Abnormal area identification device and computer readable storage medium

Abstract

This abnormal area identification device comprises: a first determination unit that determines whether an abnormal portion is included in a determination target area; a division unit that, when the first determination unit has determined that an abnormal portion is included in the determination target area, divides the determination target area, which has been determined by the first determination unit as including the abnormal portion, into a plurality of areas; a second determination unit that determines whether an abnormal portion is included in the plurality of areas divided by the division unit; and an abnormal area identification unit that, when the second determination unit has determined that no abnormal portion is not included in the plurality of areas, identifies the plurality of areas as abnormal areas.

Description

Abnormal region identifying device and computer-readable storage medium

[0001] The present disclosure relates to an abnormal region identifying device and a computer-readable storage medium.

[0002] Conventionally, whether or not an abnormal part exists on a machining surface has been evaluated using machining surface data (for example, see Patent Document 1). For example, a region indicated by the machining surface data is divided into a plurality of regions of a predetermined size, and it is determined whether or not an abnormal part exists in each of the regions.

[0003] Japanese Patent Application Laid-Open No. 2021-105825

[0004] However, conventional techniques are unable to detect the location of an abnormality within each region, so a technology is needed to accurately identify the abnormality formed on the processed surface.

[0005] The abnormal region identification device of the present disclosure includes a first determination unit that determines whether or not an abnormal portion is included in a determination target region of processed surface data; a division unit that, if the first determination unit determines that the determination target region includes an abnormal portion, divides the determination target region that the first determination unit has determined to include an abnormal portion into multiple regions; a second determination unit that determines whether or not an abnormal portion is included in the multiple regions divided by the division unit; and an abnormal region identification unit that, if the second determination unit determines that the multiple regions do not include an abnormal portion, identifies the multiple regions as abnormal regions.

[0006] The computer-readable storage medium of the present disclosure stores instructions that cause a computer to perform the following: determine whether or not a judgment target area of ​​processed surface data contains an abnormal portion; if it is determined that the judgment target area contains an abnormal portion, divide the judgment target area that is determined to contain an abnormal portion into multiple areas; determine whether or not the multiple divided areas contain an abnormal portion; and if it is determined that the multiple areas do not contain an abnormal portion, identify the multiple areas as abnormal areas.

[0007] 1 is a diagram illustrating an example of machining surface data. FIG. 2 is an example of a machining surface indicated by the machining surface data. FIG. 3 is a cross-sectional view of a portion of the machining surface. FIG. 4 is a diagram for explaining division of machining surface data. FIG. 5 is a diagram for explaining division of machining surface data. FIG. 6 is a diagram for explaining division of machining surface data. FIG. 7 is a block diagram illustrating an example of a hardware configuration of an abnormal area identifying device. FIG. 8 is a block diagram illustrating an example of functions of an abnormal area identifying device. FIG. 9 is a diagram illustrating an example of a judgment target area divided by a division unit. FIG. 10 is an example of machining surface data generated by the abnormal area identifying unit. FIG. 11 is a block diagram illustrating an example of functions of an abnormal area identifying device including a similarity determination unit and an abnormality identifying unit. FIG. 12 is an example of machining surface data indicating a moving speed of a tool during machining. FIG. 13 is an example of machining surface data in which an abnormal area is identified. FIG. 14 is an example of machining surface data in which an abnormal area is identified. FIG. 15 is a diagram illustrating machining surface data including a first abnormal area. FIG. 16 is a diagram illustrating machining surface data including a second abnormal area. FIG. 17 is a diagram illustrating machining surface data including a third abnormal area. FIG. 18 is a flowchart illustrating an example of processing executed by an abnormal area identifying device.

[0008] An abnormal region identification device and a computer-readable storage medium according to an embodiment of the present disclosure will be described below with reference to the drawings. In the following description, components having the same or similar functions will be denoted by the same reference numerals. Duplicate descriptions of those components may be omitted.

[0009] In this application, "based on XX" means "based on at least XX," and includes cases where it is based on other elements in addition to XX. Furthermore, "based on XX" is not limited to cases where XX is used directly, but also includes cases where it is based on XX that has been calculated or processed. "XX" is any element (for example, any information).

[0010] First Embodiment An abnormal region identifying device is a device that identifies an abnormal region in machining surface data. The machining surface data is two-dimensional data that indicates information about the machining surface. The information about the machining surface is, for example, unevenness of the machining surface.

[0011] 1 is a diagram showing an example of machining surface data. The machining surface data is data in which numerical values ​​N1, N2, and N3 indicating the height of the machining surface are associated with each of the areas divided into a grid and recorded. Although not shown in the figure, numerical values ​​indicating the height of the machining surface are similarly associated and recorded for other areas. This allows the machining surface data to indicate the unevenness of the machining surface.

[0012] The machining surface data is data generated by, for example, a machining simulation, which is, for example, a simulation of cutting by a processing machine.

[0013] Machining simulation is a process for generating machined surface data using a virtual model of a machining machine, a machining program, machining conditions, and the like.

[0014] The virtual model includes, for example, a model of a structure constituting the processing machine and a model of a workpiece. The model of the structure is generated based on information such as the shape, weight, strength, material, etc. of the structure. The model of the workpiece is generated based on information such as the shape, weight, strength, material, etc. of the workpiece.

[0015] The machining surface data indicating information about the machining surface may be generated based on the measurement results of the machining surface actually machined by a machining machine using a machining program. When the machining surface is actually machined by a machining machine, the machining surface may be measured using a three-dimensional scanner, a laser confocal microscope, a white light interference microscope, or the like.

[0016] 2A is an example of a machined surface indicated by the machined surface data. 2B is a cross-sectional view of a portion of the machined surface. The machined surface includes an abnormal portion. The abnormal portion is, for example, a portion where a streak is formed.

[0017] Streaks are formed, for example, due to a problem with the commands that specify the tool movement path in a machining program generated by CAM (Computer Aided Manufacturing). Streaks are elongated depressions. The depth of the streaks is, for example, about 1 μm.

[0018] Next, an overview of the method by which the abnormal region identifying device identifies an abnormal region will be described. The abnormal region identifying device divides the machining surface data into multiple regions and determines whether each region contains an abnormal part. Furthermore, the abnormal region identifying device further divides the region containing the abnormal part into multiple regions and identifies the position of the abnormal part. The abnormal region identifying device identifies the position on the machining surface where the abnormal part exists by repeatedly executing this process.

[0019] However, if the area containing the abnormal part is divided into too small areas, it may not be possible to detect whether or not each area contains an abnormal part. In other words, if the area containing the abnormal part is divided into too small areas when streaks are formed on the machined surface, the difference in height of the machined surface within the area will be small, and the abnormality will not be detected.

[0020] For example, if the width of the region containing the abnormal part is the width indicated by frame F1 in Figure 2B, the region also contains normal parts in addition to the abnormal part. In other words, the size of the region containing the abnormal part is not appropriate, so the position of the abnormal part cannot be identified with high accuracy.

[0021] Furthermore, when the width of the region containing the abnormal part is the width indicated by frame F2 in Fig. 2B, the region contains only the abnormal part. In other words, the size of the region containing the abnormal part is appropriate, so the position of the abnormal part can be identified with high accuracy.

[0022] 2B, the area containing the abnormal part contains only a part of the abnormal part. In other words, the size of the area containing the abnormal part is too small, and the abnormal part is not detected.

[0023] 3A to 3C are diagrams for explaining the division of machining surface data. The abnormal region identifying device identifies which divided region contains the abnormal part. However, the abnormal region identifying device cannot identify the position within each region where the abnormal part exists.

[0024] 3A, the size of each divided area is relatively large, so the accuracy of identifying the position of the abnormal part is not very high.

[0025] 3B, the size of each divided area is an appropriate size that matches the size of the abnormal part, so the accuracy of identifying the abnormal part is high.

[0026] 3C, the size of each divided region is too small, so the abnormal region identifying device cannot identify which region contains the abnormal part.

[0027] Therefore, when an abnormal part is no longer detected in multiple regions generated by dividing an area containing an abnormal part, the abnormal region identification device can identify the multiple regions as abnormal regions.

[0028] 4 is a block diagram showing an example of the hardware configuration of an abnormal area identification device 1. The abnormal area identification device 1 is implemented in, for example, a numerical control device, a personal computer (PC), a server, a tablet terminal, etc. The abnormal area identification device 1 includes, for example, a hardware processor 101, a bus 102, a read-only memory (ROM) 103, a random access memory (RAM) 104, a non-volatile memory 105, and an input / output device 106.

[0029] The hardware processor 101 is a processor that uses a system program to control the entire abnormal area identifying device 1. The hardware processor 101 reads the system program and the like stored in the ROM 103 via the bus 102. The hardware processor 101 is, for example, a CPU (Central Processing Unit) or an electronic circuit.

[0030] The bus 102 is a communication path that connects the various pieces of hardware in the abnormal area identifying device 1. The various pieces of hardware in the abnormal area identifying device 1 exchange data via the bus 102.

[0031] The ROM 103 is a storage device that stores system programs, etc. The ROM 103 is a computer-readable storage medium.

[0032] The RAM 104 is a storage device that temporarily stores various data and functions as a work area for the hardware processor 101 to process various data.

[0033] The nonvolatile memory 105 is a storage device that retains data even when the power to the abnormal area identification device 1 is turned off. The nonvolatile memory 105 stores, for example, processed surface data. The nonvolatile memory 105 is a computer-readable storage medium. The nonvolatile memory 105 is, for example, a battery-backed memory or an SSD (Solid State Drive).

[0034] The input / output device 106 receives various data from the hardware processor 101 and displays the data on a display. The input / output device 106 also receives input of various data and sends the data to the hardware processor 101, for example.

[0035] The input / output device 106 is, for example, a touch panel. When the input / output device 106 is a touch panel, the input / output device 106 is, for example, a capacitive touch panel. The touch panel is not limited to a capacitive touch panel and may be a touch panel of another type.

[0036] 5 is a block diagram showing an example of the functions of the abnormal area identifying device 1. The abnormal area identifying device 1 includes a processing surface data acquiring unit 111, a first determining unit 112, a dividing unit 113, a second determining unit 114, an area setting unit 115, an abnormal area identifying unit 116, and an output unit 117.

[0037] The machining surface data acquisition unit 111, the first judgment unit 112, the division unit 113, the second judgment unit 114, the area setting unit 115, the abnormal area identification unit 116, and the output unit 117 are realized, for example, by the hardware processor 101 performing arithmetic processing using a system program stored in the ROM 103 and various programs and data stored in the non-volatile memory 105.

[0038] The machining surface data acquisition unit 111 acquires machining surface data. The machining surface data acquisition unit 111 acquires the machining surface data, for example, from an external simulation device. The machining surface data acquisition unit 111 may acquire the machining surface data from a server, a PC, a numerical control device, etc. connected to the abnormal area identification device 1. The machining surface data acquisition unit 111 acquires, for example, the machining surface data shown in FIG. 2A .

[0039] The first determination unit 112 determines whether or not an abnormal part is included in a determination target area of ​​the machining surface data. The determination target area is an area for which the first determination unit 112 determines whether or not there is an abnormal part. The determination target area is, for example, the entire area of ​​the machining surface indicated by the machining surface data acquired by the machining surface data acquisition unit 111. The determination target area may be a partial area of ​​the machining surface indicated by the machining surface data.

[0040] The abnormal portion is, for example, a portion where at least one of the arithmetic mean value of the numerical values ​​indicated by the machining surface data and the value obtained by subtracting the minimum value from the maximum value of the numerical values ​​indicated by the machining surface data is equal to or greater than a predetermined threshold value. The arithmetic mean value of the numerical values ​​indicated by the machining surface data and the value obtained by subtracting the minimum value from the maximum value of the numerical values ​​indicated by the machining surface data are, for example, surface roughness Sa and maximum height Sz. The first determination unit 112 determines whether the determination target region includes an abnormal portion by, for example, evaluating whether at least one of the surface roughness Sa and maximum height Sz in the determination target region is equal to or greater than a predetermined threshold value.

[0041] When the first determination unit 112 determines that the determination target region includes an abnormal portion, the division unit 113 divides the determination target region determined by the first determination unit 112 to include an abnormal portion into a plurality of regions. In other words, each region is formed by dividing the determination target region.

[0042] 6 is a diagram showing an example of the determination target area divided by the dividing unit 113. The dividing unit 113 divides the determination target area A1 into four areas a11 to a14, for example.

[0043] The second determination unit 114 determines whether or not an abnormal portion is included in the multiple regions a11 to a14 generated by division by the division unit 113. The second determination unit 114 evaluates the surface texture of the multiple regions a11 to a14 to determine whether or not an abnormal portion is included in at least one of the multiple regions. As described above, the surface texture is, for example, the surface roughness Sa or the maximum height Sz. That is, the second determination unit 114 determines whether or not an abnormal portion is included in each of the regions a11 to a14 by determining whether or not the surface roughness Sa or the maximum height Sz of each of the regions a11 to a14 is equal to or greater than a predetermined threshold value.

[0044] When the second determination unit 114 determines that any one of the multiple regions a11 to a14 contains an abnormal portion, the region setting unit 115 sets the one of the regions determined to contain an abnormal portion as the determination target region. In the example shown in FIG. 6, the second determination unit 114 determines, for example, that region a14 contains an abnormal portion. Furthermore, the second determination unit 114 determines that regions a11 to a13 do not contain an abnormal portion. In this case, the region setting unit 115 sets region a14 as the determination target region A2.

[0045] The dividing unit 113 divides each of the areas set as the judgment target area A2 by the area setting unit 115 into a plurality of areas a21 to a24.

[0046] The second determination unit 114 determines whether or not the multiple regions a21 to a24 divided by the dividing unit 113 contain any abnormal portions. The second determination unit 114 determines, for example, that regions a21 and a24 contain abnormal portions. The second determination unit 114 also determines that regions a22 and a23 do not contain any abnormal portions. In this case, the region setting unit 115 sets region a21 as a determination target region A3. Furthermore, although not shown in the figure, the region setting unit 115 sets region a24 as another determination target region.

[0047] The dividing unit 113 divides into a plurality of regions the region set as the judgment target region A3 by the region setting unit 115. The dividing unit 113 divides the judgment target region A3 into a plurality of regions a31 to a34. Similarly, the dividing unit 113 divides other judgment target regions into a plurality of regions.

[0048] The second determination unit 114 determines whether or not the multiple regions a31 to a34 divided by the dividing unit 113 contain any abnormal portions. The second determination unit 114 determines, for example, that regions a31 and a33 contain any abnormal portions. The second determination unit 114 also determines that regions a32 and a34 do not contain any abnormal portions. In this case, the region setting unit 115 sets region a31 as a determination target region A4. Furthermore, although not shown in the figure, the region setting unit 115 sets region a33 as another determination target region.

[0049] The dividing unit 113 divides into a plurality of regions the region set as the judgment target region A4 by the region setting unit 115. The dividing unit 113 divides the judgment target region A4 into a plurality of regions a41 to a44. Similarly, the dividing unit 113 divides another judgment target region into a plurality of regions.

[0050] The second determination unit 114 determines whether or not an abnormal portion is included in the multiple regions a41 to a44 divided by the division unit 113. For example, the second determination unit 114 determines that the multiple regions a41 to a44 do not include an abnormal portion. In other words, by dividing the determination target region into smaller portions by the division unit 113, an abnormality in the determination target region is no longer detected.

[0051] When the second determination unit 114 determines that the plurality of regions a41 to a44 do not include an abnormal portion, the abnormal region identification unit 116 identifies the plurality of regions a41 to a44 as abnormal regions. The abnormal region identification unit 116 surrounds the identified abnormal region with, for example, a rectangular frame. In other words, the abnormal region identification unit 116 generates machining surface data in which the abnormal region is surrounded by a rectangular frame.

[0052] 7 shows an example of machining surface data generated by the abnormal region identifying unit 116. An abnormal region in the machining surface data is enclosed by a rectangular frame.

[0053] The output unit 117 outputs the machining surface data generated by the abnormal region identifying unit 116. The output unit 117 outputs the machining surface data to, for example, the input / output device 106. The input / output device 106 displays the machining surface data on a display. This allows the user to quantitatively evaluate the machining surface based on the machining surface data.

[0054] 8 is a flowchart showing an example of processing executed by the abnormal region identifying device 1. In the abnormal region identifying device 1, first, the processing surface data acquiring unit 111 acquires processing surface data (step S1).

[0055] Next, the first determination unit 112 determines whether or not the determination target region includes an abnormal portion (step S2). If the first determination unit 112 determines that the determination target region does not include an abnormal portion (No in step S2), the output unit 117 outputs the processed surface data (step S7), and the process ends.

[0056] If the first determination unit 112 determines that the determination target region includes an abnormal portion (Yes in step S2), the division unit 113 divides the determination target region into a plurality of regions (step S3).

[0057] Next, the second determination unit 114 determines whether or not the plurality of regions includes an abnormal portion (step S4).

[0058] If one or more of the multiple regions contains an abnormal part (Yes in step S4), the region setting unit 115 sets the region containing the abnormal part as the determination target region (step S5), and then the processes from step S3 onward are executed.

[0059] On the other hand, if it is determined that the plurality of regions does not contain an abnormal portion (No in step S4), the abnormal region identifying unit 116 identifies the plurality of regions as abnormal regions (step S6). That is, the abnormal region identifying unit 116 generates machining surface data in which the abnormal regions are identified. Thereafter, the output unit 117 outputs the machining surface data (step S7), and the process ends.

[0060] Second Embodiment In the first embodiment described above, an example was described in which the machining surface data is data indicating the unevenness of the machining surface. However, the machining surface data may be data indicating information about the machining surface other than the unevenness.

[0061] The machining surface data may be data indicating, for example, at least one of the movement speed of the tool during machining, a numerical value calculated based on the movement speed, the correction amount during machining, a numerical value calculated based on the correction amount, the detection value of the detection signal during machining, a numerical value calculated based on the detection value, the command value of the command signal, and a numerical value calculated based on the command value.

[0062] The correction amounts include, for example, a backlash correction amount, a pitch error correction amount, a spatial error correction amount, and a thermal displacement correction amount. The command signals include, for example, a command signal for enabling a function and a command signal for operating an external device. The external device is, for example, a coolant, a fan, and a chip conveying device. The detection signals include a feedback signal in response to the command signal for enabling a function, a feedback signal from the external device, and a resource usage rate signal.

[0063] The numerical values ​​calculated based on the moving speed, correction amount, detection value, and command value may be the differential or integral value of the moving speed, the differential or integral value of the correction amount, the differential or integral value of the detection value, and the differential or integral value of the command value.

[0064] Even when the machining surface data indicates information about the machining surface other than the unevenness of the machining surface, a numerical value is associated with each area divided into a grid and recorded as shown in Fig. 1. Therefore, even when the machining surface data indicates information other than the unevenness of the machining surface, the abnormal area identifying device 1 can determine the abnormal portion appearing in the machining surface data and identify the abnormal area.

[0065] When the machining surface data is any of the above-described pieces of information related to the machining surface other than the unevenness, the abnormal region identifying device 1 may identify the abnormal region based on the machining surface data indicating any of the information. For example, when the machining surface data is data indicating a backlash correction amount, the abnormal region identifying device 1 identifies the abnormal region indicating an abnormal portion of the backlash correction amount.

[0066] The abnormal area identification device 1 may further include a similarity determination unit that determines the similarity between the abnormal area identified by the abnormal area identification unit 116 when the machining surface data is data indicating first information and the abnormal area identified by the abnormal area identification unit 116 when the machining surface data is data indicating second information, and an abnormality identification unit that determines the cause of the abnormality in the abnormal part based on the determination result of the similarity determination unit.

[0067] Fig. 9 is a block diagram showing an example of the functions of an abnormal region identifying device 1 including a similarity determination unit and an abnormality identification unit. The functions of the abnormal region identifying device 1 shown in Fig. 9 other than the similarity determination unit 118 and the abnormality identification unit 119 are the same as the functions of the abnormal region identifying device 1 shown in Fig. 5. Therefore, only the similarity determination unit 118 and the abnormality identification unit 119 will be described here, and descriptions of the other functions will be omitted.

[0068] The abnormal area identification device 1 includes a processing surface data acquisition unit 111, a first judgment unit 112, a division unit 113, a second judgment unit 114, an area setting unit 115, an abnormal area identification unit 116, and an output unit 117, as well as a similarity judgment unit 118 and an abnormality identification unit 119.

[0069] The similarity determination unit 118 and the anomaly identification unit 119 are realized, for example, by the hardware processor 101 performing arithmetic processing using a system program stored in the ROM 103 and various programs and data stored in the non-volatile memory 105.

[0070] The machining surface data acquisition unit 111 acquires machining surface data. The machining surface data is data indicating first information. The first information is, for example, information indicating the height of the machining surface. In this case, as described in the first embodiment, the abnormal region identification unit 116 identifies an abnormal region including an abnormal portion related to the unevenness of the machining surface.

[0071] The machining surface data acquisition unit 111 acquires machining surface data indicating information about the machining surface other than the concaves and convexes. The machining surface data is data indicating second information. The second information is, for example, information indicating the moving speed of a tool during machining.

[0072] 10A is an example of machining surface data indicating the movement speed of a tool during machining. When the machining surface data is displayed on a screen, for example, portions where the tool movement speed is slow are displayed in dark colors. Also, portions where the tool movement speed is fast are displayed in light colors. Note that in FIG. 10A and FIG. 10B described later, for convenience, dark colored portions are displayed as hatched rectangular blocks.

[0073] The first determination unit 112, the division unit 113, the second determination unit 114, and the abnormal region identification unit 116 execute, for example, the processes of steps S2 to S6 shown in Fig. 8. As a result, an abnormal region including an abnormal part related to the moving speed of the tool during machining is identified.

[0074] 10B shows an example of machining surface data in which an abnormal region has been identified. The region identified as an abnormal region is enclosed by, for example, a rectangular frame.

[0075] The machining surface data acquisition unit 111 acquires further machining surface data indicating information about the machining surface other than the unevenness. The machining surface data is data indicating third information. The third information is, for example, information indicating a backlash correction amount during machining.

[0076] 11A is an example of machining surface data indicating the amount of backlash compensation during machining. When the machining surface data is displayed on a screen, for example, areas with a large amount of backlash compensation are displayed in dark colors. Areas with a small amount of backlash compensation are displayed in light colors. For convenience, in FIG. 11A and FIG. 11B, which will be described later, dark-colored areas are displayed as hatched rectangular blocks.

[0077] The first determination unit 112, the dividing unit 113, the second determination unit 114, and the abnormal region identifying unit 116 execute, for example, the processes of steps S2 to S6 shown in Fig. 8. As a result, an abnormal region including an abnormal portion related to the backlash correction amount during machining is identified.

[0078] 11B shows an example of machining surface data in which an abnormal region has been identified. The region identified as an abnormal region is enclosed by, for example, a rectangular frame.

[0079] The similarity determination unit 118 determines the similarity between the abnormal area identified by the abnormal area identifying unit 116 when the machining surface data is data indicating first information and the abnormal area identified by the abnormal area identifying unit 116 when the machining surface data is data indicating second information. Moreover, the similarity determination unit 118 determines the similarity between the abnormal area identified by the abnormal area identifying unit 116 when the machining surface data is data indicating second information and the abnormal area identified by the abnormal area identifying unit 116 when the machining surface data is data indicating third information. Moreover, the similarity determination unit 118 determines the similarity between the abnormal area identified by the abnormal area identifying unit 116 when the machining surface data is data indicating third information and the abnormal area identified by the abnormal area identifying unit 116 when the machining surface data is data indicating first information. Note that the types of machining surface data to be subjected to similarity determination are not limited to three, and may be two or four or more.

[0080] Fig. 12A is a diagram showing machining surface data including a first abnormal region. Fig. 12B is a diagram showing machining surface data including a second abnormal region. Fig. 12C is a diagram showing machining surface data including a third abnormal region. In Figs. 12A to 12C, each abnormal region is indicated by hatching.

[0081] The similarity determination unit 118 determines the degree of similarity by comparing the machining surface data including the first abnormal region, the machining surface data including the second abnormal region, and the machining surface data including the third abnormal region.

[0082] The similarity determination unit 118 determines the similarity using, for example, Content Based Information Retrieval (CBIR). The similarity determination unit 118 determines the similarity of the distribution of abnormal regions included in each piece of machining surface data. In other words, the similarity determination unit 118 determines the similarity by comparing the positions of the abnormal regions included in the machining surface data.

[0083] The similarity determination unit 118 may determine the degree of similarity by comparing the feature amounts indicated by the processed surface data of each abnormal region with each other using CBIR.

[0084] The regions compared by the similarity determination unit 118 are regions located in corresponding positions. For example, the regions surrounded by thick lines among the abnormal regions shown in Figures 12A and 12C are the first abnormal region and the third abnormal region located in corresponding positions.

[0085] The feature amount may be, for example, a histogram, an average hash, or a scale-invariant feature transform (SIFT). The similarity determination unit 118 may determine the similarity using an evaluation method according to the type of feature amount. The evaluation method may be, for example, an evaluation method using a Hamming distance, a cosine similarity, or the like.

[0086] The anomaly identifying unit 119 identifies the cause of the anomaly in the anomaly portion based on the determination result of the similarity determining unit 118. For example, if the position of the first anomaly region and the position of the third anomaly region included in the machining surface data are similar, the irregularity in the first anomaly region is likely to be caused by an anomaly in the backlash correction amount. In this case, the anomaly identifying unit 119 determines that the cause of the anomaly portion is the backlash correction amount. In other words, the anomaly identifying unit 119 identifies the cause of the anomaly.

[0087] Furthermore, if the characteristics of the first abnormal region and the third abnormal region are similar, the irregularity in the first abnormal region is likely to be caused by an abnormality in the backlash correction amount, and in this case, the abnormality identifying unit 119 determines that the cause of the abnormality is the backlash correction amount.

[0088] The output unit 117 outputs data indicating the cause of the abnormality identified by the abnormality identification unit 119. The output unit 117 outputs the data indicating the cause of the abnormality to, for example, the display of the input / output device 106. This allows the operator to confirm the cause of the abnormality on the machined surface.

[0089] 13 is a flowchart showing an example of processing executed by the abnormal area identifying device 1. The abnormal area identifying device 1 first performs processing to identify an abnormal area (step SA1). The processing of step SA1 is the same as the processing of steps S1 to S6 in the flowchart shown in Fig. 8. The abnormal area identifying device 1 acquires, for example, machining surface data indicating first information, machining surface data indicating second information, and machining surface data indicating third information, and identifies an abnormal area in each of the machining surface data.

[0090] Next, the similarity determination unit 118 determines the similarity of each abnormal region (step SA2). Next, the abnormality identification unit 119 identifies the cause of the abnormality (step SA3). Finally, the output unit 117 outputs the processed surface data and data indicating the cause of the abnormality (step SA4), and the process ends.

[0091] As described above, the abnormal region identifying device 1 includes the first determining unit 112 that determines whether or not an abnormal portion is included in a determination target region of the machining surface data, the dividing unit 113 that, if the first determining unit 112 determines that the determination target region includes an abnormal portion, divides the determination target region that the first determining unit 112 determines to include an abnormal portion into multiple regions, the second determining unit 114 that determines whether or not an abnormal portion is included in the multiple regions divided by the dividing unit 113, and the abnormal region identifying unit 116 that, if the second determining unit 114 determines that the multiple regions do not include an abnormal portion, identifies the multiple regions as abnormal regions. Therefore, the abnormal region identifying device 1 can accurately identify the position of an abnormal portion formed on the machining surface.

[0092] Furthermore, the abnormal region identification device 1 further includes a region setting unit 115 that, when the second determination unit 114 determines that any of the multiple regions contains an abnormal portion, sets any of the regions determined to contain an abnormal portion as a determination target region, and the division unit 113 divides any of the regions set as the determination target region by the region setting unit 115 into multiple regions. That is, the abnormal region identification device 1 may repeatedly divide the determination target region and determine whether or not the regions generated by the division contain an abnormal portion. Therefore, the abnormal region identification device 1 can accurately identify the position of the abnormal portion.

[0093] Furthermore, an abnormal portion is a portion where at least one of the arithmetic mean value of the numerical values ​​indicated by the machined surface data and the value obtained by subtracting the minimum value from the maximum value of the numerical values ​​is equal to or greater than a predetermined threshold value. Therefore, the abnormal region identifying device 1 can accurately identify, for example, the positions of portions on the machined surface where the surface roughness Sa is rough and portions on the machined surface where the maximum height Sz is high. Alternatively, the abnormal region identifying device 1 can accurately identify positions on the machined surface where there is a large fluctuation in the tool movement speed.

[0094] The abnormal region identifying device 1 further includes a similarity determination unit 118 that determines the degree of similarity between the abnormal region identified by the abnormal region identifying unit 116 when the machining surface data is data indicating the first information and the abnormal region identified by the abnormal region identifying unit 116 when the machining surface data is data indicating the second information, and an abnormality identification unit 119 that identifies the cause of the abnormality in the abnormal portion based on the determination result of the similarity determination unit 118. Therefore, the abnormal region identifying device 1 can identify the cause of the abnormality in the abnormal portion.

[0095] Although the present disclosure has been described in detail, the present disclosure is not limited to the individual embodiments described above. Various additions, substitutions, modifications, partial deletions, etc. are possible to these embodiments without departing from the gist of the present disclosure or the gist of the present disclosure derived from the content of the claims and their equivalents. Furthermore, these embodiments can also be implemented in combination.

[0096] The following are supplementary notes related to embodiments of the present disclosure. Supplementary note [1] An abnormal region identifying device comprising: a first determination unit that determines whether a determination target region of machining surface data includes an abnormal portion; a division unit that, when the first determination unit determines that the determination target region includes the abnormal portion, divides the determination target region determined by the first determination unit to include the abnormal portion into a plurality of regions; a second determination unit that determines whether the abnormal portion is included in the plurality of regions generated by division by the division unit; and an abnormal region identifying unit that, when the second determination unit determines that the plurality of regions do not include the abnormal portion, identifies the plurality of regions as abnormal regions. Supplementary note [2] The abnormal region identifying device according to Supplementary note [1] further comprises a region setting unit that, when the second determination unit determines that any of the plurality of regions includes the abnormal portion, sets any of the regions determined to include the abnormal portion as the determination target region, wherein the dividing unit divides any of the regions set as the determination target region by the region setting unit into a plurality of regions. Supplementary Note [3] The abnormal region identifying device according to Supplementary Note [2], wherein the abnormal portion is a portion where at least one of an arithmetic mean value of numerical values ​​indicated by machining surface data and a value obtained by subtracting the minimum value from the maximum value of the numerical values ​​is equal to or greater than a predetermined threshold value.Supplementary Note [4] The abnormal region identifying device according to Supplementary Note [1], further comprising: a similarity determination unit that determines a similarity between the abnormal region identified by the abnormal region identifying unit when the machining surface data is data indicating first information and the abnormal region identified by the abnormal region identifying unit when the machining surface data is data indicating second information; and an abnormality identification unit that determines a cause of the abnormality of the abnormal portion based on a determination result of the similarity determination unit.Supplementary Note [5] A computer-readable storage medium that stores instructions for causing a computer to execute the following: determining whether or not a judgment target area of ​​machining surface data includes an abnormal portion; if it is determined that the judgment target area includes the abnormal portion, dividing the judgment target area determined to include the abnormal portion into a plurality of areas; determining whether or not the abnormal portion is included in the plurality of divided areas; and if it is determined that the abnormal portion is not included in the plurality of areas, identifying the plurality of areas as abnormal areas.

[0097] REFERENCE SIGNS LIST 1 Abnormal region identifying device 101 Hardware processor 102 Bus 103 ROM 104 RAM 105 Non-volatile memory 106 Input / output device 111 Machining surface data acquisition unit 112 First determination unit 113 Dividing unit 114 Second determination unit 115 Region setting unit 116 Abnormal region identifying unit 117 Output unit 118 Similarity determining unit 119 Abnormal identifying unit

Claims

1. An abnormal region identification device comprising: a first judgment unit that judges whether or not an abnormal portion is included in a judgment target region of machining surface data; a division unit that, when the first judgment unit judges that the judgment target region contains the abnormal portion, divides the judgment target region that the first judgment unit judges to contain the abnormal portion into a plurality of regions; a second judgment unit that judges whether or not the plurality of regions generated by division by the division unit contain the abnormal portion; and an abnormal region identification unit that, when the second judgment unit judges that the plurality of regions do not contain the abnormal portion, identifies the plurality of regions as abnormal regions.

2. The abnormal area identification device of claim 1, further comprising a region setting unit which, when the second judgment unit judges that the abnormal area is contained in any one of the plurality of regions, sets the any one of the regions judged to contain the abnormal area as the judgment target area, and the division unit divides the any one of the regions set as the judgment target area by the region setting unit into a plurality of areas.

3. The abnormal region identifying device of claim 2, wherein the abnormal portion is a portion in which at least one of the arithmetic mean value of the numerical values ​​indicated by the machining surface data and the value obtained by subtracting the minimum value from the maximum value of the numerical values ​​is equal to or greater than a predetermined threshold value.

4. The abnormal area identification device of claim 1, further comprising: a similarity judgment unit that judges the similarity between the abnormal area identified by the abnormal area identification unit when the machining surface data is data indicating first information and the abnormal area identified by the abnormal area identification unit when the machining surface data is data indicating second information; and an abnormality identification unit that identifies a cause of the abnormality in the abnormal part based on a judgment result of the similarity judgment unit.

5. A computer-readable storage medium storing instructions that cause a computer to execute the following: determining whether or not a judgment target area of ​​machining surface data includes an abnormal portion; if it is determined that the judgment target area includes the abnormal portion, dividing the judgment target area determined to include the abnormal portion into a plurality of areas; determining whether or not the divided plurality of areas includes the abnormal portion; and if it is determined that the plurality of areas does not include the abnormal portion, identifying the plurality of areas as abnormal areas.

Images