Driving condition determination device and driving condition determination method

Abstract

A numerical control device (1a), which is a drive condition determination device for determining a drive condition of a machine tool for performing vibration cutting in which a shaft for changing the relative position of a workpiece and a tool vibrates during cutting to break chips, is characterized by comprising: a search condition input unit (100) for acquiring a search condition; the search conditions include a search range set for at least one drive condition among a machining condition of the machine tool (3), a vibration condition of the shaft, and a control parameter that is a parameter that determines a control characteristic of the shaft, and a reference machining condition (11) that is a reference machining condition that is determined regardless of the presence or absence of vibration cutting; and a condition search unit (102) that determines, on the basis of the search condition, an execution drive condition, which is a drive condition to be used when vibration cutting is performed, within a range in which the search range is set so as to satisfy a breaking condition, which is a condition for the occurrence of breaking of chips.

Description

Technical Field

[0001] The present invention relates to a driving condition determination device and a driving condition determination method for determining driving conditions of a machine tool that performs vibration cutting. Background Art

[0002] The machine tool used in cutting processing removes material from the surface of the object to achieve the desired shape while causing the object to interfere with the tool. At this time, the removed material is discharged as chips. If long chips are generated, the chips will be entangled with the object, tool, etc., and the processing accuracy will be reduced, or the processing surface will be damaged, which will cause poor processing.

[0003] Patent document 1 discloses a control device for a machine tool, which vibrates an axis driving a tool at a low frequency to avoid generating long chips, thereby generating a moment when the tool does not interfere with the object to be processed, and performs vibration cutting to break the chips into short pieces. The control device disclosed in Patent document 1 determines whether chip breaking occurs when vibrating a specific axis among a plurality of feed axes based on tool data, a relative positional relationship between the object to be processed and the tool, and movement data for relatively moving the object to be processed and the tool, and automatically selects the axis to be vibrated.

[0004] Patent Document 1: Japanese Patent No. 7007531 Summary of the invention

[0005] However, according to the above-mentioned conventional technology, there is a problem in that since conditions other than the axis of vibration are fixed in advance, there is a high possibility that chip breakage will not occur regardless of which axis is selected.

[0006] The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to obtain a driving condition determination device capable of increasing the possibility of chip breakage occurring when a machine tool performs vibration cutting.

[0007] In order to solve the above-mentioned problems and achieve the purpose, the driving condition determination device involved in the present invention determines the driving conditions of a working machine that performs vibration cutting in which an axis that changes the relative positions of a processing object and a tool vibrates during cutting to break chips. The driving condition determination device is characterized in that it has: an exploration condition input unit, which obtains the exploration condition, the exploration condition including an exploration range set for at least one driving condition among the processing conditions of the working machine, the vibration conditions of the axis and the parameters that determine the control characteristics of the axis, i.e., the control parameters, and a processing condition that serves as a reference determined regardless of the presence or absence of vibration cutting, i.e., the reference processing condition; and a condition exploration unit, which determines the driving condition used when performing vibration cutting, i.e., the execution driving condition, based on the exploration condition, in a manner that satisfies the condition for the occurrence of chip breaking, i.e., the breaking condition, within the range set in the exploration range.

[0008] Effects of the Invention

[0009] The drive condition determination device according to the present invention has the effect of obtaining a drive condition determination device capable of increasing the possibility of chip breakage occurring when a machine tool performs vibration cutting. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a diagram showing the functional structure of the numerical control device involved in the first embodiment.

[0011] Figure 2 Is used with Figure 1 An explanatory diagram related to vibration cutting of a machine tool shown in FIG.

[0012] Figure 3 This is a diagram for explaining the mechanism of chip breakage during vibration cutting performed by a machine tool.

[0013] Figure 4 It is used for Figure 1 The flowchart shown is used to explain the operation of the condition search unit.

[0014] Figure 5 This is a diagram showing the functional structure of an information processing device according to the second embodiment.

[0015] Figure 6 Yes means Figure 5 A diagram of an example of a mechanical model is shown.

[0016] Figure 7 It is used for Figure 5 The flowchart for explaining the first example of the operation of the condition searching unit shown in FIG.

[0017] Figure 8 It is used for Figure 5The flowchart for explaining the second example of the operation of the condition searching unit shown in FIG.

[0018] Fig. 9 This is a diagram showing the relationship between the response amplitude and the command value of the vibration amplitude derived from the experiment.

[0019] Fig.10 This is a diagram showing the functional structure of an information processing device involved in Embodiment 3.

[0020] Fig.11 Yes means Fig.10 A diagram of an example of a mechanical model is shown.

[0021] Fig.12 It is used for Fig.10 The flowchart shown is used to explain the operation of the condition search unit.

[0022] Fig.13 This is a diagram showing the functional structure of an information processing device according to a variation of the third embodiment.

[0023] Fig.14 This is a diagram showing the functional structure of a numerical control device according to a fourth embodiment.

[0024] Fig.15 It is used for Fig.14 The flowchart shown is used to explain the operation of the condition search unit.

[0025] Fig.16 This is a diagram showing the functional structure of an information processing device involved in Embodiment 5.

[0026] Fig.17 It is used for Fig.16 The flowchart shown is used to explain the operation of the condition search unit.

[0027] Fig.18 This is a diagram showing the functional structure of a numerical control device according to the sixth embodiment.

[0028] Fig.19 This is a diagram showing the functional structure of an information processing device involved in Embodiment 7.

[0029] Fig. 20 is with Fig.19 A flowchart related to the learning process of the condition search unit is shown.

[0030] Fig.21 This is a diagram showing the functional structure of a numerical control device according to the eighth embodiment.

[0031] Fig. 22 It is used for Fig.21 A flowchart for explaining the estimation processing of the numerical control device shown.

[0032] Fig.23 This is a diagram showing dedicated hardware for realizing the functions of the numerical control devices involved in embodiments 1, 4, 6, and 8.

[0033] Fig.24 This is a diagram showing the structure of a control circuit for realizing the functions of the numerical control devices involved in embodiments 1, 4, 6, and 8.

[0034] Fig.25 This is a diagram showing a configuration example of a computer system that implements the information processing device involved in embodiments 2, 3, 5, and 7. DETAILED DESCRIPTION

[0035] Hereinafter, a driving condition determination device and a driving condition determination method according to an embodiment of the present invention will be described in detail with reference to the drawings.

[0036] Implementation method 1.

[0037] Figure 1 1 is a diagram showing a functional structure of a numerical control device 1 a according to Embodiment 1. The numerical control device 1 a has a function of outputting a drive axis command 17 and a spindle command 18 , which are command values ​​for controlling the machine tool 3 according to the machining program 10 , to the drive unit 2 , thereby controlling the machine tool 3 .

[0038] The drive unit 2 supplies a spindle current 20 and a servo motor current 22 for controlling the machine tool 3 to the machine tool 3 based on the drive shaft command 17 and the spindle command 18 outputted from the numerical control device 1a. The drive unit 2 can generate the spindle current 20 based on the spindle position 21 fed back from the machine tool 3, and can generate the servo motor current 22 based on the drive shaft position 23 fed back from the machine tool 3.

[0039] The machine tool 3 has a spindle 300 and a drive shaft 301, which are axes that change the relative positions of the object 303 and the tool 304, and the spindle 300 and the drive shaft 301 are driven by the spindle current 20 and the servo motor current 22 supplied from the drive unit 2, thereby performing cutting processing on the object 303. At this time, the machine tool 3 can vibrate at least one of the axes that change the relative positions of the object 303 and the tool 304, thereby performing vibration cutting that causes chip breakage.

[0040] The numerical control device 1a is an example of a drive condition determination device that determines the drive condition of the machine tool 3. The drive condition is a condition that specifies the drive state of the machine tool 3. The machining condition for machining the object to be machined into a desired shape, the vibration condition that specifies the vibration frequency and the vibration amplitude, and the control parameter that determines the control characteristics of the axis are each one of the drive conditions.

[0041] The numerical control device 1a includes a search condition input unit 100, a machining program input unit 101, a condition search unit 102, and a command value generation unit 104. The search condition input unit 100 acquires conditions for searching for driving conditions.

[0042] The search condition input unit 100 obtains the search condition, which includes a search range set for at least one driving condition among the machining condition, the vibration condition, and the control parameter, and a reference machining condition 11 that is set as a reference regardless of the presence or absence of vibration cutting. Specifically, in the first embodiment, the search condition input unit 100 obtains the reference machining condition 11, the reference vibration condition 12 that is a reference, and the vibration condition search range 14 that is a search range set for the vibration condition. The search condition input unit 100 outputs the obtained search condition to the condition search unit 102. The search condition input unit 100 is configured using input means such as a keyboard, a button, and a mouse, and receives input from an operator to the numerical control device 1a to obtain input information.

[0043] The condition search unit 102 searches for the driving condition based on the search condition output by the search condition input unit 100, and determines the driving condition actually used, that is, the execution driving condition. In the first embodiment, the execution driving condition includes the processing condition actually used, that is, the execution processing condition 15, and the vibration condition actually used, that is, the execution vibration condition 16. The detailed method of the condition search unit 102 searching for the driving condition will be described later. The condition search unit 102 outputs the determined execution processing condition 15 and execution vibration condition 16.

[0044] The machining program input unit 101 receives an input of a machining program 10 for controlling the machine tool 3 , thereby acquiring the machining program 10 , and outputs the acquired machining program 10 to the command value generating unit 104 .

[0045] The instruction value generating unit 104 generates an instruction to the drive shaft 301 of the working machine 3, namely, a drive shaft instruction 17, and an instruction to the spindle 300, namely, a spindle instruction 18, based on the processing program 10 output by the processing program input unit 101 and the execution processing conditions 15 and the execution vibration conditions 16 determined by the condition exploration unit 102, and outputs the generated drive shaft instruction 17 and spindle instruction 18 to the drive unit 2.

[0046] The driving unit 2 includes a main-spindle servo control unit 201 and a drive-axis servo control unit 202 .

[0047] The spindle servo control unit 201 performs position control so that the spindle position 21 of the spindle motor 302 of the machine tool 3 matches the spindle command 18 and outputs the spindle current 20. However, when the spindle command 18 is a speed command, the spindle servo control unit 201 performs speed control so that the spindle speed matches the spindle command 18.

[0048] The drive shaft servo control unit 202 performs position control so that the drive shaft position 23 of the servo motor 305 of the machine tool 3 is consistent with the drive shaft instruction 17, and outputs the servo motor current 22. However, in the figure, for simplicity, only one drive shaft 301 is shown, but when the machine tool 3 has more than or equal to two drive shafts 301, the drive unit 2 can have a plurality of drive shaft servo control units 202 corresponding to each of the plurality of drive shafts 301.

[0049] The machine tool 3 is a lathe-type machine, and includes a main spindle 300 , a drive shaft 301 , a main spindle motor 302 connected to the main spindle, and a servo motor 305 connected to the drive shaft 301 .

[0050] A tool 304 is mounted on the drive shaft 301, and a processing object 303 is set on the spindle 300. A spindle current 20 generated by the spindle servo control unit 201 of the drive unit 2 based on the spindle command 18 is supplied to the spindle motor 302. A servo motor current 22 generated by the drive shaft servo control unit 202 of the drive unit 2 based on the drive shaft command 17 is supplied to the servo motor 305. The spindle motor 302 drives the spindle 300 according to the spindle current 20, and the servo motor 305 drives the drive shaft 301 according to the servo motor current 22.

[0051] However, in Figure 1 In the machine tool 3 shown, the structure of the spindle 300 and the drive shaft 301 is an example of a lathe-type machine. In the case of turning processing using a 5-axis machine tool with a rotating axis, the rotary table for setting the processing object 303 and the drive shaft for driving the tool 304 are driven by the drive shaft servo control unit 202. In addition, the machine tool 3 is not limited to a machine for turning processing, and can also be a machine for milling processing. In addition, here, the processing object 303 set on the spindle 300 can be a rotating structure, but the tool 304 can also be rotated to perform vibration cutting. For example, a method of cutting the processing object 303 by the rotating tool 304 through drilling processing, milling processing, etc.

[0052] Figure 2Is used with Figure 1 1 and 1. The main spindle 300 (in Figure 2 The processing object 303 is rotated at a constant speed according to the spindle instruction 18 (not shown). The drive shaft 301 in the Z-axis direction controls the servo motor 305 of the Z-axis based on the drive shaft instruction 17, thereby causing the tool 304 to interfere with the processing object 303 and moving the tool 304 in the Z-axis direction along the tool path specified by the processing program 10. Here, the drive shaft instruction 17 includes an instruction for the feed amount specified by the execution processing condition 15 and a vibration instruction based on the amplitude and frequency of the execution vibration condition 16. Through the drive shaft instruction 17, the drive shaft 301 causes the tool 304 to vibrate while moving along the tool path, so that relative vibration occurs between the tool 304 and the processing object 303 to break the chips. Here, the tool 304 is vibrated by the drive shaft 301, but if relative vibration can occur between the tool 304 and the processing object 303, the spindle 300 can be used to vibrate the processing object 303, or both the drive shaft 301 and the spindle 300 can be used to vibrate both the tool 304 and the processing object 303.

[0053] In addition, relative vibration may be generated by vibration transmitted to the object 303 side due to the reaction force of the motor driving the tool 304. Furthermore, relative vibration may be generated by using a drive shaft 301 (not shown) driven in the X-axis direction orthogonal to the Z-axis, or by vibrating in the inclined direction while driving the X-axis and the Z-axis at the same time.

[0054] Figure 3 It is an explanatory diagram of the mechanism of chip breakage during vibration cutting performed by the machine tool 3 . Figure 3 The horizontal axis of φ shows the rotation angle of the main shaft 300. The rotation angle is set in units of 360 degrees for one turn and then returns to 0 degrees. Figure 3 The vertical axis shows the relative distance between the tool 304 and the object 303 in the direction of the tool path. When no vibration is applied, the tool 304 moves at a feed rate F per rotation specified by the tool path. When vibration is applied, the tool 304 is vibrated while being fed in the Z-axis direction, so that the relative distance between the object 303 and the tool 304 per rotation of the spindle 300 becomes a trajectory that vibrates in the shape of a sine curve. This trajectory is called a vibration trajectory.

[0055] exist Figure 3In the example of FIG. 1 , the vibration trajectory is shown when the drive shaft 301 in the Z-axis direction is vibrated 1.5 times during one rotation of the spindle 300. The vibration trajectory is obtained by the spindle speed S (r / min), the feed speed of the tool 304 relative to the processing object 303 per one rotation of the spindle, that is, the feed amount F (mm / r) per one rotation of the spindle, the vibration amplitude A (mm) of the tool 304, and the angular frequency ω of the vibration of the tool 304. v (rad / s) determined.

[0056] exist Figure 3 In the figure, the vibration trajectory of the nth spindle rotation, i.e., the nth vibration trajectory 110, and the vibration trajectory of the n+1th spindle rotation, i.e., the n+1th vibration trajectory 111, are shown. The nth vibration trajectory 110 is a vibration trajectory from the moment when the spindle 300 passes 0 degrees, i.e., the moment t(n), to the moment when the spindle rotates one circle, i.e., the moment t(n+1). The n+1th vibration trajectory 111 is a vibration trajectory from the moment t(n+1) to the moment when the spindle rotates one circle, i.e., the moment t(n). In addition, here, the moment t(n) is set as an arbitrary reference moment, for example, the reference moment can be the moment when the machining starts. The distance between the nth vibration trajectory 110 and the n+1th vibration trajectory 111 at the same spindle angle becomes the thickness of the chips, i.e., the chip thickness. Figure 3 The area indicated by the oblique line, i.e., the chip breaking area 112, is an area where the nth vibration trajectory 110 and the n+1th vibration trajectory 111 overlap. In this area, material has been removed from the surface of the object 303 by the nth vibration trajectory 110, so in the n+1th vibration trajectory 111, material is not removed, but air cutting occurs. In the area sandwiched by the air cutting area, the chips become discontinuous, so the chips are broken. Therefore, in order to cause chip breaking, it is preferable to set the driving conditions of the vibration cutting so that the air cutting area is generated.

[0057] Here, the conditions for the occurrence of chip breakage are described. As machining conditions that should be determined in turning regardless of the presence or absence of vibration cutting, the spindle speed S (r / min) and the feed rate F (mm / r) per rotation are given, and as vibration conditions, the angular frequency ω of the spindle rotation is given when the vibration amplitude A (mm) of the relative vibration between the tool 304 and the object 303 and the angular frequency ωv (rad / s) of the vibration are given. s (rad / s) is calculated by the following formula (1).

[0058] [Formula 1]

[0059]

[0060] In addition, let the angle θ (rad) of the spindle 300 be the trajectory T (θ) along which the blade tip of the tool 304 of the n-th vibration trajectory 110 passes. n The calculation is performed by the following equation (2): Here, it is assumed that “0<θ<2π”.

[0061] [Formula 2]

[0062]

[0063] When the n-th vibration trajectory 110 and the (n+1)-th vibration trajectory 111 come into contact, that is, when the following equation (3) holds true, the chips are broken.

[0064] [Formula 3]

[0065] T(θ) n =T(θ) n+1 …(3)

[0066] Therefore, the relationship of the following equation (4) is derived from equation (2) and equation (3): wherein φ is the phase difference between the nth vibration locus 110 and the (n+1)th vibration locus 111 .

[0067] [Formula 4]

[0068]

[0069] Here, among the conditions for the following equation (5) to be satisfied, there is θ for which equation (3) and equation (4) are satisfied.

[0070] [Formula 5]

[0071]

[0072] In particular, when the following equation (6) holds true, the nth vibration locus 110 and the (n+1)th vibration locus 111 are in contact with each other through a point.

[0073] [Formula 6]

[0074]

[0075] When equation (6) holds, the condition that the two vibration loci pass through a point and come into contact is calculated. Therefore, if the vibration amplitude A is larger than the value of the vibration amplitude A that satisfies equation (4), chip breakage occurs.

[0076] As described above, the conditions for the occurrence of chip breakage can be expressed by formulas based on the machining conditions and the vibration conditions. If the conditions shown in formulas (4) and (5) can be satisfied, the machining conditions and the vibration conditions can be freely selected. On the other hand, if a driving condition is set that does not satisfy either formula (4) or formula (5), since chip breakage does not occur, vibration cutting does not occur.

[0077] Here, the exploration of machining conditions using the numerical control device 1a is described. First, the operator uses the exploration condition input unit 100 to input the spindle speed and the feed rate per blade as the reference machining condition 11. However, if the angular frequency of the spindle rotation and the feed amount per rotation can be uniquely determined, the reference machining condition 11 can be information expressed in any form. For example, the reference machining condition 11 can include information that expresses the feed rate of the Z axis by the feed amount per minute (mm / min). At this time, in the exploration condition input unit 100, the feed rate per blade is calculated based on the input feed amount per minute, and the calculation result is sent to the condition exploration unit 102. In addition, the reference machining condition 11 can directly include the angular frequency of the spindle rotation instead of the spindle speed.

[0078] Next, the operator uses the search condition input unit 100 to input the vibration frequency and vibration amplitude as the reference vibration condition 12 as the search condition. In addition, the operator uses the search condition input unit 100 to input the vibration condition search range 14 as the search range of the vibration condition as the search condition. The vibration condition search range 14 includes the search range of the vibration frequency and the search range of the vibration amplitude. In addition, as long as the vibration frequency and the vibration amplitude can be uniquely determined for the reference vibration condition 12 and the vibration condition search range 14, the reference vibration condition 12 and the vibration condition search range 14 can also be information expressed in any form. For example, the reference vibration condition 12 and the vibration condition search range 14 can include the value of the vibration frequency as the number of vibrations per rotation of the spindle, and can also include the value of the vibration amplitude as a ratio relative to the feed speed per blade. When setting values ​​other than the vibration frequency and the vibration amplitude, the search condition input unit 100 converts the input value into the vibration frequency and the vibration amplitude, and outputs them to the condition search unit 102.

[0079] The condition search unit 102 determines the driving condition used when the processing is actually performed, i.e., the execution driving condition, based on the search condition output by the search condition input unit 100. In the first embodiment, the execution driving condition determined by the condition search unit 102 includes the processing condition used when the processing is actually performed, i.e., the execution processing condition 15, and the vibration condition used when the processing is actually performed, i.e., the execution vibration condition 16.

[0080] Figure 4 It is used for Figure 1 1 and 2. The condition search unit 102 first sets the reference processing condition 11 and reference vibration condition 12 included in the search condition output by the search condition input unit 100 (step S1). The condition search unit 102 determines whether the set processing condition and vibration condition satisfy the breaking condition (step S2).

[0081] When the breaking condition is not satisfied (step S2: No), the condition exploration unit 102 determines whether the vibration amplitude can be adjusted (step S3). When the vibration amplitude can be adjusted (step S3: Yes), the vibration amplitude is changed within the vibration amplitude exploration range (step S4). After step S4, the condition exploration unit 102 returns to the process of step S2. In this case, the condition exploration unit 102 repeats the process of step S2 to step S4 from changing the vibration amplitude to determine whether the breaking condition is satisfied as long as the vibration amplitude can be adjusted until the breaking condition is satisfied. For example, the condition exploration unit 102 changes the value of the vibration amplitude from the lower limit value of the vibration amplitude exploration range, and adds a certain value each time it is repeated, thereby being able to explore from the lower limit value to the upper limit value of the vibration amplitude exploration range. In this case, when the value of the vibration amplitude after the change in step S3 does not reach the upper limit value, the condition exploration unit 102 determines that the vibration amplitude can be adjusted, and when the value of the vibration amplitude after the change reaches the upper limit value, it determines that the vibration amplitude cannot be adjusted.

[0082] If the vibration amplitude cannot be adjusted (step S3: No), the condition search unit 102 changes the vibration frequency within the vibration frequency search range (step S5). At this time, the condition search unit 102 sets the vibration amplitude value to the lower limit value of the vibration amplitude search range and returns to the process of step S2.

[0083] When the breaking condition is satisfied (step S2: Yes), the condition search unit 102 outputs the set processing condition and vibration condition as the execution drive condition (step S6).

[0084] Thus, when the reference processing condition 11 and the reference vibration condition 12 satisfy the breaking condition, the condition exploration unit 102 directly outputs the reference processing condition 11 and the reference vibration condition 12 as the execution drive condition. When the reference processing condition 11 and the reference vibration condition 12 do not satisfy the breaking condition, first, the vibration condition satisfying the breaking condition is explored while changing the vibration amplitude by a certain value from the lower limit value to the upper limit value of the vibration amplitude exploration range. When no vibration condition satisfying the breaking condition is found even when the vibration amplitude reaches the upper limit value, the condition exploration unit 102 then changes the vibration frequency to the lower limit value and the vibration amplitude to the lower limit value, and then changes the vibration amplitude by a certain value from the lower limit value to the upper limit value again, while exploring the vibration condition satisfying the breaking condition. Regarding the vibration frequency, the process of changing it in units of a certain value from the lower limit to the upper limit while determining the breaking condition is repeated, whereby the condition exploration unit 102 can explore the combination of vibration amplitude and vibration frequency that meets the breaking condition within the exploration range.

[0085] As described above, according to embodiment 1, the numerical control device 1a is a driving condition determination device that determines the driving conditions of the working machine 3 for performing vibration cutting, wherein the vibration cutting causes the axis that changes the relative position of the processing object 303 and the tool 304 during the cutting process to vibrate and break the chips. The numerical control device 1a has: an exploration condition input unit 100, which obtains the exploration conditions, and the exploration conditions include the processing conditions for the working machine 3, the vibration conditions that specify the vibration frequency and vibration amplitude of the axis that causes at least one of the drive shaft 301 and the main shaft 300 to vibrate, and the parameters that determine the control characteristics of the axis, i.e., the exploration range set by at least one driving condition among the control parameters, and the processing conditions that serve as a reference, i.e., the reference processing conditions 11, which are determined regardless of the presence or absence of vibration cutting; and a condition exploration unit 102, which determines the driving conditions used when performing vibration cutting, i.e., the execution driving conditions, based on the exploration conditions, within the range of the exploration range set in a manner to satisfy the conditions for the occurrence of chip breaking, i.e., the breaking conditions. Therefore, even if the operator cannot properly determine the driving conditions of the working machine 3 when performing vibration cutting, the driving conditions will be explored within the set exploration range, and the driving conditions for chip breakage can be automatically determined, thereby increasing the possibility of chip breakage when the working machine performs vibration cutting.

[0086] Implementation method 2.

[0087] Figure 51 is a diagram showing the functional structure of an information processing device 1b according to Embodiment 2. The information processing device 1b is an example of a drive condition determination device that determines the drive conditions of the working machine 3. Hereinafter, the same reference numerals are given to the same components as those in Embodiment 1, and detailed descriptions thereof are omitted, and the parts different from Embodiment 1 are mainly described.

[0088] The information processing device 1b is installed inside the computer and determines the driving conditions, i.e., the machining conditions and the vibration conditions, when the machine tool 3 performs vibration cutting. The information processing device 1b takes the reference machining conditions 11, the reference vibration conditions 12, and the vibration condition search range 14 as inputs, and outputs the execution machining conditions 15 and the execution vibration conditions 16 to the machining program generation device 103.

[0089] The machining program generating device 103 is CAM (Computer Aided Manufacturing) software that generates an EIA program or a machining program 10 of a machine tool 3 called a G code based on machining path generation information. Alternatively, the machining program generating device 103 may be software that additionally edits information corresponding to instructions of a spindle speed, a feed rate per blade, a vibration frequency, and a vibration amplitude for a machining program that describes a machining path for machining under a reference machining condition 11 output by CAM software, and outputs the edited machining program 10.

[0090] The information processing device 1b has an exploration condition input unit 100 and a condition exploration unit 102b. The condition exploration unit 102b uses a mechanical model 105 that simulates the characteristics of the working machine 3 to determine the execution driving conditions. In the first embodiment, it is assumed that the drive shaft 301 vibrates with the same vibration amplitude as the vibration amplitude assigned to the drive shaft servo control unit 202. However, in the actual working machine 3, due to the influence of the mechanical characteristics of the mechanical structure supporting the drive shaft 301, the drive shaft 301 not shown, the spindle 300, etc., the amplitude of the relative vibration between the tool 304 and the processing object 303 is sometimes different from the specified vibration amplitude. In the above case, even if the specified vibration amplitude and vibration frequency meet the chip breaking condition, the vibration amplitude generated in the actual processing will be different from the command value, so sometimes the chip breaking does not occur. Therefore, the condition exploration unit 102b of the information processing device 1b uses the mechanical model 105 that simulates the mechanical characteristics of the machine tool 3 to explore the vibration condition in consideration of the difference between the command value and the actual value of the vibration amplitude. Specifically, the condition exploration unit 102b uses the mechanical model 105 to estimate the actual vibration amplitude at the tip position of the tool 304 corresponding to the specified vibration amplitude, and explores the vibration condition that makes the estimated vibration amplitude value satisfy the breaking condition.

[0091] Figure 6 Yes means Figure 5 A diagram of an example of a mechanical model 105 is shown. Figure 6 The mechanical model 105 simulating the mechanical characteristics of the drive shaft servo control unit 202 and the drive shaft 301 in the frequency domain is shown. The drive shaft servo control unit 202 is a controller that controls the position x of the tip of the tool 304 in response to the drive shaft command 17r, and has a cascade control structure with a speed control loop inside the position control loop. At this time, the position control has a difference in the tool tip position x relative to the drive shaft command 17r, multiplied by the position loop gain K p The position controller that calculates the speed command vr is the P (Proportion) controller. The speed controller multiplies the speed loop gain K by the difference between the speed command vr and the tool tip speed v. vp The result is multiplied by the speed integral gain K vi The results of integration are added to calculate the speed controller, i.e., the PI (Proportion Integral) controller, which calculates the torque Tr output by the servo motor 305. The mechanical model 105 integrates the torque command by multiplying it by J, which simulates the total moment of inertia of the drive shaft 301, and calculates the tool tip speed v. The tool tip speed v is integrated to calculate the tool tip position x. Here, s is the Laplace operator, and "1 / s" represents the integration operation.

[0092] At this time, the response of the tool tip position x to the drive axis command 17r is obtained by the following equation (7).

[0093] [Formula 7]

[0094]

[0095] However, the mechanical model 105 is an example, and any model may be used as long as it is a model that can calculate the response of the tool tip position to the position command. For example, a controller other than a P controller and a PI controller may be used, or the total moment of inertia may be replaced by a transfer function to express the characteristics of the drive shaft 301. For example, a current control loop that calculates the servo motor current 22 output by the servo motor 305 may be provided inside the speed control loop. At this time, the torque Tr is calculated by multiplying the torque constant calculated for the motor torque based on the motor current by the motor current.

[0096] Furthermore, instead of a continuous system model using a Laplace operator, a model described by a discrete system involving a difference equation may be used as the mechanical model 105 .

[0097] Figure 7 It is used for Figure 5 The process of step S1 to step S6 is similar to the process of step S1 to step S6. Figure 4 Same. Figure 7 In step S2, before determining whether the breaking condition is satisfied, the condition exploration unit 102b uses the mechanical model 105 to estimate the vibration amplitude (step S7). Specifically, the condition exploration unit 102b uses the mechanical model 105 to calculate the actual vibration amplitude between the tool 304 and the object 303 according to the set vibration amplitude. Here, a method for calculating the vibration amplitude using the model represented by the above formula (7) is described. In the model described in the form of a transfer function, the response at a certain frequency ω can be calculated by substituting jω for s in formula (7). Wherein j is an imaginary unit. If jω is substituted for s in formula (7), formula (8) is derived.

[0098] [Formula 8]

[0099]

[0100] At this time, since the ratio of the output to the input corresponds to the amount of change in the vibration amplitude, if the absolute value of the equation (8) is calculated, it is expressed by the following equation (9).

[0101] [Formula 9]

[0102]

[0103] As described above, if the frequency of the input vibration is determined, the actual vibration output between the tool 304 and the object 303 can be calculated as a formula, so that the set vibration amplitude A can be used to calculate the set vibration frequency ω. v The actual vibration amplitude A′ between the tool 304 and the object 303 is calculated using equation (9).

[0104] Here, the calculation is performed assuming that the input vibration has a single frequency ω, but the input vibration waveform may also be a rectangular wave, a triangular wave, etc. For example, a rectangular wave vibration instruction uses a Fourier series, as shown in formula (10), and can be described for each n as the sum of single vibrations of frequency (2π-1)ω with a vibration amplitude of (4 / π)×(A / 2n-1).

[0105] [Formula 10]

[0106]

[0107] By using equation (10) and equation (9) to calculate the sum of the amplitude ratios for each frequency, the amplitude of the vibration command for the rectangular wave can be calculated. Alternatively, since the first frequency component is the strongest, when calculating the response to the rectangular wave input, it is sufficient to calculate the amplitude ratio by approximating it as a function of only the first frequency component. The amplitude ratio can also be calculated in the same way for commands that simulate triangular waves or other periodic vibrations.

[0108] The condition search unit 102b Figure 7 In step S2, instead of the set vibration amplitude A, the vibration amplitude A' calculated in step S7 is used to determine whether the disconnection condition is satisfied.

[0109] Figure 8 It is used for Figure 5 The flowchart for explaining the second example of the operation of the condition search unit 102b shown in FIG. Figure 7 In the first example shown, the actual vibration amplitude A' is calculated before judging whether the disconnection condition is satisfied. Figure 8 In the second example shown, after the condition search unit 102b finds the driving condition that satisfies the breaking condition, it corrects the vibration amplitude based on the mechanical model 105 before outputting it as the execution driving condition (step S8). Specifically, after the vibration amplitude A and the vibration frequency ω that satisfy the breaking condition are found, v Afterwards, the condition exploration unit 102b multiplies the inverse of formula (9) by the vibration amplitude A, thereby calculating the corrected vibration amplitude A" so that the vibration amplitude between the tool 304 and the processing object 303 is consistent with A. In step S6, the condition exploration unit 102b outputs the execution drive condition including the corrected vibration amplitude A".

[0110] Furthermore, in the above example, a formula is used to calculate the actual vibration amplitude value according to the command value of the vibration amplitude, but the method for estimating the actual vibration amplitude value is not limited to the above example. For example, the condition exploration unit 102b can estimate the actual vibration amplitude value using a formula, a data table, etc. obtained by approximating the actual vibration amplitude value experimentally derived from the command value of the vibration amplitude in an experimental manner. In addition, the condition exploration unit 102b can also replace formula (7) and use a model of a discrete system described by a differential equation, and use a transformation formula that transforms a discrete system model called a bilinear transformation or Tustin transformation into a continuous system model to transform it into a continuous system and calculate the amplitude ratio. In addition, other existing algorithms for calculating the vibration amplitude can also be used to calculate the actual vibration amplitude value. Alternatively, the condition exploration unit 102b can also calculate the response waveform according to the model through numerical simulation, and calculate the amplitude ratio according to the response waveform obtained through simulation.

[0111] Fig. 9 This is a diagram showing the relationship between the response amplitude and the command value of the vibration amplitude derived from the experiment. Fig. 9 It is a diagram called a Bode diagram, which assigns a sinusoidal signal and random noise with a continuously changing frequency to a servo motor, and is obtained by measuring the vibration between the tool 304 and the processing object 303 at this time using an acceleration sensor, a displacement sensor, etc. In this figure, the ratio of the output amplitude to the input amplitude is calculated at each frequency. However, 0dB means that the input amplitude and the output amplitude are consistent. If the gain is greater than 0dB, it means that the output amplitude is larger than the input amplitude. If the gain is less than 0dB, it means that the output amplitude is smaller than the input amplitude. By using the Bode diagram, the actual vibration amplitude corresponding to the command value of the input vibration amplitude can be obtained without using numerical calculations. For example, the characteristics of the Bode diagram can be approximated by a formula, or the actual value of the vibration amplitude can be calculated using the mechanical model 105 as table data.

[0112] As described above, in the information processing device 1b according to the second embodiment, the condition search unit 102b searches for a driving condition that satisfies the breaking condition based on the vibration amplitude estimated using the mechanical model 105 that estimates the actual vibration amplitude at the tip position of the tool 304 based on the drive axis command for the axis. By determining whether the estimated value of the actual vibration amplitude satisfies the breaking condition instead of the command value of the vibration amplitude, it is possible to more reliably increase the possibility of chip breaking when the machine tool 3 performs vibration cutting.

[0113] Implementation method 3.

[0114] Fig.101 is a diagram showing a functional configuration of an information processing device 1c according to Embodiment 3. The information processing device 1c is an example of a drive condition determination device. The information processing device 1c includes a search condition input unit 100c and a condition search unit 102c.

[0115] The search condition input unit 100 c acquires a reference machining condition 11 , a reference vibration condition 12 , a machining condition search range 13 , and a vibration condition search range 14 as search conditions.

[0116] The condition search unit 102 c searches for driving conditions using the mechanical model 105 c and the evaluation function 106 .

[0117] Fig.11 Yes means Fig.10 FIG. 1 is a diagram showing an example of a mechanical model 105c shown in FIG. 1 . The difference between the mechanical model 105c and the mechanical model 105 of Embodiment 2 is that the mechanical model 105c has a mechanical end transfer function model Gm(s). In the usual feedback control, control is performed so that the position detected by the detector is consistent with the command position. In the case of a working machine 3 that performs vibration cutting, the object whose vibration amplitude should actually be controlled is the relative vibration between the tool 304 and the object 303 to be processed, but it is difficult to install a detector that detects the relative displacement between the tool 304 and the object 303 to be processed. Therefore, feedback control is performed by using a detector installed on the servo motor 305, and the mechanical end transfer function model Gm(s) is used to estimate the actual relative vibration between the tool 304 and the object 303 to be processed.

[0118] The mechanical end model is measured by installing acceleration sensors on the tool 304 and the object to be processed 303. For example, a Bode line diagram can be drawn by taking the acceleration signal as input and the relative acceleration as output by performing a second differentiation on the detector signal when driving the motor, and identifying the transfer function by curve fitting, or by using the transfer function calculated by the finite element method (FEM) based on the structural model of the machine tool 3, or by using a state space model output from the element synthesis matrix of the FEM model.

[0119] Fig.12 It is used for Fig.101. The condition exploration unit 102c first sets the reference processing condition 11 and the reference vibration condition 12 (step S1). Next, the condition exploration unit 102c estimates the actual vibration amplitude using the mechanical model 105c (step S7). The condition exploration unit 102c determines whether the set processing condition and vibration condition satisfy the breaking condition (step S2). At this time, the condition exploration unit 102c uses the actual vibration amplitude estimated by step S7 as the vibration amplitude of the vibration condition.

[0120] If the breaking condition is not satisfied (step S2: No), the condition search unit 102c determines whether the vibration amplitude can be adjusted (step S3). If the vibration amplitude can be adjusted (step S3: Yes), the condition search unit 102c changes the vibration amplitude within the vibration amplitude search range (step S4), and returns to the process of step S7.

[0121] If the vibration amplitude cannot be adjusted (step S3: No), the condition search unit 102c determines whether the vibration frequency can be adjusted (step S9). If the vibration frequency can be adjusted (step S9: Yes), the condition search unit 102c changes the vibration frequency within the vibration frequency search range (step S5) and returns to the process of step S7.

[0122] If the vibration frequency cannot be adjusted (step S9: No), the condition search unit 102c determines whether the spindle speed can be adjusted (step S10). If the spindle speed can be adjusted (step S10: Yes), the condition search unit 102c changes the spindle speed within the speed search range (step S11), and returns to the process of step S7.

[0123] If the spindle speed cannot be adjusted (step S10: No), the condition search unit 102c determines whether the feed amount can be adjusted (step S12). If the feed amount can be adjusted (step S12: Yes), the condition search unit 102c changes the feed amount within the feed amount search range (step S13) and returns to the process of step S7.

[0124] When the feed amount cannot be adjusted (step S12: No), a search failure is output (step S14) and the process ends.

[0125] When the breaking condition is satisfied (step S2: Yes), the condition search unit 102c then determines whether the set processing condition and vibration condition satisfy the first evaluation function (step S15). When the first evaluation function is not satisfied (step S15: No), the condition search unit 102c proceeds to the processing of step S3. When the first evaluation function is satisfied (step S15: Yes), the condition search unit 102c outputs the set processing condition and vibration condition as the execution driving condition (step S6).

[0126] Here, the first evaluation function is a function for evaluating the explored machining conditions and vibration conditions. Even if the explored machining conditions and vibration conditions meet the chip breaking conditions, it is not preferred to use the machining conditions and vibration conditions described above for machining if the machining takes too much time or the load on the working machine 3 is too large. Therefore, the information processing device 1c uses an evaluation function to determine whether the combination of the explored machining conditions and vibration conditions is appropriate. The first evaluation function is represented by a linear sum greater than or equal to any one of the electric energy required for vibration, vibration acceleration, machining time, cutting volume, maximum machining load, mechanical wear, and cumulative vibration cutting time. For example, the first evaluation function can use the cutting volume V (mm 3 The cutting volume V can be expressed by the following equation (11) using the spindle speed S, the feed rate F per cutting edge, the radial feed rate h, and the diameter R (mm) of the object 303 .

[0127] [Formula 11]

[0128] V=2πRhFS …(11)

[0129] The first evaluation function may simply be that the spindle speed is greater than or equal to the lower limit value, or may be that the feed rate is greater than or equal to the lower limit value, or may be a combination of these.

[0130] In addition, as described in the first embodiment for the vibration amplitude, in the step of changing each condition within the exploration range, each condition can be changed in units of a certain value from the lower limit value to the upper limit value of the exploration range. In this case, in the step of judging whether each condition can be adjusted, it can be judged whether the adjustment can be made based on whether the changed value exceeds the upper limit value. In this case, in step S5, the vibration amplitude is changed to the lower limit value of the vibration amplitude exploration range, in step S11, the vibration amplitude is changed to the lower limit value of the vibration amplitude exploration range, the vibration frequency is changed to the lower limit value of the vibration frequency exploration range, and in step S13, the vibration amplitude is changed to the lower limit value of the vibration amplitude exploration range, the vibration frequency is changed to the lower limit value of the vibration frequency exploration range, and the spindle speed is changed to the lower limit value of the speed exploration range. Thus, it is possible to change each value contained in the processing conditions and the vibration conditions from the lower limit value to the upper limit value of the exploration range while exploring the driving conditions that satisfy the breaking conditions and the first evaluation function.

[0131] In addition, in the third embodiment, since the evaluation is performed using the first evaluation function, the exploration may fail. If the driving condition that satisfies both the segmentation condition and the first evaluation function is not found, the user is notified of the exploration failure in step S14. In this case, the user can reset the exploration range and perform the exploration again.

[0132] In addition, Fig.12 In the process, if one driving condition that satisfies both the breaking condition and the first evaluation function is found, the exploration is terminated. However, in the entire exploration range, multiple candidates for the execution driving condition that satisfies the breaking condition can be explored, and the execution driving condition is determined from the multiple candidates based on the first evaluation function. In this case, when there are multiple candidates, the condition suitable for processing can be used among the multiple candidates.

[0133] As described above, according to the information processing device 1c involved in the third embodiment, the condition exploration unit 102c explores the candidates of the execution driving condition that satisfies the breaking condition, and can judge whether to set the explored candidate as the execution driving condition based on the first evaluation function for evaluating the cutting process using the explored candidate. Thus, even if the breaking condition is satisfied, it is possible to prevent the conditions with low evaluation of the cutting process, such as taking too much time or applying too much load to the machine tool 3, from becoming the execution driving condition.

[0134] In addition, the first evaluation function can be a function represented by the linear sum of any one parameter greater than or equal to the energy required to generate vibration, vibration acceleration, machining time, cutting volume, maximum machining load, mechanical wear, and cumulative vibration cutting time. Thus, the cutting process can be evaluated based on these parameters.

[0135] For example, regarding the energy required to generate vibration, if the mass of the shaft performing vibration cutting is M, then when expressed in terms of vibration energy, “1 / 2MA 2 ω v 2 " is obtained. Alternatively, the value after the above-mentioned vibration energy is converted into electrical energy can be set as the energy required to generate vibration. In this case, the value obtained by dividing the vibration energy by the coefficient representing the efficiency of converting electrical energy into vibration energy can be set as the energy required to generate vibration. Regarding the machining time, based on the feed amount F per rotation and the spindle speed S, the axis movement amount SF per minute is calculated, and the total machining distance L for vibration cutting is divided by the axis movement amount SF per minute, thereby calculating the machining time "L / SF". The maximum machining load uses the relative cutting resistance K determined according to the material physical properties of the machining object 303. c , through "K c The mechanical wear amount is calculated by accumulating the vibration cutting time up to that point and multiplying the accumulated vibration cutting time by a coefficient determined by experiments.

[0136] By further combining these evaluation values, a more flexible evaluation function can be designed. For example, when searching for the optimal conditions for both machining time and vibration energy, the evaluation function Z can be obtained as a linear sum of two evaluation functions with coefficients P1 and P2 in a manner represented by the following formula (12). However, the coefficients P1 and P2 can take any predetermined real numbers.

[0137] [Formula 12]

[0138]

[0139] In the third embodiment, if no driving condition satisfying the segmentation condition and the first evaluation function is found within the set search range, a search failure is notified, so the user can change the search condition and search for the driving condition again.

[0140] In the above description, the mechanical model 105 is used to estimate the actual vibration amplitude value from the command value of the vibration amplitude. However, an AI (Artificial Intelligence) model may be used instead of the mechanical model 105c that simulates the physical phenomenon. Fig.13 This is a diagram showing a functional configuration of an information processing device 1 c - 2 according to a modification of the third embodiment. Fig.13 The information processing device 1c-2 shown is different from the information processing device 1c in that it has a condition exploration unit 102c-2 that uses an AI model 107 instead of a mechanical model 105c to estimate the actual value based on the command value of the vibration amplitude. The AI ​​model 107 is a trained model that outputs the actual vibration amplitude at the tip position of the tool 304 if the command value of the vibration amplitude, the spindle speed, the feed speed per blade, and the vibration frequency are input. The AI ​​model 107 can be, for example, a model trained by artificial intelligence using algorithms such as neural networks, reinforcement learning, and Q learning.

[0141] Implementation method 4.

[0142] Fig.14 This is a diagram showing a functional configuration of a numerical control device 1 d according to Embodiment 4. The numerical control device 1 d includes a search condition input unit 100 , a machining program input unit 101 , a condition search unit 102 d , and a command value generation unit 104 .

[0143] The numerical control device 1d is different from the numerical control device 1a in that the inputs to the search condition input unit 100 are the reference processing condition 11, the processing condition search range 13, and the vibration condition search range 14, and the condition search unit 102d is provided which uses the machine model 105 and the evaluation function 106 as inputs. In addition, in the numerical control device 1d, the reference vibration condition is not included in the input to the search condition input unit 100. Therefore, at the beginning of the search, the search is started based on the lower limit value of the vibration condition search range, the upper limit value of the vibration condition search range, the middle value of the vibration condition search range, or the search initial condition which is set in advance by parameters or the like.

[0144] Fig.15 It is used for Fig.14 The flowchart for explaining the operation of the condition search unit 102d is shown in FIG. Fig.12 The difference between the condition search unit 102c and the condition search unit 102d in the third embodiment shown in FIG.

[0145] If the second evaluation function is satisfied (step S17: Yes), the condition search unit 102d does not adopt the set condition, but returns to step S3 to continue searching for the condition. If the second evaluation function is not satisfied (step S17: No), the condition search unit 102d proceeds to step S6.

[0146] In addition, Fig.15 If the feed rate cannot be adjusted (step S12: No), the condition search unit 102 sets a driving condition including the reference machining condition 11 without performing vibration cutting (step S16), and proceeds to step S6.

[0147] Fig.15 Step S17 of the control functions as a safety device, and when "if vibration cutting is performed using the set processing conditions and vibration conditions, an abnormality occurs", the second evaluation function is set in a manner to "satisfy the second evaluation function". For example, when the vibration amplitude is too large, the load on the machine tool 3 and the load on the motor product equipped in the machine tool 3 increase, which may become a cause of failure, the second evaluation function is satisfied. An example of a motor product equipped in the machine tool 3 is a numerical control device, a drive device called a driver unit, a motor, an electromagnetic switch, a relay, etc.

[0148] As an example of the second evaluation function, a function using vibration energy H, power consumption, and vibration amplitude is cited. In addition, the same function can be used in the first evaluation function and the second evaluation function, and different thresholds can be set. For example, the first evaluation function sets the processing time to be greater than or equal to 3 minutes, and the second evaluation function sets the processing time to be greater than or equal to 10 minutes. In this case, the first evaluation function is not satisfied when the processing time is less than 3 minutes, and the second evaluation function is satisfied when the processing time is greater than or equal to 10 minutes, so the set driving condition is not selected. Therefore, only when the processing condition is greater than or equal to 3 minutes and less than 10 minutes, it is output as an execution driving condition.

[0149] In addition, Fig.14 Although not shown in the figure, an evaluation function setting unit in which a user sets arbitrary conditions as the first evaluation function and the second evaluation function may be provided.

[0150] In addition, the second evaluation function can be set as the same as the first evaluation function, and can be set as a function represented by the linear sum of any one of the electric energy required to generate vibration, vibration acceleration, machining time, cutting volume, maximum machining load, mechanical wear and cumulative vibration cutting time.

[0151] When the first evaluation function is satisfied, it is preferably a function that can judge whether the vibration cutting using its driving condition is appropriate. When the second evaluation function is satisfied, it is preferably a function that can judge whether the driving condition is not adopted as the execution driving condition if an abnormality occurs in the vibration cutting using its driving condition.

[0152] In addition, as an evaluation function, when using the maximum processing load, mechanical wear, cumulative vibration cutting time, etc., the processing load and wear are predicted by using the processing simulation of the mechanical model 105. In addition, with respect to the cumulative vibration cutting time, a memory is set to record the total time of the vibration cutting implemented so far. In addition, for example, an evaluation function using the number of chips that can be broken by vibration cutting per rotation is also considered. For example, if the number of chip breakages during one rotation is too many, a load is applied to the servo motor 305 and heat is generated. On the other hand, when processing a processing object 303 with a large radius, if the number of chip breakages is small, the chips become longer and the possibility of entanglement with the tool 304 etc. becomes higher. Therefore, by using the number of chip breakages per rotation as an evaluation function, the driving conditions can be explored so that the number of breakages becomes appropriate.

[0153] As described above, the condition search unit 102d of the numerical control device 1d according to the fourth embodiment evaluates the candidate of the execution driving condition that satisfies the breaking condition based on the second evaluation function for evaluating whether an abnormality occurs in the cutting process using the candidate, and does not use the candidate as the execution driving condition when it is evaluated that an abnormality occurs in the cutting process using the candidate. Thus, even if the breaking condition is satisfied, it is possible to prevent the selection of a driving condition that causes an abnormality in the cutting process.

[0154] In addition, when the condition search unit 102d does not find the driving condition to be set as the execution driving condition based on the second evaluation function, it can determine the driving condition that does not perform vibration cutting, including the reference processing condition 11, as the execution driving condition. In this case, when the driving condition that is evaluated as not causing an abnormality using the second evaluation function is not found, vibration cutting is not performed, and the cutting processing that does not cause an abnormality can be preferentially performed.

[0155] Implementation method 5.

[0156] Fig.161 is a diagram showing the functional structure of the information processing device 1e according to the fifth embodiment. The information processing device 1e is different from the third embodiment in that the information input to the search condition input unit 100 is the reference processing condition 11, the vibration condition search range 14, and the control parameter search range 19, and a condition search unit 102e is provided instead of the condition search unit 102c. The input to the information processing device 1e is the reference processing condition 11, the vibration condition search range 14, and the control parameter change range 19, and the output from the information processing device 1e is the execution processing condition 15, the execution vibration condition 16, and the control parameter 24. In addition, in the numerical control device 1e, the reference vibration condition is not included in the input to the search condition input unit 100. Therefore, at the beginning of the search, the search is started from the lower limit value of the vibration condition search range, the upper limit value of the vibration condition search range, the middle value of the vibration condition search range, or the search initial condition set in advance by parameters or the like.

[0157] Fig.17 It is used for Fig.16 The flowchart shown is used to explain the operation of the condition search unit 102e.

[0158] The condition search unit 102e sets the reference processing condition 11 (step S1e). The condition search unit 102e determines whether the set condition satisfies the breaking condition (step S2). If the breaking condition is not satisfied (step S2: No), the condition search unit 102e determines whether the vibration amplitude can be adjusted (step S3). If the vibration amplitude can be adjusted (step S3: Yes), the condition search unit 102e changes the vibration amplitude within the vibration amplitude search range (step S4), and returns to the process of step S2.

[0159] If the vibration amplitude cannot be adjusted (step S3: No), the condition search unit 102e determines whether the vibration frequency can be adjusted (step S9). If the vibration frequency can be adjusted (step S9: Yes), the condition search unit 102e changes the vibration frequency within the vibration frequency search range (step S5) and returns to the process of step S2.

[0160] If the vibration frequency cannot be adjusted (step S9: No), the condition search unit 102e determines whether the control parameter can be adjusted (step S18). If the control parameter can be adjusted (step S18: Yes), the condition search unit 102e changes the control parameter within the control parameter adjustment range (step S19) and returns to the process of step S2.

[0161] When the control parameters cannot be adjusted (step S18: No), the condition search unit 102e sets the driving conditions including the reference machining conditions 11 without vibration cutting (step S16), and outputs the set machining conditions, vibration conditions and control parameters as the execution driving conditions (step S6e).

[0162] When the disconnection condition is satisfied (step S2: Yes), the condition exploration unit 102e corrects the vibration amplitude based on the mechanical model 105 (step S8). The condition exploration unit 102e determines whether the set driving condition satisfies the second evaluation function (step S17). When the second evaluation function is satisfied (step S17: Yes), the condition exploration unit 102e proceeds to the processing of step S3. When the second evaluation function is not satisfied (step S17: No), the condition exploration unit 102e proceeds to the processing of step S6e.

[0163] like Fig.17 As shown, in the fifth embodiment, the control parameters are explored based on the vibration conditions. When the mechanical model 105 is the same as the mechanical model 105 used in the third embodiment, the control parameters are, for example, the position loop gain K p , speed loop gain K vp , and speed integral gain K vi The control parameter is a parameter that determines the control characteristics of the drive shaft 301. As shown in equation (10), the control parameter affects the calculation of the vibration amplitude. Therefore, even under the same vibration conditions, whether the chip is broken will change depending on the size of the control parameter. The only control parameter that can be changed is the position loop gain K. p , or the position loop gain K p , speed loop gain K vp The parameters to be changed may be predetermined or may be selected arbitrarily by the user each time the processing is performed. In addition, when a model other than the mechanical model 105 shown in the example is used, the control parameters affecting the vibration amplitude may be set as the object of exploration in the model used. By setting the control parameters as the object of exploration, the possibility of exploring the driving conditions that can break the chips can be increased.

[0164] When the control parameters are changed, it is preferred to use an evaluation function to determine the thresholds of phase abundance and gain abundance so that the control system does not become unstable due to the change of the control parameters. Phase abundance is the abundance of the phase from -180 degrees when the gain becomes 0 dB in the control system. Generally speaking, if the phase abundance is greater than or equal to 30 degrees, the control system is stable. In addition, gain abundance is the gain abundance from 0 dB when the phase becomes -180 degrees. By setting the evaluation function as described above, it is possible to explore the driving conditions that cause chip breakage and stabilize the control system.

[0165] As described above, according to the fifth embodiment, while maintaining the stability of the control system, the vibration conditions and control parameters that cause the breakage of chips can be determined as the execution driving conditions.

[0166] Implementation method 6.

[0167] Fig.18 1 is a diagram showing the functional structure of the numerical control device 1f according to the sixth embodiment. Compared with the numerical control device 1d according to the fourth embodiment, the numerical control device 1f has: a mechanical model updating unit 113 that updates the mechanical model 105c based on the actual value of the vibration amplitude output by the acceleration sensor 306 that measures the vibration amplitude; and a fault prediction unit 114 that predicts the fault of the working machine 3 based on the time-varying change of the mechanical model 105c. However, the mechanical model 105c is different from the mechanical model 105c. Fig.11 The same as the mechanical model 105c shown. In addition, in the numerical control device 1f, the reference vibration condition is not included in the input to the exploration condition input unit 100. Therefore, at the beginning of the exploration, the exploration is started based on the lower limit value of the vibration condition exploration range, the upper limit value of the vibration condition exploration range, the middle value of the vibration condition exploration range, or the exploration initial condition set in advance by parameters.

[0168] The vibration characteristics of the machine tool 3 gradually change due to changes over time. For example, due to loosening of the fastening of the fastening parts, fatigue damage of the structural parts, leakage of the preload of the bearings or ball screws, etc., the vibration of the mechanical structure sometimes increases. In addition, sometimes the load of the motor increases due to increased friction. Sometimes, if the vibration cutting continues while the state as described above is maintained, the machine tool 3 will fail. Therefore, the numerical control device 1f has the following function, that is, monitoring the changes over time of the machine tool 3 based on the difference between the vibration amplitude predicted by calculation and the actual vibration amplitude, predicting the failure before the failure actually occurs, thereby enabling the machine tool 3 to be repaired at an appropriate time.

[0169] The acceleration sensor 306 is disposed on the drive shaft 301. Fig.18In the embodiment, an acceleration sensor 306 is installed in the nut portion of the ball screw, but it can also be installed inside the tool holder of the tool 304, or it can be installed in the turret of the lathe where the tool is installed. The acceleration sensor 306 measures the vibration level in the vibration cutting and sends it to the mechanical model updating unit 113. The mechanical model updating unit 113 compares the calculated value of the vibration amplitude calculated according to the execution of the processing condition 15 and the execution of the vibration condition 16 with the actual measured value of the vibration amplitude. At this time, when a difference occurs between the calculated value and the measured value due to a change in the mechanical characteristics, the Gm (s) that simulates the characteristics of the drive shaft 301 in the mechanical model 105c is changed. As a result, the vibration amplitude can be accurately estimated even when the mechanical characteristics change.

[0170] The fault prediction unit 114 monitors the change history of the mechanical model 105c. When the change amount of the mechanical model 105c exceeds a predetermined threshold, it is determined that the change over time of the working machine 3 exceeds the allowable value and a fault has occurred, and a warning is output to the user. There is no particular limitation on the output method of the warning, and the warning can be output by screen display or sound output. However, the fault prediction unit 114 can make a determination based on whether the current change amount of the mechanical model 105c exceeds the threshold, or it can predict the situation that the threshold will be exceeded in a certain period of time and output a warning.

[0171] As described above, the numerical control device 1f according to the sixth embodiment changes the control parameters in accordance with the changes in the mechanical model 105c caused by the time-dependent degradation of the machine tool 3, thereby having the effect of accurately estimating the vibration amplitude even when the mechanical characteristics change. In addition, based on the amount of change in the mechanical model 105c, the occurrence of a fault or a sign of the occurrence of a fault is detected, thereby making it possible to repair or establish a repair plan in advance, and avoiding the situation where production is stopped due to an emergency fault.

[0172] Implementation method 7.

[0173] Fig.19 1 is a diagram showing a functional configuration of an information processing device 1g according to Embodiment 7. The information processing device 1g includes a search condition input unit 100, a condition search unit 102g, and a trained model storage unit 120. The condition search unit 102g includes a learning data acquisition unit 40 and a model generation unit 41.

[0174] The learning data acquisition unit 40 receives the reference machining conditions 11 and the reference vibration conditions 12 from the search condition input unit 100, and acquires the control parameters 24, the tool information 30, the processing object information 31, and the motion information 32 of each axis as vibration cutting learning data. The tool information 30 includes information indicating the tool material and the tool shape. The processing object information 31 includes information indicating the material of the processing object. The motion information 32 of each axis includes information indicating the feed speed of each axis and the acceleration of each axis.

[0175] The model generation unit 41 uses the mechanical model 105, the machined surface quality determination model 42, and the chip breaking determination model 43 to learn the execution driving condition that achieves the best machining quality of the cutting process among the execution driving conditions that satisfy the breaking conditions, based on the vibration cutting learning data obtained by the learning data acquisition unit 40. That is, the model generation unit 41 outputs the condition and control parameter exploration model 48 as the learning result, and the condition and control parameter exploration model 48 takes the reference machining condition 11, the reference vibration condition 12, the control parameter 24, the motion information 32 of each axis, the machining object information 31, and the tool information 30 as input, and outputs the combination of the execution machining condition 15, the execution vibration condition 16, and the control parameter 24 that satisfies the breaking condition and the machining quality of the cutting process satisfies the reference.

[0176] The motion information 32 of each axis is information indicating the motion position of the motor of each axis of the machine tool 3. The model generation unit 41 uses the mechanical model 105 to obtain information indicating the motion of the mechanical structure of each axis installed in the machine tool 3 from the motion information 32 of each axis and the control parameter 24 as described above. That is, the mechanical model 105 takes the motion information 32 of each axis and the control parameter 24 as input, and outputs information indicating the motion of the mechanical structure of each axis installed in the machine tool 3. In addition, the model generation unit 41 uses the chip breaking determination model 43 to determine whether the chip breaking occurs, that is, to determine whether the breaking condition is satisfied. The chip breaking determination model 43 takes the output of the mechanical model 105, the reference machining condition 11, the reference vibration condition 12, and the control parameter 24 as input, and outputs a determination result of whether the chip breaking occurs. In addition, the model generation unit 41 uses the machined surface quality determination model 42 to determine whether the machining quality of the cutting machining using the driving condition determined as the occurrence of chip breaking satisfies the reference. The machined surface quality determination model 42 determines the quality of the result after the machined object 303 is machined based on the motion information 32 of each axis. The machined surface quality determination model 42 uses the output of the mechanical model 105, the driving conditions determined to meet the breaking conditions using the chip breaking determination model 43, the workpiece information 31 and the tool information 30 as inputs, and determines the machining quality based on the smoothness of the machined surface. For example, the machined surface quality determination model 42 can evaluate the machining quality for the tool path using the average line calculated based on the arithmetic average, the fluctuation of each tool path relative to the average surface, and the maximum error amount. Alternatively, the machined surface quality determination model 42 can also consider the shape of the tool 304 to estimate the machined surface, and use the height of the cusp after the machined object 303 is removed to evaluate the machining quality.

[0177] The above-mentioned machined surface pass / fail judgment part may be modeled and included in the learning object as a machined surface quality judgment model 42. The machined surface quality judgment model 42 may calculate the machined surface of the object 303 based on the relative displacement between the tool 304 and the object 303.

[0178] The relative displacement may be a command to the motor, a motor-side response calculated using a motor characteristic model, or a machine-side response calculated using a machine characteristic model.

[0179] In addition, when calculating the processing surface of the processing object 303, the calculation can be performed only through the relative displacement between the tool 304 and the processing object 303, or the relative displacement can be corrected by geometrically considering the shape of the tool 304, or the material of the tool 304 and the processing object 303 can be further considered to calculate the quality of the processing surface based on the physical cutting phenomenon theory.

[0180] The machined surface quality judgment model 42 can be replaced with a data set obtained by the actual machine of the machine tool 3. For example, based on the machining conditions and vibration conditions, a data set of tool shape, tool material, and machining object material can be associated with information on surface quality when machining is performed under the conditions of the data set, thereby setting it as a learning data set.

[0181] As described above, the model generation unit 41 learns the combination of machining conditions, vibration conditions, and control parameters based on the aforementioned vibration cutting learning data using the mechanical model, the machined surface quality determination model 42, and the chip breaking determination model 43. That is, a trained model is generated, and the trained model infers the driving conditions including at least one of the machining conditions, vibration conditions, and control parameters based on the aforementioned vibration cutting learning data.

[0182] Next, the learning process of the model generation unit 41 is described. The learning algorithm used by the model generation unit 41 can use well-known algorithms such as teacher learning, teacher-free learning, and reinforcement learning. As an example, the case where reinforcement learning is applied is described. In reinforcement learning, the acting subject, i.e., the agent, in a certain environment observes the parameters of the environment representing the current state and decides the action to be taken. According to the action of the agent, the environment changes dynamically, and the agent is rewarded according to the change in the environment. The agent repeats this action and learns the action policy that obtains the most rewards after a series of actions. As representative methods of reinforcement learning, Q-learning and TD-Learning are known. For example, in the case of Q-learning, the general update formula of the action value function Q(s, a) is expressed by the following formula (13).

[0183] [Formula 13]

[0184]

[0185] In formula (13), s t represents the state of the environment at time t, a t represents the action at time t. According to action a t , the state changes to s t+1 . rt+1 represents the reward brought by the change of its state, γ represents the discount rate, and α represents the learning coefficient. In addition, γ is set in the range of 0<γ≤1, and α is set in the range of 0<α≤1. The selection of processing conditions, vibration conditions and control parameters becomes action a t , the determination of chip breaking and the quality information of the machined surface become status s t , for the state s at time t t The best action under a t To study.

[0186] The update formula represented by formula (13) is that if the action value Q of action a with the highest Q value at time t+1 is greater than the action value Q of action a executed at time t, then the action value Q is increased, and in the opposite case, the action value Q is decreased. In other words, the action value function Q(s, a) is updated so that the action value Q of action a at time t is close to the best action value at time t+1. Thus, the best action value in a certain environment is propagated successively to the action value in its previous environment.

[0187] As described above, when generating a trained model by reinforcement learning, the model generation unit 41 includes a reward calculation unit 44 and a function update unit 45 .

[0188] The reward calculation unit 44 calculates the reward r based on the selection of machining conditions, vibration conditions and control parameters, the determination of whether the chips are broken and the machined surface quality information. The reward judgment criteria are as follows. For example, when the machined surface quality information improves, that is, when the fluctuation of the aforementioned tool paths, the maximum error amount is reduced, or when the cusp height is reduced, the reward r is increased (for example, a reward of "1" is assigned). On the other hand, when the machined surface quality information deteriorates, the reward r is reduced (for example, a reward of "-1" is assigned). In addition, as another example, when it is determined that there is chip breakage, the reward r is increased, and when it is determined that there is no chip breakage, the reward r is reduced. The chip breakage is necessary to exert the effect of vibration cutting, so when it is determined that there is no chip breakage, the reward r is greatly reduced (for example, a reward of "-5" is assigned).

[0189] The function updating unit 45 updates the function for determining the combination of the processing condition, the vibration condition, and the control parameter according to the reward calculated by the reward calculating unit 44, and outputs the updated function to the trained model storage unit 120. For example, in the case of Q learning, the action value function Q(s) represented by equation (13) is t , a t ) is used as a function for calculating a combination of machining conditions, vibration conditions and control parameters.

[0190] The above learning is repeatedly performed. The trained model storage unit stores the action-value function Q(s) updated by the function updating unit 45. t , a t ), that is, the trained model is stored.

[0191] Next, use Fig. 20 , explaining the processing learned by the learning device. Fig. 20 is with Fig.19 Flowchart related to the learning process of the condition searching unit 102g shown.

[0192] The learning data acquisition unit 40 acquires machining conditions, vibration conditions, and control parameters as learning data (step S31 ).

[0193] The model generation unit 41 calculates the reward based on the selection of machining conditions, vibration conditions and control parameters, the determination of whether the chips are broken and the quality information of the machined surface (step S32). Specifically, the reward calculation unit 44 uses the aforementioned models based on the selection of the combination of machining conditions, vibration conditions and control parameters, obtains the determination of whether the chips are broken and the quality information of the machined surface, and determines whether to increase or decrease the reward based on a predetermined reward judgment criterion.

[0194] If the reward calculation unit 44 determines that the reward should be increased (step S32: reward increase criterion), the reward calculation unit 44 increases the reward (step S33). On the other hand, if the reward calculation unit 44 determines that the reward should be decreased (step S32: reward decrease criterion), the reward calculation unit 44 decreases the reward (step S34).

[0195] The function updating unit 45 updates the action value function Q(s) represented by the equation (13) stored in the trained model storage unit 120 based on the reward calculated by the reward calculation unit 44. t , a t ) is updated (step S35).

[0196] The condition search unit 102g repeatedly executes the above steps from step S31 to step S35, and generates the action-value function Q(s t , a t ) is stored as a trained model.

[0197] As described above, the information processing device 1g involved in Embodiment 7 stores the trained model in the trained model storage unit 120 provided inside the information processing device 1g, but the trained model storage unit 120 may also be provided outside the information processing device 1g.

[0198] In the information processing device 1g involved in the seventh embodiment, artificial intelligence learning is performed using information during machining for machining phenomena associated with complex phenomena, thereby having the following effect: even an operator who does not have the technical know-how of machining can easily find the optimal chip breaking conditions.

[0199] In addition, as a learning algorithm used in the model generation unit 41, deep learning that learns the extraction of feature quantities themselves can also be used, and machine learning can also be performed according to other well-known methods, such as neural networks, genetic programming, functional logic programming, support vector machines, etc.

[0200] In addition, the model generation unit 41 may also use learning data obtained from multiple numerical control devices to learn the combination of processing conditions, vibration conditions and control parameters. In addition, the model generation unit 41 may also obtain learning data from multiple numerical control devices used in the same area, and may also use learning data collected from multiple numerical control devices operating independently in different areas to learn the combination of processing conditions, vibration conditions and control parameters. In addition, the numerical control device that collects learning data can also be added to the object midway, or removed from the object. Furthermore, the information processing device that has learned the combination of processing conditions, vibration conditions and control parameters for a certain numerical control device can also be applied to other numerical control devices, and the combination of processing conditions, vibration conditions and control parameters for the other numerical control devices can be re-learned and updated.

[0201] Implementation method 8.

[0202] Fig.21 1 is a diagram showing a functional configuration of a numerical control device 1h according to Embodiment 8. In Embodiment 8, inference is performed using the trained model generated in Embodiment 7. The condition search unit 102h includes an inference data acquisition unit 46 and an inference unit 47.

[0203] The estimation data acquisition unit 46 acquires the vibration cutting learning data as the estimation data.

[0204] The inference unit 47 uses the trained model to infer the combination of machining conditions, vibration conditions, and control parameters based on the inference data. That is, the inference unit 47 inputs the vibration cutting learning data acquired by the inference data acquisition unit 46 into the trained model, i.e., the condition and control parameter exploration model 48, thereby being able to infer the combination of machining conditions, vibration conditions, and control parameters that can break the chips and obtain a good machined surface quality.

[0205] Regarding the method of obtaining the trained model, the trained model can be obtained from an external device different from the CNC device 1h, such as an external personal computer, a storage area on the network, etc., and the combination of processing conditions, vibration conditions and control parameters can be output based on the obtained trained model.

[0206] In this embodiment, the combination of machining conditions, vibration conditions, and control parameters is output using a trained model learned by the model generation unit 41 of the information processing device. However, a trained model generated by learning by the model generation unit 41 built into the numerical control device 1h may be used.

[0207] Next, a description will be given of a process for obtaining a combination of machining conditions, vibration conditions, and control parameters.

[0208] First, the estimation data acquisition unit 46 acquires vibration cutting learning data. At this time, the vibration cutting learning data may include at least the reference vibration condition 12 and the reference machining condition 11 as described above, but by adding information other than these, it is expected that the estimation accuracy can be improved.

[0209] Next, the inference unit 47 inputs the vibration cutting learning data as inference data to the acquired trained model, i.e., the condition and control parameter exploration model 48, and obtains a combination of machining conditions, vibration conditions, and control parameters. The inference unit 47 outputs the obtained combination of machining conditions, vibration conditions, and control parameters to the command value generation unit 104.

[0210] Then, the command value generating unit 104 generates a drive axis command and a spindle command using the combination of the output machining conditions, vibration conditions, and control parameters. Fig.21 The vibration cutting can be performed under the optimal conditions estimated by the trained model and taking both the occurrence of chip breakage and good machined surface quality into consideration.

[0211] Fig. 22 It is used for Fig.21 The flowchart for explaining the estimation processing of the numerical control device 1h shown in FIG.

[0212] The estimation data acquisition unit 46 acquires the vibration cutting learning data as estimation data (step S41 ).

[0213] The inference unit 47 inputs the vibration cutting learning data as inference data to the trained model (step S42), and obtains the combination of the execution machining condition 15, the execution vibration condition 16, and the control parameter 24. The inference unit 47 outputs the obtained combination of the execution machining condition 15, the execution vibration condition 16, and the control parameter 24 to the command value generation unit 104 (step S43).

[0214] The command value generation unit 104 generates a command value using the estimation result (step S44 ).

[0215] In the present embodiment, reinforcement learning is applied to the learning algorithm used by the estimation unit 47, but the present invention is not limited thereto. As for the learning algorithm, in addition to reinforcement learning, supervised learning, unsupervised learning, semi-supervised learning, etc. can also be applied.

[0216] As described above, in the numerical control device 1h in the eighth embodiment, vibration cutting is performed using the model learned by artificial intelligence, thereby achieving the effect that the operator can perform optimal vibration cutting without judging the state of machining.

[0217] Next, the hardware structure of the numerical control devices 1a, 1d, 1f, and 1h involved in embodiments 1, 4, 6, and 8 is described. The functional units of the numerical control devices 1a, 1d, 1f, and 1h are implemented using processing circuits. The processing circuits can be implemented by dedicated hardware or by a control circuit using a CPU (Central Processing Unit).

[0218] In the case where the above processing circuits are implemented by dedicated hardware, they are used Fig.23 This is achieved by the processing circuit 90 shown. Fig.23 It is a diagram showing dedicated hardware for realizing the functions of the numerical control devices 1a, 1d, 1f, and 1h involved in embodiments 1, 4, 6, and 8. The processing circuit 90 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof.

[0219] When the processing circuit is implemented by a control circuit using a CPU, the control circuit is, for example, Fig.24 The control circuit 91 of the structure shown. Fig.24 1 is a diagram showing the structure of a control circuit 91 for realizing the functions of the numerical control devices 1a, 1d, 1f, and 1h according to the first, fourth, sixth, and eighth embodiments. Fig.24As shown, the control circuit 91 has a processor 92 and a memory 93. The processor 92 is a CPU, which is also called a processing device, a computing device, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), etc. The memory 93 is, for example, a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable ROM), an EEPROM (registered trademark) (Electrically EPROM), a magnetic disk, a floppy disk, an optical disk, a compact disk, a mini disk, a DVD (Digital Versatile Disk), etc.

[0220] When the processing circuit is realized by using the control circuit 91, the processor 92 reads and executes the program corresponding to the processing of each component stored in the memory 93. The memory 93 is also used as a temporary memory in each processing executed by the processor 92.

[0221] Next, the hardware structure of the information processing devices 1b, 1c, 1c-2, 1e, and 1g of the present embodiment is described. In the case where the information processing devices 1b, 1c, 1c-2, 1e, and 1g are devices separate from the numerical control device, the information processing devices 1b, 1c, 1c-2, 1e, and 1g of the present embodiment are implemented, for example, by a computer system. The information processing devices 1b, 1c, 1c-2, 1e, and 1g can be implemented by one computer system or by multiple computer systems. For example, the information processing devices 1b, 1c, 1c-2, 1e, and 1g can be implemented by a cloud system. In the cloud system, the hardware of the computer system and the division of devices such as servers for each function can be arbitrarily set. For example, one computer system can have the functions of multiple devices, and multiple computer systems can also have the functions of one device.

[0222] A configuration example of a computer system that realizes the information processing devices 1 b , 1 c , 1 c - 2 , 1 e , and 1 g will be described. Fig.25 1 is a diagram showing a configuration example of a computer system for realizing information processing devices 1b, 1c, 1c-2, 1e, and 1g according to Embodiments 2, 3, 5, and 7. Fig.25 As shown, the computer system includes a control unit 401 , an input unit 402 , a storage unit 403 , a display unit 404 , a communication unit 405 , and an output unit 406 , which are connected via a system bus 407 .

[0223] exist Fig.25In the embodiment, the control unit 401 is, for example, a CPU. The control unit 401 executes a program that describes the various processes to be performed by the information processing devices 1b, 1c, 1c-2, 1e, and 1g of this embodiment. The input unit 402 is composed of, for example, a keyboard, a mouse, etc., and is used by the user of the computer system to input various information. The storage unit 403 includes various memories such as RAM (Random Access Memory) and ROM (Read Only Memory) and storage devices such as a hard disk, and stores the program to be executed by the above-mentioned control unit 401, the required data obtained during the processing, etc. In addition, the storage unit 403 is also used as a temporary storage area for the program. The display unit 404 is composed of an LCD (Liquid Crystal Display: liquid crystal display panel) and the like, and displays various screens for the user of the computer system. The communication unit 405 is a communication circuit that performs communication processing. The communication unit 405 can be composed of a plurality of communication circuits corresponding to a plurality of communication methods. The output unit 406 is an output interface that outputs data to an external device such as a printer and an external storage device.

[0224] also, Fig.25 is an example, the structure of the computer system is not limited to Fig.25 For example, the computer system may not have the output unit 406. In addition, when the information processing devices 1b, 1c, 1c-2, 1e, and 1g are implemented by a plurality of computer systems, all of these computer systems may not be Fig.25 For example, a portion of the computer system may not have Fig.25 At least one of the display unit 404, output unit 406, and input unit 402 shown.

[0225] Here, an example of the operation of a computer system until the drive condition determination program describing the processing of the information processing devices 1b, 1c, 1c-2, 1e, and 1g of the present embodiment can be executed is described. In the computer system configured as described above, the drive condition determination program is installed in the storage unit 403 from, for example, a CD-ROM or DVD-ROM provided in a CD (Compact Disc)-ROM drive or DVD-ROM drive (not shown). Furthermore, when the drive condition determination program is executed, the drive condition determination program read from the storage unit 403 is stored in an area of ​​the main storage device of the storage unit 403. In this state, the control unit 401 executes the processing of the information processing devices 1b, 1c, 1c-2, 1e, and 1g of the present embodiment according to the drive condition determination program stored in the storage unit 403.

[0226] In addition, in the above description, a CD-ROM or DVD-ROM is used as a recording medium to provide a program that records the processing in the information processing devices 1b, 1c, 1c-2, 1e, and 1g, but this is not limited to this. It is also possible to use a program provided through a transmission medium such as the Internet via the communication unit 405, for example, in accordance with the structure of the computer system, the capacity of the program to be provided, etc.

[0227] The driving condition determination program of the present embodiment is a driving condition determination program for determining the driving condition of a working machine for performing vibration cutting, wherein the vibration cutting is a process in which an axis that changes the relative positions of a processing object and a tool vibrates during cutting to separate chips. The driving condition determination program enables a computer to execute the following steps: obtaining an exploration condition, the exploration condition including an exploration range set for at least one driving condition among the processing conditions of the working machine, the vibration conditions of the axis, and the parameters that determine the control characteristics of the axis, i.e., the control parameters, and a processing condition that serves as a reference and is determined regardless of the presence or absence of vibration cutting, i.e., the reference processing condition 11; and determining, based on the exploration condition, the driving condition to be used when performing vibration cutting, i.e., the execution driving condition, within the range of the exploration range set in such a manner as to satisfy the condition for the occurrence of chip separation, i.e., the separation condition.

[0228] The search condition input unit 100 of the information processing devices 1b, 1c, 1c-2, 1e, and 1g is realized by the input unit 402. The condition search units 102b, 102c, 102c-2, and 102e of the information processing devices 1b, 1c, 1c-2, 1e, and 1g are realized by the control unit 401. The mechanical models 105 and 105c and the evaluation function 106 are stored in the storage unit 403.

[0229] In the system including the information processing devices 1b, 1c, 1c-2, 1e, and 1g, the division of functions in the devices shown in the figure is an example, and the division of functions in the devices is not limited to Figure 1 For example, the information processing devices 1 b , 1 c , 1 c - 2 , 1 e , and 1 g may be realized by the same computer system as the machining program generating device 103 .

[0230] The configuration described in the above embodiment is merely an example, and may be combined with other known technologies, the embodiments may be combined with each other, and part of the configuration may be omitted or changed without departing from the gist.

[0231] For example, in each of the above embodiments, an example of a method for exploring driving conditions is shown, but any algorithm can be used if necessary conditions such as the spindle speed, the feed rate per blade, the vibration amplitude, and the vibration frequency can be determined. For example, the order of the exploration can be changed by changing the order of the steps shown in the figure. In addition, the method is not limited to exploring driving conditions by a cyclic method, and various known optimization algorithms, such as genetic algorithms and Bayesian optimization algorithms using artificial intelligence, can be used to explore driving conditions.

[0232] In the above embodiments, the information processing device is connected to the numerical control device via a network, for example, and is described as a device separate from the numerical control device, but the information processing device may be built into the numerical control device. Alternatively, the information processing device may exist on a cloud server.

[0233] Description of the label

[0234] 1a, 1d, 1f, 1h numerical control device, 1b, 1c, 1c-2, 1e, 1g information processing device, 2 drive unit, 3 machine tool, 10 machining program, 11 reference machining condition, 12 reference vibration condition, 13 machining condition search range, 14 vibration condition search range, 15 execution machining condition, 16 execution vibration condition, 17, 17r drive axis instruction, 18 spindle instruction, 19 control parameter search range, 20 spindle current, 21 spindle position, 22 servo motor current, 23 drive axis position, 24 control parameter, 30 tool information, 31 machining object information, 32 motion information of each axis, 40 learning data acquisition unit, 41 model generation unit, 42 machined surface quality judgment model, 43 chip break judgment model, 44 feedback calculation unit, 45 function update unit, 46 inference data acquisition unit, 47 inference unit, 48 condition and control parameter search model, 90 processing circuit, 91 control circuit, 92 processor, 93 memory, 100, 100c exploration condition input unit, 101 machining program input unit, 102, 102b, 102c, 102c-2, 102d, 102e, 102g, 102h condition exploration unit, 103 machining program generation device, 104 instruction value generation unit, 105, 105c mechanical model, 106 evaluation function, 107 AI model, 110 nth vibration trajectory, 111 n+1th vibration trajectory, 112 chip breaking area, 1 13 Mechanical model updating unit, 114 Fault prediction unit, 120 Trained model storage unit, 201 Spindle servo control unit, 202 Drive shaft servo control unit, 300 Spindle, 301 Drive shaft, 302 Spindle motor, 303 Processing object, 304 Tool, 305 Servo motor, 306 Acceleration sensor, 401 Control unit, 402 Input unit, 403 Storage unit, 404 Display unit, 405 Communication unit, 406 Output unit, 407 System bus.

Claims

1. A drive condition determination device for determining the drive conditions of a machine tool for performing vibration cutting in which a shaft that changes the relative position of a workpiece and a tool is vibrated during cutting to break chips, The driving condition determining device is characterized by having: an exploration condition input unit for obtaining exploration conditions, the exploration conditions including an exploration range set for at least one driving condition among the machining conditions of the working machine, the vibration conditions of the shaft, and the parameters determining the control characteristics of the shaft, i.e., control parameters, and the machining conditions that serve as a benchmark determined regardless of the presence or absence of vibration cutting, i.e., a benchmark machining condition; and A condition search unit determines the driving condition, i.e., the execution driving condition, used when performing the vibration cutting within the range set for the search range in a manner that satisfies the conditions for the occurrence of chip breakage, i.e., the breaking conditions, based on the search conditions.

2. The driving condition determination device according to claim 1, characterized in that: The condition search unit searches for the drive condition that satisfies the breaking condition based on the vibration amplitude estimated using a mechanical model that estimates the vibration amplitude at the tip position of the tool in accordance with a drive command to the axis.

3. The driving condition determination device according to claim 1 or 2, characterized in that: The condition search unit searches for a candidate of the execution driving condition that satisfies the breaking condition, and determines whether to set the candidate as the execution driving condition based on a first evaluation function for evaluating a cutting process using the searched candidate.

4. The driving condition determination device according to claim 3, characterized in that: The condition search unit searches for a plurality of candidates for the execution driving condition that satisfies the separation condition, and determines the execution driving condition based on the first evaluation function from among the plurality of searched candidates.

5. The driving condition determination device according to claim 3 or 4, characterized in that: The first evaluation function is represented by a linear sum greater than or equal to any one of electric energy required to generate vibration, vibration acceleration, machining time, cutting volume, maximum machining load, machine wear amount, and cumulative vibration cutting time.

6. The driving condition determination device according to any one of claims 1 to 5, characterized in that: Regarding the candidates of the execution driving conditions that satisfy the break condition, the condition exploration unit evaluates the candidates of the execution driving conditions based on a second evaluation function for evaluating whether an abnormality occurs in the cutting process using the candidate, and when it is evaluated that an abnormality occurs in the cutting process using the candidate, the candidate is not set as the execution driving condition.

7. The driving condition determination device according to claim 6, characterized in that: When the condition searching unit does not find the driving condition to be the execution driving condition based on the second evaluation function, the condition searching unit determines the driving condition including the reference machining condition and not performing the vibration cutting as the execution driving condition.

8. The driving condition determination device according to claim 6 or 7, characterized in that: The second evaluation function is represented by a linear sum greater than or equal to any one of the electric energy required to generate vibration, vibration acceleration, machining time, cutting volume, maximum machining load, machine wear amount, and cumulative vibration cutting time.

9. The driving condition determination device according to claim 2, characterized in that: The device further includes a mechanical model updating unit configured to update the mechanical model based on the measured value of the vibration amplitude.

10. The driving condition determination device according to claim 1, characterized in that: The condition exploration unit has: an estimation data acquisition unit that acquires the machining condition, the vibration condition, the material of the object to be machined, the material of the tool, the shape of the tool, the feed speed of the axis, and the acceleration of the axis as estimation data; and An inference unit, which uses a trained model for inferring the execution driving condition that best achieves the best processing quality of the cutting process among the execution driving conditions that satisfy the separation condition based on the processing conditions, the vibration conditions, the material of the processing object, the material of the tool, the shape of the tool, the feed speed of the axis and the acceleration of the axis, and infers the execution driving condition based on the inference data.

11. The driving condition determination device according to claim 1, characterized in that: The machining conditions include the rotation speed of the axis, the feed direction, and the feed amount per rotation of the axis.

12. The driving condition determination device according to any one of claims 1 to 11, characterized in that: The numerical control device further includes a command value generating unit that generates a drive command for the axis based on the execution drive condition determined by the condition searching unit.

13. A method for determining driving conditions of a machine tool for performing vibration cutting in which a shaft that changes the relative position of a workpiece and a tool is vibrated during cutting to break chips, The driving condition determination method is characterized by comprising the following steps: Acquiring exploration conditions, the exploration conditions including an exploration range set for at least one driving condition among the machining conditions of the working machine, the vibration conditions of the shaft, and the parameters determining the control characteristics of the shaft, i.e., control parameters, and the machining conditions, i.e., reference machining conditions, which are determined as a reference regardless of the presence or absence of vibration cutting; and Based on the search condition, the drive condition used when performing the vibration cutting, namely, the execution drive condition, is determined within the range set for the search range so as to satisfy the breaking condition, namely, the condition for the breaking of the chips.

Images