502.

[0123] In step S506, CPU1 updates state data 22 so that the temporary state data prepared in step S503 becomes state data 22 (S506). After completing step S506, CPU1 returns to step S501.

[0124] In this modified example, the workpiece thickness estimation process S50 is executed in the post-polishing state estimation process S6 and the post-replacement state estimation process S7, prioritizing it over the estimation of other parameters, that is, prioritizing it over the clearance estimation process in the example shown in the first embodiment.

[0125] As described above, this modified method allows for more accurate estimation of the workpiece thickness than conventional methods.

[0126] (Second Embodiment) Figure 15 is a block diagram showing an example configuration of a machining state estimation device 200 according to a second embodiment of the present disclosure. Compared with the machining state estimation device 100 in Figure 1, in the machining state estimation device 200, the storage device 2 further stores a complementary waveform library 24 and a complementary processing setting value 25. The complementary waveform library 24 is generated by the CPU 1 based on the waveform library 23 and the complementary processing setting value 25.

[0127] Figure 16 is a schematic diagram illustrating the overview of the generation process of the interpolation waveform library 24 (hereinafter referred to as "interpolation process") performed by the CPU 1 of the machining state estimation device 200 shown in Figure 15. Each parameter of the state data 22 can take on multiple discrete values ​​within a predetermined range. In the example given in the first embodiment, the candidate values ​​for punch wear amount and die wear amount are one of 0 μm, 2 μm, 4 μm, 6 μm, 8 μm, 10 μm, and 12 μm, and the candidate values ​​for clearance are one of 3 μm, 4 μm, 5 μm, 6 μm, and 7 μm. In the example given in the modified example of the first embodiment, the candidate values ​​for workpiece thickness are one of 46 μm, 48 μm, 50 μm, 52 μm, and 54 μm. There are 1225 possible combinations of these four parameters, and waveforms corresponding to each of the 1225 combinations are registered in the waveform library 23.

[0128] As illustrated in Figure 16, the interpolation settings 25 define, for example, the minimum, maximum, and pitch values ​​for each parameter. The interpolation settings 25 may also define only the pitch value for each parameter.

[0129] CPU1 generates interpolated waveforms (interpolated reference data) corresponding to the interpolated values ​​between candidate parameter values, based on the waveform library 23 and the interpolation processing settings 25. In other words, CPU1 generates interpolated waveforms that are not pre-registered in the waveform library 23, based on the waveform library 23 and the interpolation processing settings 25.

[0130] Figure 17 is a flowchart illustrating the procedure of the interpolation process performed by the CPU 1 of the machining state estimation device 200 shown in Figure 15. First, the CPU 1 obtains the waveform library 23 and the interpolation process setting value 25 from the storage device 2 (S21).

[0131] Next, CPU1 generates an interpolated waveform corresponding to the combination of parameters specified in the interpolation processing setting value 25 (S22), and stores the generated interpolated waveform in memory device 2 (S23).

[0132] Figure 18 is a schematic diagram illustrating an example of interpolation processing. In Figure 18, the horizontal axis represents punch wear, and the vertical axis represents die wear. As in the example above, when the punch wear and die wear can take values ​​of 0 μm, 2 μm, 4 μm, 6 μm, 8 μm, 10 μm, and 12 μm, the waveform library 23 registers waveforms corresponding to the parameters indicated by the black circles in Figure 18.

[0133] On the other hand, the waveform library 23 does not contain a waveform corresponding to point M1 where the punch wear is 1.5 μm and the die wear is 2.0 μm. The CPU 1 synthesizes the waveforms in the waveform library 23 corresponding to points P1 and P2 located near point M1 with respect to punch wear, respectively, to generate an intermediate waveform corresponding to point M1.

[0134] Such synthesis is performed, for example, by calculating a weighted average of the waveform corresponding to point P1 and the waveform corresponding to point P2. For example, in the example in Figure 18, the distance between point M1 and point P1 is three times the distance between point M1 and point P2, so a weighted average is taken by calculating the average of the waveform corresponding to point P1 and the waveform corresponding to point P2, which is weighted three times more than the waveform corresponding to point P1.

[0135] Similarly, CPU1 can synthesize waveforms in waveform library 23 corresponding to points P3 and P4 located near point M2 with respect to punch wear to generate an intermediate waveform corresponding to point M2. Furthermore, CPU1 can synthesize intermediate waveforms corresponding to points M1 and M2 to generate a complementary waveform corresponding to point C1.

[0136] Using the interpolated waveform generation method described above, CPU1 can generate interpolated waveforms corresponding to all the intersection points of the straight lines shown as solid and dashed lines in Figure 18.

[0137] Figure 18 shows a two-dimensional graph to facilitate understanding of the explanation, but the interpolation process can also be applied to waveforms with three or more dimensions. In the example shown in Figure 16, there are 1125 possible combinations of values ​​for the four parameters in the waveform library 23, but each parameter can be interpolated at the pitch set by the interpolation setting value 25. As a result, the total number of waveforms in the waveform library 23 and the interpolated waveform library 24 becomes 230,625.

[0138] As described above, in the processing state estimation device 200 according to this embodiment, each of the multiple parameters may be a variable selected from a plurality of discrete values. The CPU 1 generates an interpolated waveform corresponding to the parameter between the plurality of discrete values ​​based on the plurality of reference waveforms. In the search process, the CPU 1 searches for a reference waveform or interpolated waveform from among the plurality of reference waveforms and interpolated waveforms that has the greatest degree of agreement with the measured waveform.

[0139] This configuration allows for more accurate estimation of the machining state by searching for the waveform that best matches the measured waveform, not only from multiple reference waveforms but also from complementary waveforms.

[0140] Furthermore, the machining state estimation device 200 according to this embodiment can supplement the complementary waveform library 24 from a waveform library 23 in which a limited number of reference waveforms are registered. Reference waveforms are created based on actual measurements or simulations, but creating waveforms that correspond to many combinations of state values ​​would require a huge number of combinations, thus incurring significant man-hours and costs. Therefore, by creating a limited number of reference waveforms that correspond to representative combinations of state values ​​and generating complementary waveforms from these reference waveforms, it is possible to maintain or improve estimation accuracy while reducing man-hours and costs.

[0141] (Third embodiment) Figure 19 is a block diagram showing an example configuration of a machining state estimation device 300 according to the third embodiment of this disclosure. Compared with the machining state estimation device 100 in Figure 1, in the machining state estimation device 300, the storage device 2 stores a waveform library 323 instead of a waveform library 23, and also stores time-slide position relationship data 26. In addition, the machining state estimation device 300 acquires measurement results from the slide position sensor 12 via the input interface 3.

[0142] The slide position sensor 12 is installed, for example, between the slide 52 and the bolster 51 of the press machine 50 in Figure 2, and measures the position of the slide 52. The slide position sensor 12 is, for example, an eddy current displacement sensor or a laser displacement sensor, but is not limited to these.

[0143] In the first embodiment, the waveforms pre-registered in the waveform library 23 represent the relationship between time and press load, whereas in this embodiment, the waveforms pre-registered in the waveform library 323 represent the relationship between the slide position and press load during processing by the press machine 50.

[0144] The time-slide position relationship data 26 defines the relationship between time and slide position during processing by the press machine 50. Figure 20 is a schematic graph showing an example of the time-slide position relationship data 26 in Figure 19. In the graph of Figure 20, the horizontal axis represents time, and the vertical axis represents slide position. In the graph of Figure 20, the relationship between time and slide position in the first press machine 50 is shown by a solid line, and the relationship between time and slide position in the second press machine 50, which is different from the first press machine 50, is shown by a dashed line.

[0145] As shown in Figure 20, even if multiple press machines 50 are set to perform the same sliding motion, the movement of the slide 52 may differ from machine to machine. By registering waveforms showing the relationship between the slide position and press load during processing in the waveform library 323, the differences between machines can be reduced. Therefore, even if there is a change in the processing state, such as when the press machine 50 is changed, the processing state can be estimated stably and with high accuracy using the same waveform library 323.

[0146] Compared to the first embodiment, in this embodiment, CPU1 executes punch wear amount estimation process S51a instead of punch wear amount estimation process S51 in the first embodiment. Figure 21 is a flowchart illustrating punch wear amount estimation process S51a in this embodiment. Compared to punch wear amount estimation process S51 in the first embodiment, in punch wear amount estimation process S51 of this embodiment, CPU1 executes steps S520 and S511a instead of step S511, and steps S521 and S514a instead of step S514.

[0147] In the punch wear amount estimation process S51a, the CPU 1 first performs a unit conversion process on the waveform in the waveform library 323 corresponding to the state data 22 (S520). The unit conversion process converts the waveform in the waveform library 323 into a waveform (secondary reference data) that shows the relationship between time and press load during processing by the press machine 50.

[0148] Figure 22 is a schematic diagram illustrating the unit conversion process in step S520. Based on the time-slide position relationship data 26, the CPU 1 converts the waveforms in the waveform library 323, which show the relationship between slide position and load, into waveforms that show the relationship between time and load.

[0149] Returning to Figure 21, CPU1 calculates the degree of agreement between the converted waveform, which was converted in step S520, and the measured waveform, which was acquired in step S1 (S511a).

[0150] The unit conversion process in step S521 and the matching degree calculation process in step S514a are the same as described above. Furthermore, the other steps in Figure 21 are the same as the punch wear amount estimation process S51 in the first embodiment shown in Figure 7.

[0151] Similarly, in this embodiment, CPU1 performs the unit conversion process and degree of agreement calculation process described above in place of steps S521 and S524, respectively, of the die wear amount estimation process S52 in Figure 8. Also, in this embodiment, CPU1 performs the unit conversion process and degree of agreement calculation process described above in place of steps S631 and S634, respectively, of the clearance estimation process after polishing S63 in Figure 10. Furthermore, in this embodiment, CPU1 performs the unit conversion process and degree of agreement calculation process described above in place of steps S71 and S74, respectively, of the state estimation process after replacement S7 in Figure 11.

[0152] As described above, in the processing state estimation device 300 according to this embodiment, each of the multiple reference waveforms indicates the relationship between the slide position of the press machine 50 and the press load during processing by the press machine 50. The measured waveform indicates the relationship between time and press load during processing by the press machine 50 as measured. The storage device 2 further stores time-slide position relationship data 26 that defines the relationship between time and slide position. In the process of searching for a reference waveform, the CPU 1 converts the multiple reference waveforms into multiple converted waveforms that each indicate the relationship between time and press load during processing by the press machine 50, based on the time-slide position relationship data 26, and searches for the converted waveform that has the greatest degree of agreement with the measured waveform from among the multiple converted waveforms, thereby searching for a reference waveform that has the greatest degree of agreement with the measured waveform from among the multiple reference waveforms.

[0153] With this configuration, even if there is a change in the processing state, such as when the press machine 50 is changed, the processing state can be estimated stably and with high accuracy using the same waveform library 323.

[0154] (Other embodiments) As described above, the above embodiments have been explained as examples of the technology disclosed in this application. However, the technology in this disclosure is not limited to these embodiments and can be applied to embodiments that have been modified, replaced, added, or omitted as appropriate. Therefore, other embodiments will be described below as examples.

[0155] In the third embodiment, an example was described in which the CPU1 converts the waveform in the waveform library 323, which shows the relationship between slide position and load, into a waveform that shows the relationship between time and load, based on the time-slide position relationship data 26. However, in this disclosure, it is sufficient that the waveform in the waveform library 323 can be compared with the measured waveform acquired in step S1, and it is not essential to convert the waveform in the waveform library 323 into a waveform that shows the relationship between time and load. For example, contrary to the conversion in the third embodiment, the CPU1 may convert the measured waveform into a waveform (converted measurement data) that shows the relationship between slide position and press load in processing by the press machine 50, based on the time-slide position relationship data 26. In this case, the degree of agreement is calculated by comparing the waveform in the waveform library 323 corresponding to the value of each parameter of the state data 22 with the converted measurement data.

[0156] In the above embodiment, a press machine 50 was described as an example of a processing machine, but the processing machine is not limited to this. For example, the processing machine may be a molding device. This disclosure is applicable to these processing machines, and the processing state can be estimated based on parameters that affect abnormal processing in these processing machines. [Industrial applicability]

[0157] This disclosure is applicable to processing machines. [Explanation of Symbols]

[0158] 2 Storage device 3. Input Interface 4 Output Interfaces 11. Load sensor 12 Slide position sensor 21 Programs 22 Status Data 23,323 waveform libraries 24 Interpolation Waveform Library 25. Interpolation processing settings 26-hour slide positional relationship data 50 Press Machines 51 Bolster 52 slides 61 Die Backing Plate 62 Die Plates 63 Die 71 Punch backing plate 72 Punch Plates 73 punches 74 Stripper Plate 80 Work 100, 200, 300 Processing state estimation device

Claims

1. Equipped with a memory device and a processor, The aforementioned storage device is Multiple parameters that define the machining state of a machining machine, The system stores multiple reference data corresponding to each of the multiple parameter combinations, The aforementioned processor, Measurement data showing the measurement results of the processing load by the aforementioned processing machine is obtained, The degree of similarity between the aforementioned multiple reference data and the aforementioned measurement data is determined, Based on the determined similarity, the processing state is estimated. Processing state estimation device.

2. The aforementioned processor, From the aforementioned multiple reference data, search for the reference data with the highest similarity. One of the multiple combinations of parameters corresponding to the searched reference data is determined as an estimated parameter set representing the processing state of the measurement data at the time of measurement. The processing state estimation device according to claim 1.

3. The processing state estimation device according to claim 2, wherein the processor, in the process of searching for the reference data, sequentially changes the plurality of parameters within a predetermined range based on the estimated parameter set already determined by the processor, and searches for a portion of the plurality of reference data that has the greatest similarity to the measurement data.

4. The aforementioned multiple parameters are, A wear parameter that defines the degree of wear of the tool of the aforementioned processing machine, Includes a clearance parameter that defines the clearance of the processing machine, The aforementioned set of estimated parameters is: The estimated wear parameter, which was estimated as the wear parameter at the time of measurement of the measurement data, The measurement data includes the estimated clearance parameter estimated as the clearance parameter at the time of measurement, The processing state estimation device according to claim 2 or 3.

5. In the process of searching for the reference data, the processor sequentially changes the wear parameter within a range of values ​​greater than or equal to the estimated wear parameter to search for the reference data that has the greatest similarity to the measurement data. The processing state estimation device according to claim 4.

6. The aforementioned wear parameters are: A punch wear parameter that defines the degree of wear of the punch of the aforementioned processing machine, Includes a die wear parameter that defines the degree of wear of the die of the processing machine, The estimated wear parameters are: The estimated punch wear parameter, which was estimated as the punch wear parameter at the time of measurement of the measurement data, The measurement data includes the estimated die wear parameter, which is estimated as the die wear parameter at the time of measurement. In the process of searching for the reference data, the processor sequentially changes the punch wear parameter within a value greater than or equal to the estimated punch wear parameter, and sequentially changes the die wear parameter within a value greater than or equal to the estimated die wear parameter, in order to search for reference data that has the greatest similarity to the measurement data. The processing state estimation device according to claim 4 or 5.

7. In the process of searching for the reference data, the processor The die wear parameter is fixed to the estimated die wear parameter, and the punch wear parameter is sequentially varied within a value greater than or equal to the estimated punch wear parameter to search for reference data that has the maximum first similarity to the measured data. The punch wear parameter corresponding to the reference data with the highest first similarity is determined as the estimated punch wear parameter. The punch wear parameter is fixed to the determined estimated punch wear parameter, and the die wear parameter is sequentially varied within a value greater than or equal to the estimated die wear parameter to search for reference data that maximizes the second similarity with the measurement data. The die wear parameter corresponding to the reference data with the highest second similarity is determined as the estimated die wear parameter. The processing state estimation device according to claim 6.

8. The aforementioned plurality of parameters further include a workpiece thickness parameter that defines the thickness of the workpiece processed by the processing machine, In the process of searching for the reference data, the processor sequentially changes the workpiece thickness parameter to search for the reference data that has the greatest similarity to the measurement data. A processing state estimation device according to any one of claims 4 to 7.

9. The aforementioned wear parameters are: A punch wear parameter that defines the degree of wear of the punch of the aforementioned processing machine, Includes a die wear parameter that defines the degree of wear of the die of the processing machine, The estimated wear parameters are, The estimated punch wear parameter, which was estimated as the punch wear parameter at the time of measurement of the measurement data, The measurement data includes the estimated die wear parameter, which is estimated as the die wear parameter at the time of measurement. In the process of searching for the reference data, the processor sequentially changes the punch wear parameter within a value greater than or equal to the estimated punch wear parameter, and sequentially changes the die wear parameter within a value greater than or equal to the estimated die wear parameter, in order to search for reference data that has the greatest similarity to the measurement data. The processing state estimation device according to claim 8.

10. The estimated parameter set includes the estimated work thickness parameter estimated as the work thickness parameter at the time of measurement of the measurement data, In the process of searching for the reference data, the processor The die wear parameter and the punch wear parameter are fixed to the estimated die wear parameter and the estimated punch wear parameter, respectively, and the workpiece thickness parameter is sequentially changed to search for reference data that has the maximum first similarity to the measured data. The workpiece thickness parameter corresponding to the reference data with the highest first similarity is determined as the estimated workpiece thickness parameter. The workpiece thickness parameter and the die wear parameter are fixed to the determined estimated workpiece thickness parameter and the estimated die wear parameter, respectively, and the punch wear parameter is sequentially changed within a value greater than or equal to the estimated punch wear parameter to search for reference data that maximizes the second similarity with the measurement data. The punch wear parameter corresponding to the reference data with the highest second similarity is determined as the estimated punch wear parameter. The punch wear parameter is fixed to the determined estimated punch wear parameter, and the die wear parameter is sequentially varied within a value greater than or equal to the estimated die wear parameter to search for reference data that maximizes the third similarity with the measurement data. The die wear parameter corresponding to the reference data with the highest similarity is determined as the estimated die wear parameter. The processing state estimation device according to claim 9.

11. The machining state estimation device according to any one of claims 4 to 10, wherein the processor sets the estimated wear parameter to an initial value when it receives a signal indicating that the tool has been replaced or when it receives a signal indicating that the tool has been polished.

12. The processor sets the estimated wear parameter to an initial value and then continues until a predetermined period of time has elapsed. The estimated wear parameters are kept at their initial values. The processing state estimation device according to claim 11, wherein in the process of searching for the aforementioned reference data, the clearance parameter is sequentially changed to search for the reference data from among the plurality of reference data that has the greatest similarity to the measurement data.

13. Each of the aforementioned parameters is a variable selected from a plurality of discrete values, The processor generates complementary reference data corresponding to the parameters between the plurality of discrete values ​​based on the plurality of reference data, In the search process, the processor searches for the reference data or complementary reference data from among the plurality of reference data and complementary reference data that has the greatest similarity to the measurement data. A processing state estimation device according to any one of claims 1 to 12.

14. Each of the aforementioned multiple reference data shows the relationship between time and processing load during processing by the processing machine, The aforementioned measurement data shows the relationship between time and processing load during processing by the processing machine as measured. A processing state estimation device according to any one of claims 1 to 13.

15. Each of the aforementioned multiple reference data shows the relationship between the sliding position of the processing machine and the processing load during processing by the processing machine. The aforementioned measurement data shows the relationship between time and processing load during processing by the processing machine as measured. The storage device further stores time-slide position relationship data that defines the relationship between time and the slide position, In the process of searching for the reference data, the processor Based on the time-slide position relationship data, the plurality of reference data are converted into a plurality of secondary reference data each showing the relationship between time and machining load during machining by the machining machine, The system searches for the secondary reference data that has the greatest similarity to the measurement data from among the multiple secondary reference data. A processing state estimation device according to any one of claims 1 to 13.

16. Each of the aforementioned multiple reference data shows the relationship between the sliding position of the processing machine and the processing load during processing by the processing machine. The aforementioned measurement data shows the relationship between time and processing load during processing by the processing machine as measured. The storage device further stores time-slide position relationship data that defines the relationship between time and the slide position, In the process of searching for the reference data, the processor Based on the time-slide position relationship data, the measurement data is converted into converted measurement data that shows the relationship between the slide position and the processing load during processing by the processing machine. By searching for the reference data with the greatest similarity to the converted measurement data from among the multiple reference data, the system searches for the reference data with the greatest similarity to the measurement data from among the multiple reference data. A processing state estimation device according to any one of claims 1 to 13.

17. The processor acquires measurement data showing the measurement results of the machining load by the machining machine, The processor performs the step of determining a similarity, which is an indicator of the degree of similarity between a plurality of reference data corresponding to all combinations of a plurality of parameters defining the machining state of a machining tool and the measurement data, The processor estimates the processing state based on the determined similarity, A method for estimating the processing state, including the method described above.

Images