JP2021171833A - Surface roughness estimation system - Google Patents

Abstract

To provide a surface roughness estimation system capable of estimating surface roughness of a polishing face of a workpiece with good precision.SOLUTION: A surface roughness estimation system 1 is assembled with an estimation calculating device 20. The estimation calculating device 20 is assembled with a correlation relationship lead-out part 82 for obtaining a correlation relationship between a roughness value which shows surface roughness of a polishing face of a workpiece in a case where measuring is conducted by using a detector 13 with a detection probe and an actual roughness value showing a surface roughness of a polishing face of a workpiece in a case where measuring is conducted by using a roughness meter 2 with a small diameter probe having a small diameter; an estimation data acquiring part 83 for acquiring an estimation data showing a roughness value based on a state data having been output from the detector 13; and a roughness value estimation part 84 for estimating an actual roughness value by using the estimation data and the correlation relationship.SELECTED DRAWING: Figure 6

Description

The present invention relates to a surface roughness estimation system.

Patent Document 1 describes the presence or absence of chatter vibration by comparing the magnitude of vibration in a certain frequency band and the degree of change in vibration within a certain period of time with a discrimination reference value for the vibration of a grinding machine during grinding. The technique for determining the frequency is disclosed. As a result, in particular, chatter vibration at the initial stage of generation is detected.

Japanese Unexamined Patent Publication No. 2000-23338

In general, when a workpiece is ground, vibration may occur due to the surface roughness of the ground surface of the workpiece. If chatter vibration occurs during grinding, it may cause deterioration of grinding quality, and it is extremely important to grasp the surface roughness of the ground surface of the workpiece during grinding, that is, in-process. ..

In this respect, in the technique disclosed in Patent Document 1, although it is possible to grasp the surface roughness of the ground surface of the workpiece by in-process, it is necessary to provide a dedicated vibration meter in the grinding device. The configuration of the grinding machine becomes complicated. By the way, in the grinding device, for example, the shape of a workpiece is measured in-process using a sizing device. Since the sizing device measures the shape of the work by bringing the stylus into contact with the ground surface of the work, for example, the surface roughness of the work is estimated based on the displacement value measured by the sizing device. It is possible.

However, in general, the stylus of the sizing device is set to have a large diameter in order to suppress wear due to contact with the ground surface of the workpiece. Therefore, for example, there is a risk that the measurement accuracy will be lower than the surface roughness measured using a roughness meter having a small diameter of the stylus when offline.

An object of the present invention is to provide a surface roughness estimation system capable of accurately estimating the surface roughness of a ground surface of a workpiece.

The surface roughness estimation system has a grinding device in which the grindstone grinds the workpiece under the control of the control device, and a detection stylus that comes into contact with the ground surface of the workpiece. A detector that detects and outputs state data related to the surface roughness of the surface roughness of the workpiece according to the possible operation of the detection stylus, and the grinding of the workpiece when measuring using the detection stylus. Correlation between the roughness value indicating the surface roughness of the surface and the actual roughness value indicating the surface roughness of the ground surface of the workpiece when measured using a small-diameter stylus with a diameter smaller than that of the detection stylus. Correlation to be derived Using the derivation unit, the estimation data acquisition unit that acquires the estimation data representing the roughness value based on the state data output from the detector, and the estimation data and the correlation, the actual roughness is achieved. It is provided with a roughness value estimation unit for estimating the bulk value.

According to this, the roughness value when measured using the detection stylus of the detector provided in the grinding device and the actual roughness value when measured using the small diameter stylus with a smaller diameter than the detection stylus. The correlation with can be derived. As a result, even when the surface roughness of the ground surface of the workpiece is measured using a detector with a large diameter, the ground surface of the workpiece being ground is based on the derived correlation. The surface roughness, that is, the actual roughness value in the above can be estimated accurately.

It is a figure which shows the structure of the surface roughness estimation system. It is a figure which shows an example of a grinding apparatus. It is a figure which shows the functional block composition of an example of a detector. It is a figure for demonstrating the roughness data in the circumferential direction detected by a detector. It is a figure for demonstrating the surface roughness data generated based on the roughness data in the circumferential direction of FIG. It is a figure which shows the functional block composition of the surface roughness estimation system. It is a figure which shows the solid line roughness curve measured by a roughness meter. It is a figure for demonstrating the central locus line derived based on the envelope. It is a figure which shows the state which superposed the central locus line on the solid line roughness curve. It is a graph which shows the correlation between the actual roughness value (estimated roughness value) and the center locus roughness value (approximate waveform roughness value). It is a figure which shows the line roughness approximate waveform. It is a figure which shows the structure of the surface roughness estimation system which concerns on 1st alternative example. It is a figure which shows the functional block composition of the learning processing apparatus of FIG. It is a figure which shows the amplitude for each frequency extracted by a frequency analysis. It is a figure which shows the functional block composition of the estimation arithmetic unit of FIG.

(1. Grinding equipment to which the surface roughness estimation system is applied)
The surface roughness estimation system estimates the surface roughness of the ground surface of a workpiece ground by a grinding device. As the grinding machine, a grinding machine having various configurations such as a cylindrical grinding machine, a cam grinding machine, and a surface grinding machine can be applied. The grinding device can grind the workpiece through a grinding process such as a rough grinding process, a fine grinding process, a fine grinding process, and a spark out.

(2. Outline of the configuration of the surface roughness estimation system 1)
The outline of the configuration of the surface roughness estimation system 1 will be described with reference to FIG. The surface roughness estimation system 1 includes at least one grinding device 10 and one arithmetic unit 20. The grinding device 10 may be targeted at one unit, or may be targeted at a plurality of units as shown in FIG. In this example, the surface roughness estimation system 1 includes a case where a plurality of grinding devices 10 are provided as an example.

The grinding device 10 includes at least a detector 13 that detects observable state data during grinding of the workpiece W. The arithmetic unit 20 estimates the surface roughness of the workpiece W having a correlation with the state data by using the state data detected by the detector 13. Here, the correlation is the state data related to the surface roughness detected by the detector 13 when the grinding device 10 is grinding the workpiece W, and, for example, grinding the workpiece W after grinding. Line roughness of the workpiece W based on the actual displacement value detected by the small diameter stylus of the roughness meter 2 provided separately from the device 10 (hereinafter, the value representing this line roughness is referred to as "actual roughness value". Represents the correlation with).

(3. Details of the configuration of the surface roughness estimation system 1)
The configuration of the surface roughness estimation system 1 will be described in more detail with reference to FIG. The surface roughness estimation system 1 includes a plurality of grinding devices 10 and a plurality of arithmetic devices 20 provided one-to-one in each grinding device 10. In this example, the arithmetic unit 20 is configured as an estimation arithmetic unit 20.

Each grinding device 10 mainly includes a grinding machine 11 that grinds a workpiece W using a grindstone T, a control device 12 that controls the grinding machine 11, a detector 13, and an interface 14. The control device 12 includes a CNC device, a PLC device, and the like, and controls a drive device and the like in the grinding machine 11. The interface 14 is a device that enables communication with the outside with the grinding machine 11, the control device 12, and the detector 13. In addition, outside, a roughness meter 2 that detects the surface roughness of the ground surface of the workpiece W ground by each grinding device 10 as needed, a learning processing device 60 and a common display device 70, which will be described later, and the like. Is included.

As the state data, the detector 13 detects the state data related to the surface roughness of the ground surface during the grinding of the workpiece W by the grinding machine 11. The detector 13 is, for example, an acceleration sensor or a displacement sensor that detects acceleration data or displacement data due to the surface roughness of the ground surface of the workpiece W. That is, the state data detected by the detector 13 is acceleration data (vibration data), displacement data (roughness data, and surface roughness data described later) and the like.

The estimation arithmetic unit 20 includes a processor 21, a storage device 22, an interface 23, and the like. Each estimation arithmetic unit 20 is communicably connected to the corresponding grinding apparatus 10 and the learning processing apparatus 60 as a server described later.

The estimation arithmetic unit 20 is arranged at a position close to each grinding device 10, and functions as a so-called edge computer. The estimation arithmetic unit 20 estimates (predicts) the surface roughness of the ground surface of the workpiece W based on the surface roughness-related data detected by the detector 13 during the grinding of the workpiece W. Regarding the arrangement of the estimation calculation device 20, it is not always necessary to arrange the estimation calculation device 20 at a position close to the grinding device 10, and it is also possible to arrange the estimation calculation device 20 so as to be separate from the grinding device 10.

The surface roughness estimation system 1 includes a plurality of individual display devices 30. However, the surface roughness estimation system 1 may be configured not to include the individual display device 30. The individual display device 30 is arranged corresponding to each of the grinding devices 10.

(4. Example of grinding machine 11)
As the grinding machine 11 of this example, as shown in FIG. 2, a grindstone base traverse type cylindrical grinding machine 40 is taken as an example. As the grinding machine 11, a table traverse type can also be used.

The cylindrical grinding machine 40 is a machine for grinding the peripheral surface of the workpiece W. The cylindrical grinding machine 40 mainly includes a bed 41, a headstock 42, a tailstock 43, a traverse base 44, a grindstone 45, a grindstone 46 (grindstone T), a sizing device 47, a grindstone correction device 48, and a coolant. The device 49 is provided.

The bed 41 is fixed on the installation surface. The headstock 42 is provided on the upper surface of the bed 41 on the front side in the X-axis direction (lower side in FIG. 2) and one end side in the Z-axis direction (left side in FIG. 2). The headstock 42 rotatably supports the workpiece W around the Z axis. The workpiece W is rotated by driving a motor 42a provided on the headstock 42. The tailstock 43 is located on the upper surface of the bed 41 at a position facing the headstock 42 in the Z-axis direction, that is, the front side in the X-axis direction (lower side in FIG. 2) and the other end side in the Z-axis direction (the lower side in FIG. 2). It is provided on the right side of FIG. 2). As a result, the geographic structure W is rotatably supported at both ends by the headstock 42 and the tailstock 43.

The traverse base 44 is provided on the upper surface of the bed 41 so as to be movable in the Z-axis direction. The traverse base 44 is moved by driving a motor 44a provided on the bed 41. The grindstone base 45 is provided on the upper surface of the traverse base 44 so as to be movable in the X-axis direction. The grindstone base 45 is moved by driving a motor 45a provided on the traverse base 44. The grindstone wheel 46 is rotatably supported by the grindstone base 45. The grindstone wheel 46 is rotated by driving a motor 46a provided on the grindstone base 45. The grindstone 46 is configured by fixing a plurality of abrasive grains with a bond material.

The sizing device 47 measures the dimension (diameter) of the workpiece W. The sizing device 47 is provided on the upper surface of the bed 41 so as to be movable in the Z-axis direction. The position of the sizing device 47 in the Z-axis direction is controlled by the feed mechanism 47a provided on the bed 41. The sizing device 47 includes a detection stylus 47b that comes into contact with the ground surface of the workpiece W. The detection stylus 47b is always in contact with the ground surface of the rotating workpiece W in the grinding process, and it is necessary to suppress wear due to friction with the ground surface. Therefore, the diameter of the detection stylus 47b is formed so as to suppress wear, and specifically, the diameter is larger (about several mm) than the diameter of the small diameter stylus of the roughness meter 2 (about several μm). It is formed.

Further, in the sizing device 47, an acceleration sensor 47c is provided on an arm that supports the detection stylus 47b. The acceleration sensor 47c outputs acceleration data representing the acceleration detected in a state where the center of the detection stylus 47b is in contact with the ground surface of the rotating workpiece W. Instead of using the acceleration sensor 47c, displacement data representing the displacement value (corresponding to the surface roughness) detected when the center of the detection stylus 47b is in contact with the ground surface of the rotating workpiece W is used. It is also possible to use a displacement sensor that outputs.

The grindstone correction device 48 corrects the shape of the grindstone 46. The grindstone correction device 48 is a device for performing the growing of the grindstone 46. The grindstone correction device 48 may be a device for dressing the grindstone 46 in addition to or in place of the trueing. Here, the trueing is a reshaping work, a work of forming the grindstone 46 according to the shape of the workpiece W when the grindstone 46 is worn by grinding, and a work of removing the runout of the grindstone 46 due to uneven wear. be. Dressing is a work of sharpening (sharpening), adjusting the amount of protrusion of abrasive grains, and creating a cutting edge of abrasive grains. Dressing is the work of correcting fouling, clogging, spillage, etc., and is usually performed after truing.

The coolant device 49 supplies the coolant to the grinding point of the workpiece W by the grindstone 46. The coolant device 49 cools the recovered coolant to a predetermined temperature and supplies it to the grinding point again.

The control device 12 provided in the cylindrical grinding machine 40 is based on an NC program generated based on operation command data such as the shape of the workpiece W, the machining conditions, the shape of the grindstone 46, and the coolant supply timing information. Control the drive unit. That is, the control device 12 inputs the operation command data and generates an NC program based on the operation command data.

Then, the control device 12 grinds the workpiece W by controlling the motors 42a, 44a, 45a, 46a, the coolant device 49, and the like based on the NC program. In particular, the control device 12 grinds the workpiece W until it has a finished shape based on the diameter of the workpiece W measured by the sizing device 47. Further, the control device 12 corrects the grindstone 46 (truing and dressing) by controlling the motors 42a, 44a, 45a, 46a, the grindstone repair device 48, and the like at the timing of modifying the grindstone 46. I do.

(5. Example of detector 13)
In this example, an example is a chatter evaluation device 50 in which a detector 13 integrally provided on a cylindrical grinding machine 40 evaluates chatter vibration generated according to the surface roughness of a ground surface of a workpiece W in grinding. Listed in. The detector 13 can also be configured by using a displacement sensor or an acceleration sensor to output displacement data or acceleration data (vibration data).

As shown in FIG. 3, the chatter evaluation device 50 mainly includes an actual data acquisition unit 51, a displacement conversion unit 52, a surface roughness data generation unit 53, and an output unit 54.

The actual data acquisition unit 51 acquires the acceleration data detected by the acceleration sensor 47c provided in the sizing device 47 in time series. Here, the actual data acquisition unit 51 spirally moves the contact position of the workpiece W on the ground surface by the detection stylus 47b of the sizing device 47 for each predetermined angle of the workpiece W that rotates. Get time-series data about the spiral position of.

That is, the detection stylus 47b of the sizing device 47 moves in the Z-axis direction, which is the axial direction of the workpiece W, by the feed mechanism 47a in a state where the workpiece W is rotated by the grinding process. In this case, since the detection stylus 47b of the sizing device 47 is in contact with the ground surface of the workpiece W, the contact position between the center of the detection stylus 47b and the workpiece W is the ground surface of the workpiece W. It moves in a spiral trajectory on the top. Therefore, the actual data acquisition unit 51 acquires the detected acceleration data while the detection stylus 47b moves relative to the ground surface in a spiral shape.

The displacement conversion unit 52 FFT (Fast Fourier Transform) the acceleration data acquired from the acceleration sensor 47c in time series, and obtains data on acceleration having a rotation frequency component (specific frequency component) corresponding to the rotation speed of the grindstone wheel 46. Extract. Then, the displacement conversion unit 52 reverse-FFTs the data regarding the acceleration having the extracted specific frequency component. As a result, the displacement conversion unit 52 causes the displacement data (roughness data) regarding the displacement value of the detection stylus 47b of the sizing device 47 having a specific frequency component, that is, the unevenness (surface roughness) on the ground surface of the workpiece W. ). The specific frequency component is a frequency component that is an integral multiple of the rotation speed and the rotation speed of the grindstone 46.

Here, the cylindrical grinding machine 40 grinds the workpiece W while rotating the grindstone 46. Therefore, the surface shape of the grindstone 46 is transferred at each rotation cycle of the grindstone 46, that is, at each rotation speed, and appears on the ground surface of the workpiece W. Specifically, when there are large protruding abrasive grains on the surface of the grindstone 46, a recess is formed on the ground surface of the workpiece W in which the portion in contact with the abrasive grains is largely scraped off. In this case, the recesses formed in the workpiece W are formed at equal intervals in the rotation direction, and the spacing between the recesses in the circumferential direction of the workpiece W coincides with the rotation cycle (for each rotation number) of the grindstone wheel 46. Therefore, by extracting the data related to the acceleration having a specific frequency component, it is possible to extract the unevenness, that is, the surface roughness on the ground surface of the workpiece W.

The surface roughness data generation unit 53 generates a series of surface roughness data using the roughness data (displacement data) in the circumferential direction on the ground surface of the workpiece W. Specifically, the plurality of circumferential roughness data generated by the displacement conversion unit 52 are generated for each angle at which the angles of the workpiece W with respect to the rotation axis are different from each other. That is, as shown in FIG. 4, the roughness data in the circumferential direction at the adjacent axial positions is the data of the positions deviated from each other in the circumferential direction of the workpiece W. Roughness data, that is, unevenness (surface roughness) on the ground surface of the work W is repeatedly observed on the ground surface of the work W every rotation cycle of the grindstone 46, as described above. Therefore, the surface roughness data generation unit 53 moves each roughness data for each different angle in the circumferential direction (the direction of the arrow shown in FIG. 4). As a result, as shown in FIG. 5, the surface roughness data generation unit 53 arranges the roughness data for each different angle in parallel in the axial direction with the same angle of the workpiece W. Generate a series of surface roughness data.

Here, when the roughness data (displacement data) in the circumferential direction obtained by converting from the acceleration data acquired in a spiral shape is divided, the unevenness represented by each of the divided roughness data may shift. Therefore, when the surface roughness data generation unit 53 generates the surface roughness data, the unevenness at the end points of the respective roughness data is continuous along the axial direction (Z-axis direction) of the workpiece W. The relative position of each roughness data is corrected so that it has the property. Then, the surface roughness data generation unit 53 arranges each of the position-corrected roughness data in the axial direction to generate the surface roughness data as state data.

By the way, the workpiece W of this example has a cylindrical shape (or a columnar shape) having a peripheral surface around the Z axis. Therefore, when each roughness data is arranged in parallel in the axial direction to generate a series of surface roughness data, it is necessary to take into consideration the influence of the change in the outer diameter of the ground surface of the workpiece W. Therefore, the surface roughness data generation unit 53 combines the surface roughness data and the outer diameter data representing the outer diameter of the workpiece W to add the influence of the outer diameter change to the surface roughness data. To generate.

That is, the planar roughness data generation unit 53 acquires the outer diameter data based on the signal from the sizing device 47 (detection transducer 47b), and FFT (Fast Fourier Transform) the acquired outer diameter data. Then, the planar roughness data generation unit 53 extracts the outer diameter data having a low frequency component in a specific frequency region (for example, 50 Hz or less) that does not include the frequency component of chatter vibration from the FFT outer diameter data. do. As a result, the surface roughness data generation unit 53 extracts the change in the outer diameter of the workpiece W in the axial direction.

Then, the surface roughness data generation unit 53 reverse-FFTs the extracted outer diameter data having the low frequency component, and synthesizes the reverse FFT outer diameter data and the surface roughness data, that is, the circumference of the workpiece W. The unevenness in the direction (surface roughness) and the change in the outer diameter in the axial direction are combined. As a result, the surface roughness data generation unit 53 accurately generates unevenness (surface roughness), that is, surface roughness data on the ground surface of the workpiece W in consideration of the change in the outer diameter of the workpiece W. be able to.

The output unit 54 outputs the surface roughness data generated by the surface roughness data generation unit 53 to the estimation calculation device 20 as state data. The output unit 54 can also output image data that can be generated based on the surface roughness data generated by the surface roughness data generation unit 53.

(6. Functional block configuration of surface roughness estimation system 1)
The functional block of the surface roughness estimation system 1 will be described with reference to FIG. The surface roughness estimation system 1 includes a control device 12, a detector 13 (chatter evaluation device 50), an estimation calculation device 20, and a display device 30.

As described above, the detector 13, that is, the chatter evaluation device 50 outputs the surface roughness data which is the observable state data in the cylindrical grinding machine 40 during the grinding process of the workpiece W. The surface roughness data is detected and output from the start to the end of machining in one workpiece W for each grinding process, that is, the rough grinding process, the fine grinding process, the fine grinding process, and the spark-out process. Will be done.

Further, the control device 12 of the cylindrical grinding machine 40 includes a counter. The counter counts the number of workpieces W ground by the cylindrical grinding machine 40. In this example, the counter counts the number of times the workpiece W has been machined since the grindstone repairing device 48 slewed or dressed the grindstone 46. In addition to the control device 12, the counter may be provided by the detector 13 itself or by the roughness meter 2.

In the corresponding cylindrical grinding machine 40, the estimation calculation device 20 uses a line roughness curve approximate waveform in an arbitrary axial direction based on the surface roughness data acquired from the detector 13, that is, the chatter evaluation device 50 (FIG. 11 described later). (See) is obtained, and the estimated roughness value on the ground surface is calculated. The estimation calculation device 20 includes a line roughness curve conversion unit 81, a correlation derivation unit 82, an estimation data acquisition unit 83, a roughness value estimation unit 84, and an output unit 85.

As shown in FIG. 7, the line roughness curve conversion unit 81 connects the actual displacement values measured by the small diameter stylus of the roughness meter 2 along the axial direction of the workpiece W (hereinafter, line roughness curve). The "solid line roughness curve LB") is converted into a line roughness curve when measured by the detection stylus 47b of the sizing device 47. Therefore, as shown in FIG. 6, the line roughness curve conversion unit 81 includes a solid line roughness curve acquisition unit 81a, an envelope calculation unit 81b, an intersection calculation unit 81c, and a center locus calculation unit 81d.

As shown in FIG. 6, the solid line roughness curve acquisition unit 81a acquires line roughness curve data representing the solid line roughness curve LB from the roughness meter 2 arranged outside the machine of the cylindrical grinding machine 40. Here, as described above, the diameter of the small diameter stylus of the roughness meter 2 is about several μm. As a result, the roughness meter 2 can accurately measure the unevenness of the ground surface in the axial direction of the workpiece W, that is, the actual displacement value, and as shown in FIG. 7, the surface roughness of the ground surface can be accurately measured. The solid line roughness curve LB reflected in can be obtained.

The envelope calculation unit 81b is the diameter of the detection stylus 47b of the sizing device 47 in the solid line roughness curve LB represented by the line roughness curve data acquired by the solid line roughness curve acquisition unit 81a from the roughness meter 2. Calculate the envelope using the same circle as (a few mm). Specifically, as shown in FIG. 8, the envelope calculation unit 81b has a solid line roughness curve LB (indicated by a thick broken line in FIG. 8) and a measurement interval line of the roughness meter 2 (two-dot chain line in FIG. 8). Circles C1 to C7 centered on the intersection points P1 to P7, that is, the actual displacement value, and the diameter of the detection stylus 47b are provided at the intersection points P1 to P7. Subsequently, the envelope calculation unit 81b calculates the envelopes L1 to L6 tangent to the adjacent circles for each of the circles C1 to C7.

The intersection calculation unit 81c calculates the intersections Q1 to Q7 between the envelopes L1 to L6 calculated by the envelope calculation unit 81b and the measurement interval line. By the way, the envelopes L1 to L6 may intersect at two points with respect to the same measurement interval line. For example, in FIG. 8, the measurement interval line on which the intersection Q5 is set intersects the envelope L4 and the envelope L5. In this case, in the envelope calculation unit 81b, the side of the measurement interval line where the displacement value when the detection stylus 47b measures is large, that is, the envelope L4 on the upper side in FIG. 8 intersects the measurement interval line. The side is the intersection Q5.

The center locus calculation unit 81d has a line connecting the intersections Q1 to Q7 calculated by the intersection calculation unit 81c, that is, the center of the large-diameter detection stylus 47b represented by a circle as shown by the thick solid line in FIG. The central locus line LC representing the locus of the detection stylus 47b when moving along the solid line roughness curve LB is calculated. Here, as shown in FIG. 9, the central locus line LC represents the roughness value by being associated with the solid line roughness curve LB (indicated by the broken line in FIG. 9). Therefore, in the following description, the roughness value represented by the central locus line LC is referred to as "center locus roughness value".

The correlation derivation unit 82 measures the actual roughness value based on the actual displacement value measured by the small diameter stylus of the roughness meter 2 and the center locus roughness value based on the above-mentioned center locus line LC, that is, the detection stylus 47b. Calculate the correlation with the central locus roughness value corresponding to the case of. As shown in FIG. 6, the correlation derivation unit 82 includes an actual roughness value acquisition unit 82a, a center locus roughness value calculation unit 82b, and a roughness value relational expression calculation unit 82c.

The actual roughness value acquisition unit 82a acquires the actual roughness value based on the actual displacement value measured by the roughness meter 2. Here, in this example, as the actual roughness value, for example, the case where the arithmetic mean roughness (Ra) calculated from each curve (displacement value of the measurement point) is used is illustrated, but the ten-point average roughness (Rz) is illustrated. Etc. may be used.

The center locus roughness value calculation unit 82b calculates the center locus roughness value based on the center locus data representing the center locus line LC calculated by the center locus calculation unit 81d. That is, the center locus roughness value calculation unit 82b calculates the center locus roughness value corresponding to the actual roughness value acquired by the actual roughness value acquisition unit 82a.

The roughness value relational expression calculation unit 82c calculates the correlation between the actual roughness value acquired by the actual roughness value acquisition unit 82a and the center locus roughness value calculated by the center locus roughness value calculation unit 82b. That is, the roughness value relational expression calculation unit 82c correlates the actual roughness value and the center locus roughness value corresponding to each other by, for example, performing statistical calculation processing. Guide the relationship. As a result, as shown in FIG. 10, for example, the correlation between the actual roughness value and the central locus roughness value is derived by a linear function (y = ax + b). However, the correlation between the actual roughness value and the center locus roughness value is not limited to the linear function, but may be a correlation represented by a quadratic function, a cubic function, or another function. Needless to say.

The estimation data acquisition unit 83 acquires estimation data based on the surface roughness data which is the state data output from the detector 13, that is, the chatter evaluation device 50. As shown in FIG. 6, the estimation data acquisition unit 83 includes an approximate waveform acquisition unit 83a and an approximate waveform roughness value derivation unit 83b. The approximate waveform acquisition unit 83a acquires the surface roughness data output from the chatter evaluation device 50, and as shown in FIG. 11, shows an arbitrary axial direction (thick solid line in FIG. 5) in the acquired surface roughness data. The line roughness approximate waveform LA in (see) is acquired.

The approximate waveform roughness value derivation unit 83b derives the approximate waveform roughness value by calculating the approximate waveform roughness value based on the line roughness approximate waveform LA acquired by the approximate waveform acquisition unit 83a. The approximate waveform roughness value deriving unit 83b calculates, for example, the arithmetic mean roughness (Ra) of the line roughness approximate waveform LA.

Here, the line roughness approximate waveform LA is based on the surface roughness data detected by the detection stylus 47b of the sizing device 47. That is, the line roughness approximate waveform LA is the line roughness of the ground surface detected when the center of the detection stylus 47b of the sizing device 47 comes into contact with the ground surface of the workpiece W in an arbitrary axial direction. It is an approximation of (surface roughness). Therefore, the approximate waveform roughness value derived (calculated) by the approximate waveform roughness value derivation unit 83b can be regarded as the central locus roughness value.

The roughness value estimation unit 84 is a correlation derivation unit with the approximate waveform roughness value derived (calculated) by the approximate waveform roughness value derivation unit 83b of the estimation data acquisition unit 83, that is, the center locus roughness value. The estimated roughness value corresponding to the center locus roughness value (approximate waveform roughness value) is estimated (converted) by using the correlation derived by the roughness value relational expression calculation unit 82c of 82. Here, the estimated roughness value corresponds to the actual roughness value as shown in FIG. That is, the estimated roughness value corresponds to the roughness value measured by the roughness meter 2 having a small diameter of the stylus, and the central locus roughness value (approximate waveform roughness value) using the detection stylus 47b having a large diameter. More accurate than. Then, the output unit 85 outputs the estimated roughness value estimated (converted) by the roughness value estimation unit 84 to the display device 30.

The display device 30 displays the estimated roughness value output from the output unit 85 of the estimation calculation device 20. As a result, the operator or the control device 12 of the cylindrical grinding machine 40 (grinding machine 11) can confirm the quality of the workpiece W to be machined in the grinding process (in the in-process).

As can be understood from the above description, according to the surface roughness estimation system 1, more specifically, the detection stylus 47b of the sizing device 47 than the detector 13 provided on the cylindrical grinding machine 40 which is the grinding device 10. Roughness value (surface roughness data output from the chatter evaluation device 50) measured by using The correlation with the actual roughness value can be derived. As a result, even when the surface roughness of the ground surface of the workpiece W is measured using the detection stylus 47b, which corresponds to a large-diameter stylus with a large diameter, grinding is performed based on the derived correlation. The surface roughness, that is, the actual roughness value on the ground surface of the workpiece W being machined can be estimated accurately.

(7. First alternative example)
In the above-mentioned example, the correlation deriving unit 82 of the estimation arithmetic unit 20 derives the correlation between the central locus roughness value and the displacement value measured by the roughness meter 2.

In the first alternative example, the amplitude of each frequency is extracted as a feature amount by performing frequency analysis on the calculated center locus line LC. Then, in the first alternative example, the correlation between the extracted amplitude and the actual roughness value based on the actual displacement value measured by the roughness meter 2 is learned by machine learning, and the trained model generated by machine learning is used. Estimate the estimation data using.

(7-1. Details of the configuration of the surface roughness estimation system 100 of the first alternative example)
The configuration of the surface roughness estimation system 100 of the first alternative example will be described in detail with reference to FIG. The surface roughness estimation system 100 is different from the surface roughness estimation system 1 of this example described above in that it includes an arithmetic unit 60 and a common display device 70. The arithmetic unit 60 is a learning processing device 60. The learning processing device 60 has a so-called server function, and is installed in the above-mentioned estimation calculation device 20 and a plurality of grinding devices 10 (cylindrical grinding machines 40), for example, in other factories separated from each other. It is communicatively connected to the grinding device 10. Although the estimation calculation device 20 and the learning processing device 60 are shown as independent configurations as a first alternative example in FIG. 12, they can also be one device. The common display device 70 is arranged with respect to the learning processing device 60.

The learning processing device 60 includes a processor 61, a storage device 62, an interface 63, and the like. The learning processing device 60 performs machine learning based on the line roughness curve data output from the roughness meter 2 and the actual roughness value based on the actual displacement output from the roughness meter 2. Then, the learning processing device 60 generates a trained model for predicting the estimation data.

In particular, the learning processing device 60 calculates the amplitude for each frequency extracted by performing FFT (Fast Fourier Transform), which is a frequency analysis, on the center locus data representing the center locus line LC calculated based on the line roughness curve data. It is a feature quantity. Then, the learning processing device 60 generates a trained model using the extracted feature amount and the actual roughness value.

(7-2. Details of learning processing device 60)
As shown in FIG. 13, the learning processing device 60 includes a training data set acquisition unit 91, a training data set storage unit 92, and a model generation unit 93.

The training data set acquisition unit 91 acquires a training data set for performing machine learning. The training data set acquisition unit 91 includes a solid line roughness curve acquisition unit 91a, an envelope calculation unit 91b, an intersection point calculation unit 91c, a center locus calculation unit 91d, a learning frequency analysis unit 91e, and an actual roughness value acquisition unit 91f.

The solid line roughness curve acquisition unit 91a acquires line roughness curve data from the roughness meter 2. The envelope calculation unit 91b detects the sizing device 47 in the solid line roughness curve LB (see FIG. 7) represented by the line roughness curve data acquired by the solid line roughness curve acquisition unit 91a from the roughness meter 2. The envelope is calculated using the circle corresponding to the diameter of the stylus 47b (see FIG. 8). The intersection calculation unit 91c calculates the intersections Q1 to Q7 between the envelopes L1 to L6 calculated by the envelope calculation unit 91b and the measurement interval line (see FIG. 8). The center locus calculation unit 91d calculates the center locus line LC connecting the intersections Q1 to Q7 calculated by the intersection calculation unit 91c (see FIG. 9).

Here, the solid line roughness curve acquisition unit 91a, the envelope calculation unit 91b, the intersection point calculation unit 91c, and the center locus calculation unit 91d are the solid line roughness curve acquisition units 81 constituting the line roughness curve conversion unit 81 of the estimation calculation device 20. The same processing as that of 81a, the envelope calculation unit 81b, the intersection calculation unit 81c, and the center locus calculation unit 81d is performed. In this example, the solid line roughness curve acquisition unit 91a, the envelope calculation unit 91b, the intersection calculation unit 91c, the center locus calculation unit 91d, the solid line roughness curve acquisition unit 81a, and the envelope of the training data set acquisition unit 91. The calculation unit 81b, the intersection calculation unit 81c, and the central locus calculation unit 81d will be described as separate elements.

However, the solid line roughness curve acquisition unit 91a, the envelope calculation unit 91b, the intersection point calculation unit 91c, and the center locus calculation unit 91d of the training data set acquisition unit 91 are combined with the solid line roughness curve acquisition unit 81a, the envelope calculation unit 81b, and the intersection point. It can also be used as the calculation unit 81c and the center locus calculation unit 81d. That is, the functions of the elements 91a, 91b, 91c, 91d in the learning processing device 60 are also used as a part of the functions of the estimation calculation device 20.

The learning frequency analysis unit 91e performs frequency analysis (specifically, FFT (Fast Fourier)) on the center locus data representing the center locus line LC calculated by the center locus calculation unit 91d, that is, the center locus data having frequency characteristics. Convert)). As a result, as shown in FIG. 14, the learning frequency analysis unit 91e extracts the amplitudes of the plurality of frequency components included in the central locus data as feature quantities which are explanatory variables in machine learning.

The actual roughness value acquisition unit 91f acquires the actual roughness value due to the actual displacement measured by the roughness meter 2. Here, as the actual roughness value, for example, the arithmetic mean roughness (Ra) is used. However, ten-point average roughness (Rz) or the like can also be used.

The training data set storage unit 92 stores the training data set acquired by the training data set acquisition unit 91. Specifically, the training data set storage unit 92 contains amplitude data representing the amplitudes of each of the plurality of frequency components extracted by the learning frequency analysis unit 91e, and the actual roughness acquired by the actual roughness value acquisition unit 91f. Store in association with the value.

The model generation unit 93 performs machine learning using the training data set stored in the training data set storage unit 92. Specifically, the model generation unit 93 performs machine learning using each amplitude, which is a feature quantity extracted by the learning frequency analysis unit 91e, as an explanatory variable and an actual roughness value as an objective variable. Then, the model generation unit 93 generates a trained model for deriving the correlation, that is, deriving the estimated roughness value.

As shown in FIG. 15, the estimation arithmetic unit 20 in the first alternative example is partially modified from the estimation arithmetic unit 20 of this example described above. That is, in the estimation calculation device 20 in the first alternative example, the line roughness curve conversion unit 81 and the correlation derivation unit 82 described in this example are omitted. Then, in the first alternative example, the configuration of the estimation data acquisition unit 83 is changed, so that the function as the correlation derivation unit is exhibited.

Specifically, the estimation data acquisition unit 83 of the first alternative example is provided with a model storage unit 83c that stores a learned model for deriving the correlation generated by the learning processing device 60. Further, the estimation data acquisition unit 83 of the first alternative example is provided with a frequency analysis unit 83d instead of the approximate waveform roughness value derivation unit 83b.

The frequency analysis unit 83d performs frequency analysis (specifically, FFT (Fast Fourier Transform)) on the line roughness approximate waveform LA acquired by the approximate waveform acquisition unit 83a, and extracts the amplitude for each frequency as a feature quantity. do. The roughness estimation unit 84 derives an estimated roughness value using each amplitude (feature amount) extracted by the frequency analysis unit 83d and the learned model stored in the model storage unit 83c.

That is, in the first alternative example, the frequency analysis unit 83d of the estimation data acquisition unit 83 extracts the amplitude, which is a feature amount, by frequency analysis and outputs it to the roughness value estimation unit 84. Then, the roughness value estimation unit 84 predicts the estimated roughness value by using the trained model generated by the learning processing device 60 as the correlation derivation unit and the amplitude which is the output feature amount. Can be done. Therefore, even in this case, by using the trained model generated by the learning processing device 60, the estimated roughness value can be estimated in the same manner as in this example described above, so that the same effect as in this example described above can be obtained. can get.

(9. Others)
In this example and the first alternative example described above, the surface roughness data is acquired as the state data from the chatter evaluation device 50 as the detector 13. However, the state data output from the detector 13 is not limited to the surface roughness data. For example, the workpiece W is not rotated but moved only in the Z-axis direction, and is detected by the sizing device 47. It may be line roughness data that can be acquired based on the displacement data.

1,100 ... Surface roughness estimation system, 2 ... Roughness meter, 10 ... Grinding device, 11 ... Grinding machine, 12 ... Control device, 13 ... Detector, 14 ... Interface, 20 ... Estimating calculation device, 21 ... Processor, 22 ... storage device, 23 ... interface, 30 ... individual display device, 40 ... cylindrical grinder, 41 ... bed, 42 ... headstock, 42a ... motor, 43 ... tailstock, 44 ... traverse base, 44a ... motor, 45 ... Grinding table, 45a ... Motor, 46 ... Grinding wheel (grind T), 46a ... Motor, 47 ... Dimensioning device, 47a ... Feed mechanism, 47b ... Detection stylus, 47c ... Acceleration sensor, 48 ... Grinding wheel correction device, 49 ... Coolant device, 50 ... Chatter evaluation device, 51 ... Actual data acquisition unit, 52 ... Displacement conversion unit, 53 ... Surface roughness data generation unit, 54 ... Output unit, 60 ... Learning processing device (correlation derivation unit) , 61 ... Processor, 62 ... Storage device, 63 ... Interface, 70 ... Common display device, 81 ... Line roughness curve conversion unit, 81a ... Solid line roughness curve acquisition unit, 81b ... Entourage line calculation unit, 81c ... Intersection point calculation unit , 81d ... Central locus calculation unit, 82 ... Correlation derivation unit, 82a ... Actual roughness value acquisition unit, 82b ... Center locus roughness value calculation unit, 82c ... Roughness value relational expression derivation unit, 83 ... Estimating data acquisition Unit, 83a ... Approximate curve acquisition unit, 83b ... Approximate waveform roughness value derivation unit, 83c ... Model storage unit (learned model storage unit), 83d ... Frequency analysis unit, 84 ... Roughness value estimation unit, 85 ... Output unit , 91 ... Training data set acquisition unit, 91a ... Solid line roughness curve acquisition unit, 91b ... Entourage line calculation unit, 91c ... Intersection point calculation unit, 91d ... Center locus calculation unit, 91e ... Learning frequency analysis unit, 92 ... Training data Set storage unit, 93 ... Model generation unit, LA ... Line roughness approximation curve, LB ... Solid line roughness curve, LC ... Center locus line, P1 to P7 ... Intersection, Q1 to Q7 ... Intersection, L1 to L6 ... Envelopment line, C1 to C7 ... circle, T ... grindstone, W ... workpiece

Claims (15)

A grinding device in which the grindstone grinds the workpiece under the control of the control device,
It has a detection stylus that comes into contact with the ground surface of the workpiece, and the surface roughness of the ground surface of the workpiece depends on the operation of the detection stylus that can be observed in the grinding process by the grinding device. A detector that detects and outputs state data related to
The roughness value representing the surface roughness of the ground surface of the work piece when measured using the detection stylus, and the work when measurement is performed using a small diameter stylus having a diameter smaller than that of the detection stylus. A correlation derivation unit that derives a correlation with the actual roughness value representing the surface roughness of the ground surface of the object,
An estimation data acquisition unit that acquires estimation data representing the roughness value based on the state data output from the detector, and an estimation data acquisition unit.
A roughness value estimation unit that estimates the actual roughness value using the estimation data and the correlation, and a roughness value estimation unit.
A surface roughness estimation system. A line roughness curve conversion unit that converts a solid line roughness curve that connects actual displacement values measured using the small diameter stylus into a line roughness curve that connects the displacement values measured using the detection stylus. With
The correlation derivation unit is
The surface roughness estimation system according to claim 1, wherein the correlation between the roughness value obtained from the converted line roughness curve and the actual roughness value is derived. The line roughness curve conversion unit is
A solid line roughness curve acquisition unit that acquires the solid line roughness curve, and
In the solid line roughness curve, the envelope is calculated using a circle centered on a point having the same diameter as the detection stylus and representing each of the actual displacement values on the solid line roughness curve. Envelope calculation unit and
An intersection calculation unit that calculates the displacement value represented by the intersection of the calculated envelope and each measurement interval line that represents the measurement interval when measuring the actual displacement value.
A central locus calculation unit that connects the calculated displacement values and calculates a central locus line representing the locus of the circle when the center of the circle moves on the solid line roughness curve.
With
The correlation derivation unit is
The surface roughness estimation system according to claim 2, wherein the correlation between the calculated roughness value of the center locus line and the actual roughness value is derived. The intersection calculation unit
The surface roughness estimation system according to claim 3, wherein when two envelopes intersect the same measurement interval line, the side having a large displacement value is calculated as the intersection. The estimation data acquisition unit
An approximate waveform acquisition unit that acquires a line roughness approximate waveform representing the line roughness of the workpiece based on the state data, and an approximate waveform acquisition unit.
An approximate waveform roughness value deriving unit that derives the roughness value of the acquired approximate waveform as the estimation data, and
The surface roughness estimation system according to any one of claims 1-4. The estimation data acquisition unit
An approximate waveform acquisition unit that acquires a line roughness approximate waveform representing the line roughness of the workpiece when measuring using the detection transducer based on the state data, and an approximate waveform acquisition unit.
Machine learning using the explanatory variable and the training data set including the objective variable, with the feature amount of the line roughness curve as the explanatory variable and the actual roughness value as the objective variable when measuring using the detection stylus. A trained model storage unit that stores the trained model generated by performing
A roughness value deriving unit that derives the roughness value predicted using the feature amount extracted from the approximate waveform and the trained model, and a roughness value deriving unit.
The surface roughness estimation system according to any one of claims 1-4. The approximate waveform and the line roughness curve have frequency characteristics and have frequency characteristics.
The surface roughness estimation system according to claim 6, further comprising a frequency analysis unit that extracts the feature amount by frequency-analyzing the approximate waveform and the line roughness curve. The seventh aspect of claim 7, wherein the frequency analysis unit divides the approximate waveform and the line roughness curve into a plurality of frequency components by frequency analysis, and extracts the amplitude of each of the plurality of frequency components as the feature amount. Surface roughness estimation system. Line roughness curve conversion that converts the solid line roughness curve that connects the actual displacement values measured using the small diameter stylus into the line roughness curve that connects the displacement values when measuring using the detection stylus. Department and
It is provided with a learning frequency analysis unit that extracts the feature amount for learning by frequency-analyzing the line roughness curve converted by the line roughness curve conversion unit, and is extracted by the learning frequency analysis unit. Learning to generate the trained model by performing machine learning using the explanatory variable and the training data set including the explanatory variable and the objective variable with the feature amount as the explanatory variable and the actual roughness value as the objective variable. The surface roughness estimation system according to any one of claims 6 to 8, further comprising a completed model generator. The surface roughness estimation system according to any one of claims 1 to 9, wherein the detector is integrally provided with the grinding device. The detection stylus
The surface roughness estimation system according to claim 10, which is included in the sizing device provided in the grinding device. The state data is any one of claims 1-11, which is acceleration data or displacement data generated in the detection stylus corresponding to the surface roughness of the workpiece to which the grinding process has been performed. The surface roughness estimation system described in. The surface roughness estimation system according to claim 12, wherein the state data is calculated by statistically calculating the acceleration data or the displacement data at a plurality of locations of the workpiece. When the workpiece has a peripheral surface around the axis,
The surface roughness estimation system according to claim 12 or 13, wherein the state data includes outer diameter data representing the outer diameter of the workpiece detected by the detection stylus. When the workpiece has a peripheral surface around the axis,
The state data regards the acceleration data or the displacement data in a plurality of circumferential directions for each different angle of the workpiece as a plurality of the acceleration data or the displacement data at the same angle of the workpiece, and the outer diameter. The surface roughness estimation system according to claim 14, which is a plurality of the acceleration data or the displacement data obtained by synthesizing the data and arranging them in parallel in the axial direction of the workpiece.

Images