Processing inference apparatus

Abstract

The present invention relates to a machining inference device. The machining inference device (3, 5, 303) infers at least one of a state of a workpiece (W) or a tool (T, T2) at the time of machining, a shape of the workpiece (W), a shape of the tool (T, T2), and a mechanical state of a machining device (2, 302) using contact dynamic stiffness data (Ci, Ki) between the workpiece (W) and the tool (T, T2) exhibited by contact of the workpiece (W) with the tool (T, T2) at the time of machining in the machining device (2, 302) that machines the workpiece (W) with the tool (T, T2).

Description

Technical Field

[0001] This invention relates to a processing inference apparatus. Background Technology

[0002] Patent Document 1 describes a grinding simulation apparatus. The grinding simulation is performed by repeatedly calculating the amount of material removed from the workpiece based on the relative position of the workpiece and the grinding wheel, calculating the grinding resistance based on the amount of material removed, and calculating a correction amount for the relative position based on the grinding resistance. Furthermore, the calculation of the correction amount utilizes pre-determined support rigidity of the workpiece and support rigidity of the grinding wheel.

[0003] Patent document 2 describes a method for calculating the depth of grinding marks on a workpiece when it is machined by grinding with a grinding wheel, taking into account the static contact rigidity between the workpiece and the grinding wheel. The static contact rigidity used here is not a value measured when the grinding wheel is stationary, but is calculated using the theoretical static contact rigidity during grinding. The static contact rigidity is represented by the spring constant K between the workpiece and the grinding wheel.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2018-153907

[0005] Patent Document 2: Japanese Patent Application Publication No. 2015-208812

[0006] In the grinding simulation apparatus described in Patent Document 1, in addition to grinding resistance, the support rigidity of the workpiece and the support rigidity of the grinding wheel are used to calculate the correction amount for the relative position between the workpiece and the grinding wheel. However, using only the support rigidity of the workpiece and the support rigidity of the grinding wheel is insufficient to calculate the correction amount for the relative position with high accuracy. It is known that one of the reasons for this is the influence of the contact rigidity between the workpiece and the grinding wheel.

[0007] It is believed that by using the contact static rigidity described in Patent Document 2, the relative position of the workpiece and the grinding wheel can be calculated with high accuracy. However, even when using the contact static rigidity between the workpiece and the grinding wheel, there is room for improvement in accuracy.

[0008] Furthermore, the objects of inference are not limited to the correction amount for the relative position of the workpiece and the grinding wheel, but also take into account the shape of the workpiece, the shape of the grinding wheel, the condition of the workpiece or grinding wheel during grinding, and the mechanical condition of the grinding machine. In addition to grinding, this also applies to cutting processes. Summary of the Invention

[0009] The present invention provides a processing inference device capable of inferring the object of inference with higher accuracy.

[0010] One embodiment of the present invention provides a machining inference apparatus that, in a machining apparatus that processes a workpiece by means of a tool, uses dynamic contact rigidity data between the workpiece and the tool, as exhibited during machining, to infer at least one of the following during machining: the state of the workpiece or the tool, the shape of the workpiece, the shape of the tool, and the mechanical state of the machining apparatus.

[0011] The aforementioned contact dynamic rigidity data is generated from the relationship between the vibration force and the displacement of the workpiece while processing the workpiece with the aforementioned tool and applying a vibration force to the workpiece.

[0012] According to the above embodiment, contact dynamic rigidity data between the workpiece and the tool is used to infer at least one of the following during machining: the state of the workpiece or tool, the shape of the workpiece, the shape of the tool, and the mechanical state of the machining apparatus. The contact dynamic rigidity data is represented by the spring constant and attenuation coefficient between the workpiece and the tool, as exhibited through their contact. Thus, various inference objects can be inferred with high accuracy by using contact dynamic rigidity data including the spring constant and attenuation coefficient.

[0013] As described above, according to the above embodiments, a processing inference apparatus capable of inferring the inference object with higher accuracy can be provided.

[0014] Furthermore, the reference numerals in parentheses in the technical solutions indicate the correspondence with the specific mechanisms described in the embodiments described below, and do not limit the technical scope of the present invention. Attached Figure Description

[0015] Figure 1 This is a diagram showing a processing system that includes the processing apparatus and the processing inference apparatus of Embodiment 1.

[0016] Figure 2 This is a functional block diagram of the processing inference device in Implementation Method 1.

[0017] Figure 3 This is a schematic diagram illustrating the interference state between the workpiece and the grinding wheel during grinding.

[0018] Figure 4 It is a diagram that uses radial line segments to represent the shape of the workpiece in the grinding simulation. It is a diagram that shows the interference state between the workpiece, represented by radial line segments, and the outer periphery of the grinding wheel during grinding.

[0019] Figure 5 This is a schematic diagram representing the dynamic rigidity of contact, workpiece support, and tool support in grinding processes.

[0020] Figure 6A graph showing the contact dynamic stiffness table obtained through actual measurements.

[0021] Figure 7 It is a graph representing the contact dynamic stiffness table obtained through actual measurement and interpolation.

[0022] Figure 8 It is a diagram showing the dynamic rigidity of the workpiece support.

[0023] Figure 9 This is a flowchart representing the acquisition and processing of contact dynamic rigid data.

[0024] Figure 10 This is a top view of the grinding machine when obtaining contact dynamic rigidity data.

[0025] Figure 11 It is a diagram showing the state of the grinding machine in a part of the process of acquiring and processing dynamic rigid data.

[0026] Figure 12 This is a functional block diagram of the processing inference device in Implementation Method 2.

[0027] Figure 13 It is a graph representing the total dynamic stiffness table obtained through actual measurement and interpolation.

[0028] Figure 14 This is a diagram showing a processing system that includes the processing apparatus and the processing inference apparatus of Embodiment 3.

[0029] Figure 15 It is a schematic diagram representing the dynamic rigidity of contact, workpiece support, and tool support in cutting processes. Detailed Implementation

[0030] (Implementation Method 1)

[0031] 1. Structure of Processing Inference System 1

[0032] Reference Figure 1 Let me explain the machining inference system 1. The machining inference system 1 takes grinding machining using the grinding machine 2 as its object. The machining inference system 1 includes the grinding machine 2 as a machining device and the machining inference device 3.

[0033] Grinding machine 2 rotates the workpiece W, causing the grinding wheel T, which is a rotating body or tool, to rotate as well. The grinding wheel T approaches the workpiece W relative to the workpiece W in a direction intersecting the axis of the workpiece W, thereby grinding the outer or inner circumferential surface of the workpiece W. Grinding machine 2 can be a table-moving type grinding machine, a wheel-moving type grinding machine, etc. Furthermore, grinding machine 2 can be a cylindrical grinding machine, a cam grinding machine, etc.

[0034] In this embodiment, such as Figure 1As shown, the workpiece W is illustrated in the example of a shaft portion Wa that is not machined, and multiple machined portions Wb whose outer peripheral surfaces are to be ground. For example, the machined portions Wb have a cylindrical outer peripheral surface coaxial with the shaft portion Wa. However, Figure 1 The workpiece W shown is an example. The grinding machine 2 is capable of grinding workpieces with various shapes.

[0035] The machining inference device 3 infers at least one of the following during the grinding process in the grinding machine 2: the state of the workpiece W or the grinding wheel T, the shape of the workpiece W, the shape of the grinding wheel T, and the mechanical state of the grinding machine 2. The machining inference device 3 performs inference processing on the inference object by simulating the information used in the grinding process.

[0036] The machining inference device 3 can function as a simulation device independent of the grinding machine 2, or it can function as a simulation device that operates in conjunction with the grinding machine 2. In the former case, for example, it can determine the optimal grinding conditions without actually grinding the workpiece W. In the latter case, the machining inference device 3 can process the workpiece W in parallel with the grinding of the workpiece W performed by the grinding machine 2, thereby, for example, correcting the grinding conditions or performing actions to influence various controls. Furthermore, the machining inference device 3 can also be configured as an embedded system of the grinding machine 2.

[0037] 2. Structure of Grinding Machine 2

[0038] Reference Figure 1 Here is an example of the structure of the grinding machine 2. The grinding machine 2 is exemplified as a cylindrical grinding machine with a traversing table. That is, this grinding machine 2 has a structure in which the workpiece W moves along the axial direction of the workpiece W, and the grinding wheel T moves in a direction intersecting the axial direction of the workpiece W. Furthermore, in this embodiment, the grinding machine 2 is exemplified by grinding the outer circumferential surface of the cylindrical workpiece W using the grinding wheel T.

[0039] The grinding machine 2 includes a bed 10, a worktable 20, a spindle assembly 30, a tailstock assembly 40, a grinding wheel holder 50, a sizing device 60, and a control device 70. The bed 10 is mounted on a mounting surface. The bed 10 supports the front side in the X-axis direction (…). Figure 1 The width (length in the Z-axis direction) of the lower side is made longer, making the back side (in the X-axis direction) longer. Figure 1 The width of the upper side is relatively short.

[0040] The bed 10 has a Z-axis guide surface 11 extending along the Z-axis direction on its upper surface on the front side in the X-axis direction. Furthermore, a Z-axis drive mechanism 12 is provided on the bed 10, which drives along the Z-axis guide surface 11. In this embodiment, the Z-axis drive mechanism 12 is exemplified by having a ball screw mechanism 12a and a Z-axis motor 12b. The ball screw mechanism 12a extends parallel to the Z-axis guide surface 11, and the Z-axis motor 12b drives the ball screw mechanism 12a.

[0041] To drive the Z-axis drive mechanism 12, a Z-axis drive circuit (not shown) and a Z-axis detector 12c are provided. The Z-axis drive circuit includes an amplifier circuit and drives the Z-axis motor 12b. In this embodiment, the Z-axis detector 12c is, for example, an angle detector such as an encoder, which detects the angle of the rotation axis of the Z-axis motor 12b. Alternatively, the Z-axis drive mechanism 12 can replace the structure with the ball screw mechanism 12a described above, and a linear motor or the like can be used.

[0042] Furthermore, the bed 10 has a guide surface 13 extending along a direction intersecting the Z-axis on its upper surface on the back side in the X-axis direction. In this embodiment, the guide surface 13 is an X-axis guide surface extending along the X-axis direction orthogonal to the Z-axis. The bed 10 also includes an X-axis drive mechanism 14 that drives along the X-axis guide surface 13. In this embodiment, the X-axis drive mechanism 14 is exemplified by having a ball screw mechanism 14a and an X-axis motor 14b. The ball screw mechanism 14a extends parallel to the X-axis guide surface 13, and the X-axis motor 14b drives the ball screw mechanism 14a.

[0043] To drive the X-axis drive mechanism 14, an X-axis drive circuit (not shown) and an X-axis detector 14c are provided. The X-axis drive circuit includes an amplifier circuit and drives the X-axis motor 14b. In this embodiment, the X-axis detector 14c is, for example, an angle detector such as an encoder, which detects the angle of the rotation axis of the X-axis motor 14b. Alternatively, the X-axis drive mechanism 14 can replace the structure with the ball screw mechanism 14a described above, and a linear motor or the like can be used.

[0044] The worktable 20 is elongated and supported by the Z-axis guide surface 11 of the bed 10, allowing it to move along the Z-axis direction (horizontal left and right direction). Furthermore, the worktable 20 is fixed to the ball screw nut of the Z-axis ball screw mechanism 12a and moves along the Z-axis direction via the rotation of the Z-axis motor 12b.

[0045] The spindle assembly 30 constitutes a workpiece support device. The spindle assembly 30 supports the workpiece W and drives the workpiece W to rotate. The spindle assembly 30 is disposed on one end of the worktable 20 in the Z-axis direction. The spindle assembly 30 includes a spindle housing 31, a spindle 32, a spindle motor 33, a spindle center 34, a spindle detector 35, and a spindle drive circuit (not shown).

[0046] The spindle housing 31 is fixed to the worktable 20. The spindle 32 is supported by the spindle housing 31 via bearings and is rotatable. The spindle 32 is driven to rotate by a motor 33. The spindle center 34 supports the end face of one axial end of the workpiece W. The spindle center 34 is fixed to the spindle 32 and is configured to rotate relative to the spindle housing 31. However, if the spindle assembly 30 has a turning component such as a carrier (not shown), the spindle center 34 may also be fixed to the spindle housing 31 and is configured not to rotate relative to the spindle housing 31. Alternatively, the spindle assembly 30 may, instead of the spindle center 34, have a chuck for holding the workpiece W. Furthermore, the chuck is driven to rotate by being connected to the spindle 32.

[0047] A spindle detector 35 and a spindle drive circuit are provided for driving the spindle motor 33. In this embodiment, the spindle detector 35 is, for example, an angle detector such as an encoder, which detects the angle of the rotation axis of the spindle motor 33. The spindle drive circuit includes an amplifier circuit and drives the spindle motor 33.

[0048] The tailstock assembly 40, together with the spindle assembly 30, constitutes a workpiece support device. The tailstock assembly 40 is located on the other end of the worktable 20 in the Z-axis direction. The tailstock assembly 40 is configured to move along the Z-axis on the worktable 20. The tailstock assembly 40 has a tailstock center 41. The tailstock center 41 supports the end face of the other axial end of the workpiece W. The tailstock center 41 may be configured to be non-rotatable or rotatable. Furthermore, when the grinding machine 2 performs grinding on the inner circumferential surface of the workpiece W, the tailstock assembly 40 is not required.

[0049] Alternatively, the tailstock center 41 can be positioned fixed relative to the workpiece W, or it can be configured to move axially toward the workpiece W relative to the workpiece W. In the latter case, the tailstock center 41 can also be configured to adjust the axial pressing force relative to the workpiece W. The pressing force can be controlled by a mechanism for adjusting spring force, a mechanism for adjusting fluid pressure, or the like.

[0050] The grinding wheel holder 50 includes a grinding wheel T and drives the grinding wheel T to rotate. In addition to the grinding wheel T, the grinding wheel holder 50 also includes a grinding wheel holder body 51, a grinding wheel shaft 52, a grinding wheel motor 53, and a grinding wheel drive circuit (not shown).

[0051] The grinding wheel T is formed in a disc shape. The grinding wheel T is used to grind the outer or inner circumferential surface of the workpiece W. The grinding wheel T is constructed by fixing multiple abrasive grains together with a bonding agent. The abrasive grains can be general abrasive grains formed from ceramic materials such as alumina and silicon carbide, or ultra-abrasive grains such as diamond and CBN.

[0052] The binders include glass (V), resin (B), rubber (R), silicate (S), shellac (E), metal (M), electrophoresis (P), and magnesium cement (Mg). Furthermore, grinding wheels T can have porous or non-porous structures. Depending on the type of binder and the presence or absence of pores, grinding wheels T can have structures capable of elastic deformation or structures that are almost inelastically indeformable. Among the elastically deformable grinding wheels T, the elastic modulus varies depending on the type of binder, the presence or absence of pores, and the porosity.

[0053] The grinding wheel holder body 51 is, for example, rectangular in a top view, and is supported by the X-axis guide surface 13 of the bed 10, allowing it to move along the X-axis direction (horizontal back-and-forth direction). Furthermore, the grinding wheel holder body 51 is fixed to the ball screw nut of the X-axis ball screw mechanism 14a and is driven to move along the X-axis direction by the rotation of the X-axis motor 14b. The grinding wheel holder body 51 constitutes a grinding wheel support device that supports the grinding wheel T.

[0054] The grinding wheel spindle 52 is supported by the grinding wheel housing body 51 via bearings, enabling it to rotate. A grinding wheel T is fixed to the front end of the grinding wheel spindle 52, and the grinding wheel T rotates due to the rotation of the grinding wheel spindle 52. The grinding wheel is driven by a motor 53 to rotate the grinding wheel spindle 52. The bearings used are hydrostatic bearings, rolling bearings, etc.

[0055] The grinding wheel motor 53 transmits rotational driving force to the grinding wheel shaft 52, for example, via a belt. However, the grinding wheel motor 53 can be configured coaxially with the grinding wheel shaft 52. Generally, the rotational speed of the grinding wheel T generated by the drive of the grinding wheel motor 53 is higher than the rotational speed of the workpiece W generated by the drive of the spindle motor 33. A grinding wheel drive circuit is provided to drive the grinding wheel motor 53. The grinding wheel drive circuit includes an amplifier circuit to drive the grinding wheel motor 53.

[0056] A sizing device 60 is disposed on the upper surface of the bed 10 and measures the outer diameter of the workpiece W. The sizing device 60, for example, has a pair of contacts that can contact the outer peripheral surface of the workpiece W and measures the outer diameter of the contact portion with the workpiece W.

[0057] The control device 70 is a CNC (Computer Numerical Control) device and a PLC (Programmable Logic Controller) device that performs machining control. Specifically, based on the grinding program and the measurement results of the sizing device 60, the control device 70 drives the Z-axis drive mechanism 12 and the X-axis drive mechanism 14, which act as moving devices, to control the position of the worktable 20 and the grinding wheel holder 50. In other words, by controlling the position of the worktable 20 and the grinding wheel holder 50, the control device 70 brings the workpiece W and the grinding wheel T closer together and separates. Furthermore, the control device 70 controls the spindle assembly 30 and the grinding wheel holder 50. Specifically, the control device 70 controls the rotation of the spindle 32 and the grinding wheel T.

[0058] 3. Structure of processing inference device 3

[0059] Reference Figure 2 The structure of the machining inference device 3 will be explained below. The machining inference device 3 includes an instruction value acquisition unit 101, an inference unit 102, a contact dynamic rigidity table storage unit 103, a workpiece support dynamic rigidity table storage unit 104, a grinding wheel support dynamic rigidity table storage unit 105, a machining condition acquisition unit 106, a dynamic rigidity determination unit 107, a correction amount calculation unit 108, and an output unit 109.

[0060] The command value acquisition unit 101 acquires command values ​​used to control the grinding machine 2 during grinding. When the machining inference device 3 is a simulation device independent of the grinding machine 2, the command value acquisition unit 101 generates command values ​​for controlling various parts of the grinding machine 2 using calculations based on the grinding program and the input structural information of the grinding machine 2. Furthermore, when the machining inference device 3 functions as a simulation device that operates in conjunction with the grinding process performed by the grinding machine 2, the command value acquisition unit 101 can directly acquire command values ​​from the control device 70 of the grinding machine 2.

[0061] The inference unit 102 uses the instruction value obtained by the instruction value acquisition unit 101 to perform a grinding simulation, thereby inferring at least one of the following during the grinding process: the state of the workpiece W or the grinding wheel T, the shape of the workpiece W, the shape of the grinding wheel T, and the mechanical state of the grinding machine 2. The state of the workpiece W includes, for example, its vibration state and temperature state. The state of the grinding wheel T includes, for example, its vibration state, temperature state, grinding resistance generated at each part of the outer circumference of the grinding wheel T, the sharpness of the grinding wheel T, and the state of the abrasive grains constituting the grinding wheel T. The state of the abrasive grains includes, for example, the average protrusion of the abrasive grains and the abrasive grain distribution. The shape of the workpiece W includes the shape during the intermediate stage of the grinding process and the shape during the final stage of the grinding process. The shape of the grinding wheel T includes the shape during the intermediate stage of the grinding process and the shape during the final stage of the grinding process. The mechanical state of the grinding machine 2 includes the vibration state and temperature state of the parts constituting the grinding machine 2.

[0062] In this embodiment, the inference unit 102 simulates the gradual change in the shape of the workpiece W through grinding, and thus takes the shape of the workpiece W, the state of the workpiece W, and the mechanical state of the grinding machine 2 as inference objects. In this embodiment, the grinding wheel T is set as a non-deformable component for grinding simulation. In addition to the above-mentioned inference objects, the inference unit 102 can also infer the grinding resistance generated at each part of the outer peripheral surface of the grinding wheel T.

[0063] The inference unit 102 includes an interference calculation unit 111, a grinding efficiency calculation unit 112, a grinding characteristic determination unit 113, and a grinding resistance calculation unit 114.

[0064] The interference calculation unit 111 calculates the interference amount between the workpiece W and the grinding wheel T based on the relative position of the workpiece W and the grinding wheel T, the outer peripheral surface shape of the workpiece W, and the outer peripheral surface shape of the grinding wheel T obtained by the command value acquisition unit 101. The interference amount corresponds to the radial grinding amount of the workpiece W at each circumferential location. In other words, the interference amount is the amount of material removed by the grinding wheel T; more specifically, it is the radial grinding amount of the workpiece W at each circumferential location. Figure 3 As shown, the interference amount is the portion of the interference between the workpiece W and the grinding wheel T. Figure 3 The volume of the slanted portion (the interference region).

[0065] The interference calculation unit 111 calculates the interference geometrically through computational processing. Here, the interference calculation unit 111 stores the outer peripheral surface shape of the workpiece W and the outer peripheral surface shape of the grinding wheel T. For example... Figure 4As shown on the right side, the shape of the outer peripheral surface of workpiece W is represented by multiple radial line segments in polar coordinates with the rotation center Ow of workpiece W as the origin. That is, the interference calculation unit 111 stores the division points ( ) on the outer peripheral surface of workpiece W that are divided by equal angles (α). Figure 4 A group of line segments connecting the white dot (the point on the surface) to the rotation center Ow (origin) of the workpiece W. Figure 4 The white dots represent the dividing points, which are stored as the outer circumferential surface shape of the workpiece W before removal by the grinding wheel T.

[0066] The interference calculation unit 111 determines the intersection points of each line segment of the workpiece W and the line representing the shape of the outer circumference of the grinding wheel T based on the relative position (interaxial distance) of the workpiece W and the grinding wheel T. Figure 4 The black dots). The interference calculation unit 111 will determine the intersection point ( Figure 4 The black dots are stored as the outer peripheral surface shape of the workpiece W after it has been removed by the grinding wheel T. That is, the interference calculation unit 111 changes the stored outer peripheral surface shape of the workpiece W.

[0067] Furthermore, the interference calculation unit 111 subtracts the area of ​​the triangle △Ow-a1-a2 formed by the points b1 and b2 (intersection with the grinding wheel T) and the origin Ow from the area of ​​the triangle △Ow-a1-a2 formed by the points of the outer peripheral surface shape of the workpiece W before removal and the definition of removal. The area after subtraction is calculated for all adjacent points defining the outer peripheral surface shape of the workpiece W.

[0068] Furthermore, the interference calculation unit 111 accumulates the areas after each subtraction, and multiplies the total accumulated area by the thickness of the workpiece W to calculate the interference amount (removal amount). Additionally, as described above, the areas of the two triangles are calculated, and the difference between their areas is calculated, thereby calculating the area of ​​the removed portion. Alternatively, the area of ​​the removed portion can also be calculated by directly calculating quadrilateral a1-a2-b1-b2.

[0069] like Figure 2 As shown, the grinding efficiency calculation unit 112 calculates the grinding efficiency Z′ based on the interference amount calculated by the interference amount calculation unit 111. The grinding efficiency Z′ is calculated as the interference amount per unit time, that is, the volume of the workpiece W ground by the grinding wheel T per unit time.

[0070] The grinding characteristic determination unit 113 determines the grinding characteristic kc based on the material of the workpiece W, the abrasive grains of the grinding wheel T, the type of bonding agent, and the condition of the outer peripheral surface of the grinding wheel T. The condition of the outer peripheral surface of the grinding wheel T is represented, for example, by the wear state of the abrasive grains of the grinding wheel T or an index indicating sharpness. Here, the grinding characteristic determination unit 113 stores the grinding characteristics under each condition through prior experiments, analysis, etc.

[0071] The grinding resistance calculation unit 114 calculates the grinding resistance Fn in the normal direction (X-axis direction) of the outer peripheral surface of the workpiece W based on the grinding efficiency Z′ and the grinding characteristic kc. The grinding resistance Fn is obtained by multiplying the grinding efficiency Z′ by the grinding characteristic kc (Fn=kc×Z′).

[0072] Furthermore, the grinding characteristic kc exhibits a roughly linear relationship, such that the greater the grinding efficiency Z′, the greater the grinding resistance Fn in the normal direction (X-axis direction). However, this relationship changes, for example, when the grinding wheel T wears. For instance, when the grinding wheel T wears, the grinding resistance Fn in the normal direction increases relative to the grinding efficiency Z′.

[0073] The contact dynamic rigidity table storage unit 103 stores the contact dynamic rigidity data Ci and Ki between the workpiece W and the grinding wheel T. Specifically, the contact dynamic rigidity table storage unit 103 stores the correspondence between machining conditions and the contact dynamic rigidity data Ci and Ki. The workpiece support dynamic rigidity table storage unit 104 stores the workpiece support dynamic rigidity data Cw and Kw for the spindle device 30 and tailstock device 40, which serve as workpiece support devices. Specifically, the workpiece support dynamic rigidity table storage unit 104 stores the correspondence between machining conditions and the workpiece support dynamic rigidity data Cw and Kw. The grinding wheel support dynamic rigidity table storage unit 105 stores the grinding wheel support dynamic rigidity data Ct and Kt (tool support dynamic rigidity data) for the grinding wheel seat body 51, which serves as a grinding wheel support device. Specifically, the grinding wheel support dynamic rigidity table storage unit 105 stores the correspondence between machining conditions and the grinding wheel support dynamic rigidity data Ct and Kt.

[0074] Reference Figure 5 This section explains contact dynamic rigidity, workpiece support dynamic rigidity, and grinding wheel support dynamic rigidity. Contact dynamic rigidity is the dynamic rigidity between the workpiece W and the grinding wheel T, exhibited during grinding through the contact between them. Contact dynamic rigidity is defined by the attenuation coefficient Ci and the spring constant Ki. The attenuation coefficient Ci represents the relationship between the relative velocity of the workpiece W and the grinding wheel T and the external force acting on either the workpiece W or the grinding wheel T. The spring constant Ki represents the relationship between the relative position of the workpiece W and the grinding wheel T and the external force acting on either the workpiece W or the grinding wheel T.

[0075] In particular, the dynamic rigidity of the contact corresponds to the contact arc length L between the grinding wheel T and the workpiece W during grinding. For example... Figure 3 As shown, the contact arc length L is the length of the arc that contacts the workpiece W on the outer circumference surface of the grinding wheel T during grinding, within the cross-section of the grinding wheel T along the orthogonal direction of the axis. The contact arc length L varies depending on the feed rate of the grinding wheel T along the X-axis, the outer diameter of the grinding wheel T, and the outer diameter of the workpiece W.

[0076] Furthermore, the contact dynamic stiffness is caused by the elastic deformation of the grinding wheel T. That is, the contact dynamic stiffness is represented by the spring constant caused by the elastic deformation of the grinding wheel T and the attenuation coefficient caused by the elastic deformation of the grinding wheel T.

[0077] The workpiece support dynamic rigidity is the support-related dynamic rigidity of the spindle assembly 30 and the tailstock assembly 40, and is the dynamic rigidity exhibited by the spindle assembly 30 and the tailstock assembly 40, which constitute the workpiece support device of the grinding machine 2, when supporting the workpiece W. The workpiece support dynamic rigidity is defined by the attenuation coefficient Cw and the spring constant Kw. The attenuation coefficient Cw is a value representing the relationship between the relative velocity of the workpiece W with respect to the reference position of the spindle assembly 30 and the tailstock assembly 40 and the external force acting on the workpiece W. The spring constant Kw is a value representing the relationship between the relative position of the workpiece W with respect to the reference position of the spindle assembly 30 and the tailstock assembly 40 and the external force acting on the workpiece W.

[0078] The dynamic rigidity of the grinding wheel support is the dynamic rigidity related to the support within the grinding wheel holder body 51, and is the dynamic rigidity exhibited by the grinding wheel holder body 51, which serves as the grinding wheel support device constituting the grinding machine 2, when supporting the grinding wheel T. The dynamic rigidity of the grinding wheel support is defined by the attenuation coefficient Ct and the spring constant Kt. The attenuation coefficient Ct is a value representing the relationship between the relative velocity of the grinding wheel T relative to the reference position of the grinding wheel holder body 51 and the external force acting on the grinding wheel T. The spring constant Kt is a value representing the relationship between the relative position of the grinding wheel T relative to the reference position of the grinding wheel holder body 51 and the external force acting on the grinding wheel T.

[0079] Reference Figure 6 as well as Figure 7 The contact dynamic stiffness table storage unit 103 is described in detail. The contact dynamic stiffness table storage unit 103 stores a contact dynamic stiffness table containing contact dynamic stiffness data Ci and Ki. Specifically, as follows... Figure 6 As shown, the contact dynamic rigidity table is a table that sets the processing conditions, contact arc length L, mass Mi, attenuation coefficient Ci, and spring constant Ki as items. Figure 6 The contact dynamic stiffness table shown is a data table obtained through actual measurements, as described later.

[0080] exist Figure 6 In this context, processing conditions are, for example, the feed rate of the grinding wheel T along the X-axis, the outer diameter of the grinding wheel T, and the outer diameter of the workpiece W. As mentioned above, the contact arc length L is the length of the arc that contacts the workpiece W on the outer circumferential surface of the grinding wheel T during grinding, within a cross-section perpendicular to the grinding wheel T's axis. The contact arc length L is determined by the feed rate of the grinding wheel T along the X-axis, the outer diameter of the grinding wheel T, and the outer diameter of the workpiece W. The mass Mi includes the mass information of the workpiece W and the mass information of the grinding wheel T. The attenuation coefficient Ci and the spring constant Ki represent different values ​​depending on the processing conditions, i.e., the contact arc length L.

[0081] In addition to the processing conditions affecting the contact arc length L, the dynamic contact rigidity is also affected by the rotational speed of the grinding wheel T, the rotational speed of the workpiece W, and the amount of cooling. These are the main causes of temperature changes in the grinding wheel T and the workpiece W. In particular, when the bonding agent of the grinding wheel T is temperature-dependent, the dynamic contact rigidity changes with the temperature of the grinding wheel T. Therefore, the dynamic contact rigidity storage unit 103 has processing conditions that affect temperature as a unit. Figure 6 The table shown at the very beginning.

[0082] here, Figure 6 The contact dynamic rigidity table shown is a data table obtained by actual measurement of the first processing conditions (condition 1, condition 2, condition 3...), so it becomes a table limited to the contact arc length L corresponding to the first processing conditions that were measured.

[0083] However, the actual processing conditions vary widely, so in Figure 6 The contact dynamic stiffness table shown may be insufficient. Therefore, as... Figure 7 As shown, the measured contact dynamic stiffness table is used and interpolated to supplement the data.

[0084] That is, as a premise, such as Figure 6 As shown, the contact dynamic rigidity table storage unit 103 pre-stores the correspondence between the first processing conditions (condition 1, condition 2, condition 3, ...) and the measured contact dynamic rigidity data. Furthermore, by interpolating the contact dynamic rigidity data for the first processing conditions (condition 1, condition 2, condition 3, ...), contact dynamic rigidity data for second processing conditions (condition 1h, condition 2h, condition 3h), which are different from the first processing conditions, is generated. Moreover, the contact dynamic rigidity table storage unit 103 additionally stores the generated contact dynamic rigidity data.

[0085] Interpolation can be performed using experimental formulas that define the relationship between the contact arc length L, the attenuation coefficient Ci, and the spring constant Ki. Alternatively, interpolation can also be performed using machine learning, theoretical calculations, etc.

[0086] However, as mentioned above, the contact arc length L is determined by the feed rate of the grinding wheel T in the X-axis direction, the outer diameter of the grinding wheel T, and the outer diameter of the workpiece W. However, the contact arc length L changes during the grinding process due to the bending of the workpiece W and the grinding wheel T. Therefore, calculations under the same processing conditions may result in different contact arc lengths L depending on the amount of bending of the workpiece W and the grinding wheel T. Therefore, the contact dynamic rigidity table storage unit 103 can also store contact dynamic rigidity data Ci and Ki for multiple contact arc lengths L.

[0087] Reference Figure 8 The workpiece support dynamic rigidity table storage unit 104 is described in detail. The workpiece support dynamic rigidity table storage unit 104 stores the workpiece support dynamic rigidity data Cw and Kw. Specifically, as follows... Figure 8 As shown, the workpiece support dynamic rigidity table is a table that sets the processing conditions (condition 21, condition 22, condition 23...), mass Mw, attenuation coefficient Cw, and spring constant Kw as items.

[0088] For example, when the pressing force of the tailstock center 41 on the workpiece W in the axial direction of the workpiece W can be controlled, Figure 8 The processing conditions are set as the pressing force of the tailstock center 41. Furthermore, the workpiece support dynamic rigidity data Cw and Kw are data that change with the change in the contact state between the tailstock center 41 and the workpiece W, which varies according to the pressing force of the tailstock center 41. Therefore, the workpiece support dynamic rigidity table storage unit 104 stores the mass Mw, attenuation coefficient Cw, and spring constant Kw for each of the multiple pressing forces of the tailstock center 41.

[0089] The dynamic rigidity data of the workpiece support, Cw and Kw, can be obtained, for example, by hammering tests with the workpiece W supported by the spindle center 34 and the tailstock center 41. Furthermore, by performing hammering tests on changes in the pressure generated by the tailstock center 41, the dynamic rigidity data can be obtained... Figure 8 The workpiece support dynamic stiffness data Cw and Kw are shown. Additionally, the workpiece support dynamic stiffness table storage unit 104 has data for each type of workpiece W. Figure 8 The table shown at the very beginning.

[0090] As described above, the dynamic rigidity table storage unit 105 stores dynamic rigidity data Ct and Kt for the grinding wheel support. For example, the dynamic rigidity table storage unit 105 stores the dynamic rigidity data Ct and Kt in units of grinding wheel type T. Furthermore, in a structure where the grinding wheel T is supported by a hydrostatic bearing, when the pressure of the hydrostatic bearing can be controlled, the dynamic rigidity data Ct and Kt sometimes become data that varies according to the pressure of the hydrostatic bearing. Therefore, the dynamic rigidity table storage unit 105 can also store mass Mt, attenuation coefficient Ct, and spring constant Kt based on the pressure of the hydrostatic bearing, which is a machining condition.

[0091] like Figure 2As shown, the machining condition acquisition unit 106 acquires the machining conditions during grinding by the grinding machine 2. More specifically, the machining condition acquisition unit 106 acquires the machining conditions used for inference by the inference unit 102. The machining conditions acquired by the machining condition acquisition unit 106 are information used by the dynamic rigidity determination unit 107 to calculate various dynamic rigidities. The acquired machining conditions include, for example, the outer diameter of the workpiece W, the outer diameter of the grinding wheel T, the feed rate of the grinding wheel T in the X-axis direction, the rotational speed of the grinding wheel T, the rotational speed of the workpiece W, the coolant dosage, the pressing force of the tailstock center 41, and the pressure of the hydrostatic bearing of the grinding wheel T.

[0092] When the machining inference device 3 is a simulation device independent of the grinding machine 2, the machining condition acquisition unit 106 acquires the machining conditions contained in the grinding program by inputting the grinding program. Alternatively, when the machining inference device 3 functions as a simulation device that operates in conjunction with the grinding process performed by the grinding machine 2, the machining condition acquisition unit 106 can also acquire the machining conditions by inputting the grinding program from the control device 70, or it can directly acquire information related to the machining conditions from the control device 70 of the grinding machine 2.

[0093] The dynamic rigidity determination unit 107 determines the dynamic rigidity data that affects the grinding process. The dynamic rigidity determination unit 107 determines the contact dynamic rigidity data, the workpiece support dynamic rigidity data, and the grinding wheel support dynamic rigidity data, respectively. That is, the dynamic rigidity determination unit 107 includes a contact dynamic rigidity determination unit 121, a workpiece support dynamic rigidity determination unit 122, and a grinding wheel support dynamic rigidity determination unit 123.

[0094] The contact dynamic rigidity determination unit 121 obtains the contact dynamic rigidity table stored in the contact dynamic rigidity table storage unit 103 and the processing conditions obtained by the processing condition acquisition unit 106. Furthermore, the contact dynamic rigidity determination unit 121 determines the contact dynamic rigidity data Ci and Ki corresponding to the acquired processing conditions from the contact dynamic rigidity table.

[0095] The contact arc length L also varies depending on the outer peripheral surface shape of the workpiece W, which changes due to the grinding process, the workpiece W caused by the grinding resistance, and the bending of the grinding wheel T. Furthermore, the interference calculation unit 111 of the inference unit 102 sequentially calculates the outer peripheral surface shape of the workpiece W and the relative position of the workpiece W and the grinding wheel T during the grinding process.

[0096] Therefore, the contact dynamic rigidity determination unit 121 can also obtain the simulated current moment outer peripheral surface shape of the workpiece W and the relative position of the workpiece W and the grinding wheel T from the interference quantity calculation unit 111, and calculate the contact arc length L. Moreover, the contact dynamic rigidity determination unit 121 can also determine the contact dynamic rigidity data Ci and Ki corresponding to the calculated contact arc length L from the contact dynamic rigidity table.

[0097] The workpiece support dynamic rigidity determination unit 122 acquires the workpiece support dynamic rigidity table stored in the workpiece support dynamic rigidity table storage unit 104, and the processing conditions acquired by the processing condition acquisition unit 106. Furthermore, the workpiece support dynamic rigidity determination unit 122 determines the workpiece support dynamic rigidity data Cw and Kw corresponding to the acquired processing conditions from the workpiece support dynamic rigidity table.

[0098] The grinding wheel support dynamic rigidity determination unit 123 acquires the grinding wheel support dynamic rigidity table stored in the grinding wheel support dynamic rigidity table storage unit 105, and the processing conditions acquired by the processing condition acquisition unit 106. Furthermore, the grinding wheel support dynamic rigidity determination unit 123 determines the grinding wheel support dynamic rigidity data Ct and Kt corresponding to the acquired processing conditions from the grinding wheel support dynamic rigidity table.

[0099] The correction calculation unit 108 calculates the correction amount for the relative displacement of the grinding wheel T and the workpiece W in the X-axis direction due to the grinding resistance Fn, based on the dynamic rigidity data determined by the dynamic rigidity determination unit 107. The correction amount related to displacement can be obtained from the dynamic rigidity data and the grinding resistance Fn. That is, the correction amount related to displacement can be calculated based on the grinding resistance Fn, the contact dynamic rigidity data Ci and Ki, the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt.

[0100] The correction calculation unit 108 outputs the calculated correction amount to the inference unit 102. As described above, the inference unit 102 infers the object to be inferred based on the relative position of the workpiece W and the grinding wheel T, the outer peripheral surface shape of the workpiece W, and the outer peripheral surface shape of the grinding wheel T obtained by the command value acquisition unit 101. However, the relative position of the workpiece W and the grinding wheel T becomes a different position than the relative position based on the command value due to the grinding resistance Fn.

[0101] Therefore, when the inference unit 102 infers the object of inference, it uses the relative position of the workpiece W and the grinding wheel T, in addition to the relative position obtained by the command value acquisition unit 101, a relative position with a correction amount calculated by the correction amount calculation unit 108. That is, the inference unit 102 infers the object of inference based on the relative position of the command value and the correction amount calculated using each dynamic rigidity data.

[0102] In particular, in this embodiment, the correction calculation unit 108 outputs the calculated correction amount to the interference calculation unit 111 of the inference unit 102. As described above, the interference calculation unit 111 calculates the interference amount between the workpiece W and the grinding wheel T based on the relative position of the workpiece W and the grinding wheel T, the outer peripheral surface shape of the workpiece W, and the outer peripheral surface shape of the grinding wheel T obtained by the command value acquisition unit 101. However, the relative position between the workpiece W and the grinding wheel T is different from the relative position of the command value due to the grinding resistance Fn.

[0103] Therefore, the interference calculation unit 111 uses the relative position of the workpiece W and the grinding wheel T for calculating the interference amount, in addition to the relative position obtained by the command value acquisition unit 101, and also uses the relative position with the correction amount calculated by the correction amount calculation unit 108. That is, the interference calculation unit 111 calculates the interference amount based on the relative position of the command value and the correction amount calculated using each dynamic rigidity data.

[0104] In order for the interference calculation unit 111 to calculate the interference amount taking into account the correction amount, the grinding efficiency calculation unit 112, the grinding characteristic determination unit 113, and the grinding resistance calculation unit 114 obtain the grinding efficiency Z′, grinding characteristic kc, and grinding resistance Fn based on the interference amount taking into account the correction amount.

[0105] The output unit 109 outputs the inferred object deduced by the inference unit 102. That is, the output unit 109 infers at least one of the following during grinding: the state of the workpiece W or the grinding wheel T, the shape of the workpiece W, the shape of the grinding wheel T, and the mechanical state of the grinding machine 2. The output unit 109 may also teach the inference result to a teaching device (not shown). Alternatively, the output unit 109 may also output the inference result to the control device 70 of the grinding machine 2. In this case, the control device 70 can use the inference result, for example, to correct the grinding conditions.

[0106] 4. Processing of dynamic rigidity data in contact

[0107] Reference Figures 9-11 This will illustrate the acquisition and processing of contact dynamic rigidity data. For example... Figure 9 As shown, the process of acquiring and processing contact dynamic rigidity data first involves mounting the measuring fixture 4 onto the grinding machine 2 and the workpiece W (S1). The measuring fixture 4 is a non-contact vibrator, a device that applies vibrational power to the workpiece W. Figure 10 As shown, the measuring fixture 4 is mounted on the upper surface of the worktable 20. The measuring fixture 4 can be adjusted to a fixed position in the Z-axis direction on the upper surface of the worktable 20.

[0108] The measuring fixture 4 holds the workpiece W in a state where it is inserted. Specifically, a portion of the shaft Wa, which is a non-machined part of the workpiece W, is inserted into the measuring fixture 4, while the multiple machined parts Wb, which are the objects of grinding, are located outside the measuring fixture 4. The workpiece W, inserted into and held by the measuring fixture 4, is supported by the spindle assembly 30 and the tailstock assembly 40, as in normal grinding operations.

[0109] Here, refer to Figure 11 (a)~ Figure 11(c) The structure of the measuring fixture 4 will be described. The measuring fixture 4 includes a housing 131, an electromagnet 132, a rotor 133, a locking nut 134, a displacement sensor 135, and a control device 136. The housing 131 is fixed to the upper surface of the worktable 20 of the grinding machine 2. Furthermore, the housing 131 has a hole 131a extending along the Z-axis direction.

[0110] An electromagnet 132 is embedded in the housing 131. A rotor 133 is mounted on the outer peripheral surface of the workpiece W, integrally formed with the workpiece W. The rotor 133 is formed of a magnetic material and moves by the magnetic force generated by the electromagnet 132. The rotor 133 is cylindrical, and its outer peripheral surface is positioned with a predetermined gap relative to the inner peripheral surface of the housing 131. This gap determines the distance the rotor 133 can move relative to the housing 131. The inner peripheral surface of the rotor 133 is formed according to the shape of the outer peripheral surface of the workpiece W. A locking nut 134 is used to fix the rotor 133 to the workpiece W. The method of fixing the rotor 133 is not limited to a mechanism using the locking nut 134; various mechanisms can be employed.

[0111] The displacement sensor 135 is positioned near the inner circumferential surface of the housing 131 to measure the distance from the outer circumferential surface of the rotor 133. That is, when the rotor 133 is vibrated by the electromagnet 132, the displacement sensor 135 measures the displacement of the rotor 133 relative to the inner circumferential surface of the housing 131 in the direction of approaching and separating (hereinafter referred to as radial displacement).

[0112] like Figure 11 As shown in (c), the control device 136 supplies a drive current to the electromagnet 132 in order to apply a vibration force to the electromagnet 132. In addition, the control device 136 obtains the displacement measured by the displacement sensor 135, that is, the radial displacement of the rotor 133.

[0113] Therefore, in Figure 9 In S1, such as Figure 11 As shown in (a), the shaft portion Wa, which is the non-machined part of the workpiece W, is inserted into the rotor 133 of the measuring fixture 4. Furthermore, as... Figure 11 As shown in (b), the rotor 133 is fixed to the workpiece W by locking nut 134.

[0114] Furthermore, the housing 131 of the measuring fixture 4 is mounted on the worktable 20. The workpiece W, on which the rotor 133 is mounted, is then supported by the spindle assembly 30 and the tailstock assembly 40. At this time, as... Figure 11 As shown in (b), the position of the housing 131 is adjusted so that the outer peripheral surface of the rotor 133 is opposite to the inner peripheral surface of the housing 131 of the measuring fixture 4.

[0115] Next, the grinding process begins (S2). That is, while the workpiece W and the grinding wheel T are rotating, the grinding wheel T is moved along the X-axis to grind the outer peripheral surface of the machining part Wb of the workpiece W.

[0116] Next, a vibration force is applied by measuring fixture 4 (S3). The application of the vibration force by measuring fixture 4 is performed simultaneously with the grinding of workpiece W by grinding wheel T. The applied vibration force can be pulse vibration or scanning vibration with continuously varying vibration frequency. The application of the vibration force is performed by supplying current to electromagnet 132 through control device 136 of measuring fixture 4. Moreover, the vibration force is controlled by the current supplied to electromagnet 132 by control device 136.

[0117] Next, while applying vibration during grinding, the radial displacement of the rotor 133 is measured by the displacement sensor 135 of the measuring fixture 4 (S4). Here, the displacement of the rotor 133 on the workpiece W coincides with the radial displacement of the part fixed to the rotor 133. Therefore, the displacement sensor 135 of the measuring fixture 4 measures the radial displacement generated by the workpiece W when vibration is applied to the workpiece W.

[0118] Next, the grinding process ends when the displacement sensor 135 finishes measuring (S5).

[0119] Next, the total dynamic rigidity data Ccom and Kcom during grinding are calculated (S6). The total dynamic rigidity data Ccom and Kcom are the sum (composite) dynamic rigidity data represented by the contact dynamic rigidity data Ci and Ki, the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt. The total dynamic rigidity data Ccom and Kcom are expressed as the sum of the contact dynamic rigidity data Ci and Ki, the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt.

[0120] As described above, the radial displacement measured by the displacement sensor 135 of the measuring fixture 4 is measured while a vibration force is applied to the workpiece W during grinding. Therefore, the measured displacement is influenced by the contact dynamic rigidity data Ci, Ki, the workpiece support dynamic rigidity data Cw, Kw, and the grinding wheel support dynamic rigidity data Ct, Kt. Therefore, the calculation of the sum of the dynamic rigidity data Ccom, Kcom becomes data generated from the relationship between the vibration force and the radial displacement of the workpiece W while a vibration force is applied to the workpiece W during grinding.

[0121] Next, the dynamic rigidity data Cw and Kw of the workpiece support and the dynamic rigidity data Ct and Kt of the grinding wheel support were obtained (S7). The dynamic rigidity data Cw and Kw of the workpiece support and the dynamic rigidity data Ct and Kt of the grinding wheel support were obtained in advance through hammering tests, etc.

[0122] Next, the contact dynamic rigidity data Ci and Ki (S8) are calculated. The contact dynamic rigidity data Ci and Ki are obtained by subtracting the workpiece support dynamic rigidity data Cw and Kw and the grinding wheel support dynamic rigidity data Ct and Kt from the total dynamic rigidity data Ccom and Kcom.

[0123] Next, interpolation processing of the contact dynamic rigidity data Ci and Ki is performed (S9). The interpolation processing is described in detail below. Figure 7 As described above, interpolation is the process of using measured contact dynamic rigidity data Ci and Ki to derive contact dynamic rigidity data Ci and Ki under grinding conditions different from the measured values. Interpolation can apply experimental formulas, machine learning, theoretical calculations, etc. In this way, it is possible to obtain... Figure 7 The contact dynamic rigidity data Ci and Ki are shown.

[0124] 5. Effects

[0125] In this embodiment, contact dynamic rigidity data Ci and Ki are used to infer at least one of the following during grinding: the state of the workpiece W or the grinding wheel T, the shape of the workpiece W, the shape of the grinding wheel T, and the mechanical state of the grinding machine 2. The contact dynamic rigidity data Ci and Ki are represented by the spring constant Ki and the attenuation coefficient Ci between the workpiece W and the grinding wheel T, as exhibited by the contact between the workpiece W and the grinding wheel T. Thus, by using contact dynamic rigidity data Ci and Ki that includes the spring constant Ki and the attenuation coefficient Ci, various inference objects can be inferred with high accuracy. Furthermore, contact static rigidity data, which differs from the contact dynamic rigidity data, is represented only by the spring constant K and does not include the attenuation coefficient C.

[0126] In particular, the contact dynamic rigidity data Ci and Ki are data caused by the elastic deformation of the grinding wheel T, and correspond to the contact arc length L of the grinding wheel T contacting the workpiece W during grinding. Even when significantly affected by the elastic deformation of the grinding wheel T, the object being analyzed can be inferred with high accuracy by using the aforementioned contact dynamic rigidity data Ci and Ki.

[0127] Additionally, as referenced Figure 9 As shown in the process of obtaining contact dynamic rigidity data, the contact dynamic rigidity data Ci and Ki are data generated from the relationship between the vibration force and the displacement of the workpiece W when the workpiece W is being ground by the grinding wheel T and a vibration force is being applied to the workpiece W. Therefore, the contact dynamic rigidity data Ci and Ki during grinding can be reliably obtained.

[0128] Additionally, as referenced Figure 6 as well as Figure 7 As explained, the contact dynamic rigidity table storage unit 103 pre-stores a contact dynamic rigidity table that represents at least one of the processing conditions, including the relative position of the workpiece W and the grinding wheel T, the rotational speed of the grinding wheel T, and the rotational speed of the workpiece W, and the correspondence between these conditions and the contact dynamic rigidity data Ci and Ki. Furthermore, the inference unit 102 uses the processing conditions at the time of inference and the correspondence stored in the contact dynamic rigidity table storage unit 103 to infer the object to be inferred during the grinding process. In this way, by pre-storing the relationships between the contact dynamic rigidity data Ci and Ki and various processing conditions, the object to be inferred can be inferred with high accuracy using the contact dynamic rigidity data Ci and Ki that correspond to the actual processing conditions at the time of inference.

[0129] In addition, such as Figure 7 As shown, in the contact dynamic rigidity table storage unit 103, contact dynamic rigidity data Ci and Ki, which are different from the measured first processing conditions, are added and stored by interpolation processing. In this way, contact dynamic rigidity data Ci and Ki for the second processing conditions that are not measured can be generated, and the inferred object can be inferred with high accuracy.

[0130] The correction calculation unit 108 calculates the correction amount using not only the contact dynamic rigidity data Ci and Ki, but also the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt. Furthermore, the inference unit 102 uses the correction amount calculated using the contact dynamic rigidity data Ci and Ki, the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt to infer the object being inferred. In this way, not only the contact dynamic rigidity data, but also the dynamic rigidity data for the workpiece support rigidity and the grinding wheel support rigidity are used. Therefore, the object can be inferred with high accuracy.

[0131] like Figure 2 As shown, the dynamic rigidity determination unit 107 of the machining inference device 3 determines the contact dynamic rigidity data Ci, Ki, the workpiece support dynamic rigidity data Cw, Kw, and the grinding wheel support dynamic rigidity data Ct, Kt, respectively. Specifically, the machining inference device 3 stores the dynamic rigidity data of each object in the contact dynamic rigidity table storage unit 103, the workpiece support dynamic rigidity table storage unit 104, and the grinding wheel support dynamic rigidity table storage unit 105. Furthermore, the correction amount calculation unit 108 uses the determined contact dynamic rigidity data Ci, Ki, workpiece support dynamic rigidity data Cw, Kw, and grinding wheel support dynamic rigidity data Ct, Kt to calculate the correction amount.

[0132] In this way, by separately storing the contact dynamic rigidity data Ci and Ki, the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt, and then performing the decision-making process, versatility can be improved. For example, when changing the type of workpiece W, it can be used directly without changing the grinding wheel support dynamic rigidity data Ct and Kt. Furthermore, when changing the pressing force of the tailstock center 41, only the workpiece support dynamic rigidity data Cw and Kw need to be changed.

[0133] Furthermore, in cylindrical grinding machines with the workpiece W supported at the other axial end by the tailstock center 41, the tailstock center 41 is often configured to control the pressing force relative to the workpiece W along its axial direction. In this case, the workpiece support dynamic rigidity data Cw and Kw become data that change with the contact state between the tailstock center 41 and the workpiece W, which changes due to the change in the pressing force of the tailstock center 41. That is, by setting the workpiece support dynamic rigidity data Cw and Kw to take into account the pressing force of the tailstock center 41, the object being analyzed can be deduced with higher accuracy.

[0134] (Implementation Method 2)

[0135] Reference Figure 12 as well as Figure 13 The processing inference apparatus 5 of Embodiment 2 will now be described. Here, the processing inference apparatus 5 differs from the processing inference apparatus 3 of Embodiment 1 in that the dynamic stiffness determination unit 207 and the total dynamic stiffness table storage unit 203 are different. In the processing inference apparatus 5, the other structures are the same as in Embodiment 1.

[0136] The overall dynamic rigidity table storage unit 203 stores overall dynamic rigidity data Ccom and Kcom. Specifically, the overall dynamic rigidity table storage unit 203 stores the correspondence between processing conditions and the overall dynamic rigidity data Ccom and Kcom. The overall dynamic rigidity data Ccom and Kcom are obtained from contact dynamic rigidity data as described in Embodiment 1. That is, the overall dynamic rigidity data Ccom and Kcom are composite dynamic rigidity data represented by contact dynamic rigidity data Ci and Ki, workpiece support dynamic rigidity data Cw and Kw, and grinding wheel support dynamic rigidity data Ct and Kt.

[0137] The dynamic rigidity determination unit 207 determines the dynamic rigidity data that affects the grinding process. In this embodiment, the dynamic rigidity determination unit 207 includes a combined dynamic rigidity determination unit 221. The combined dynamic rigidity determination unit 221 determines a combined dynamic rigidity data Ccom, Kcom that combines the contact dynamic rigidity data Ci, Ki, the workpiece support dynamic rigidity data Cw, Kw, and the grinding wheel support dynamic rigidity data Ct, Kt.

[0138] The overall dynamic rigidity determination unit 221 obtains the overall dynamic rigidity table stored in the overall dynamic rigidity table storage unit 203 and the processing conditions obtained by the processing conditions acquisition unit 106. Furthermore, the overall dynamic rigidity determination unit 221 determines the overall dynamic rigidity data Ccom and Kcom corresponding to the acquired processing conditions from the overall dynamic rigidity table.

[0139] According to this embodiment, by using the aggregate dynamic rigidity data Ccom and Kcom, which combine the contact dynamic rigidity data Ci and Ki, the workpiece support dynamic rigidity data Cw and Kw, and the grinding wheel support dynamic rigidity data Ct and Kt, the object to be inferred can be determined with high accuracy. However, when there are multiple types of workpieces W and grinding wheels T, multiple processing conditions, and multiple pressing pressures controlled by the tailstock center 41, the data in the aggregate dynamic rigidity table often becomes large. In such cases, it is preferable to process each dynamic rigidity data separately, as described in Embodiment 1 above.

[0140] (Implementation Method 3)

[0141] Reference Figure 14 The machining inference system 301 of this embodiment will be described below. The machining inference system 301 takes cutting machining using a lathe 302 as its object. The machining inference system 301 includes a lathe 302 as a machining device and a machining inference device 303.

[0142] The lathe 302 turns the workpiece W by rotating the workpiece W and moving the cutting tool T2 relative to the workpiece W. The machining inference device 303 infers at least one of the following during the cutting process of the lathe 302: the state of the workpiece W or the cutting tool T2, the shape of the workpiece W, the shape of the cutting tool T2, and the mechanical state of the lathe 302. The machining inference device 303 performs the inference process by inputting information used in the cutting process and performing a simulation.

[0143] The lathe 302 includes, for example, a bed 310, a spindle assembly 320, a tailstock assembly 330, a support assembly 340, a tool table 350, and a control device 360. The spindle assembly 320 constitutes a workpiece support device. The spindle assembly 320 is fixed to the upper surface of the bed 310, supports one end of the workpiece W, and drives the workpiece W to rotate. The spindle assembly 320 includes a spindle housing 321, a spindle 322, a spindle motor 323, a chuck 324, a spindle detector 325, and a spindle drive circuit (not shown).

[0144] The spindle housing 321 is fixed to the bed 310. The spindle 322 is supported by the spindle housing 321 via bearings and is rotatable. The spindle motor 323 drives the spindle 322 to rotate. The chuck 324 is fixed to the spindle 322 and holds one end of the workpiece W. The spindle detector 325 and the spindle drive circuit are provided for driving the spindle motor 323.

[0145] The tailstock assembly 330, together with the spindle assembly 320, constitutes a workpiece support device. The tailstock assembly 330 is mounted on the bed 310 and positioned opposite the spindle assembly 320 in the Z-axis direction. The tailstock assembly 330 is configured to be movable along the Z-axis on the bed 310. The tailstock assembly 330 has a tailstock center 331 that supports the other end of the workpiece W.

[0146] The support device 340 is fixed to the bed 310 and supports the outer peripheral surface of the axial intermediate portion of the workpiece W. In particular, the support device 340 is positioned to overcome the cutting load on the workpiece W from the cutting tool T2.

[0147] The tool table 350 includes a Z-axis slide 351, an X-axis slide 352, a rotary table (rotary tool post) 353, and multiple cutting tools T2. The Z-axis slide 351 is supported by the Z-axis guide surface 311 of the bed 310 so that it can move along the Z-axis direction, and moves along the Z-axis direction by the Z-axis drive mechanism 312 provided on the bed 310.

[0148] The X-axis slide table 352 is supported by the X-axis guide surface 351a on the Z-axis slide table 351, enabling it to move along the X-axis direction. It is further moved along the X-axis direction by the X-axis drive mechanism 351b mounted on the Z-axis slide table 351. A rotary table 353 is mounted on the X-axis slide table 352 in a manner capable of rotating about an axis parallel to the Z-axis direction. Multiple cutting tools T2 are fixed to the outer circumferential surface of the rotary table 353. The multiple cutting tools T2 can be of different types.

[0149] The control device 360 ​​is a CNC (Computer Numerical Control) device and a PLC (Programmable Logic Controller) device that performs machining control. Specifically, based on the cutting machining program, the control device 360 ​​drives the Z-axis drive mechanism 312 and the X-axis drive mechanism 351b, which act as moving devices, to control the position of the cutting tool T2. In other words, by controlling the position of the cutting tool T2, the control device 360 ​​causes the workpiece W to move relative to the cutting tool T2. Furthermore, the control device 360 ​​controls the rotation of the spindle 322 and the rotary table 353.

[0150] The processing inference device 303 of this embodiment and Figure 2The structure of the machining inference device 3 in Embodiment 1 is the same. However, the grinding in Embodiment 1 is changed to cutting, and the grinding wheel T is changed to the cutting tool T2.

[0151] Next, in this embodiment, refer to Figure 15 This section explains contact dynamic rigidity, workpiece support dynamic rigidity, and tool support dynamic rigidity. Contact dynamic rigidity is the dynamic rigidity between the workpiece W and the cutting tool T2, and it is the dynamic rigidity exhibited during machining through the contact between the workpiece W and the cutting tool T2. Contact dynamic rigidity is defined by the attenuation coefficient Ci and the spring constant Ki. The attenuation coefficient Ci is a value representing the relationship between the relative velocity of the workpiece W and the cutting tool T2 and the external force acting on the workpiece W or the cutting tool T2. The spring constant Ki is a value representing the relationship between the relative position of the workpiece W and the cutting tool T2 and the external force acting on the workpiece W or the cutting tool T2.

[0152] In particular, the dynamic rigidity of the contact corresponds to the contact length L between the cutting tool T2 and the workpiece W during machining. The contact length L varies depending on the size of the cutting tool T2, the shape of the front end of the cutting tool T2, the rake angle and clearance angle of the cutting tool T2, the depth of cut, and the outer diameter of the workpiece W.

[0153] The workpiece support dynamic rigidity is the support-related dynamic rigidity of the spindle assembly 320, tailstock assembly 330, and support assembly 340, and is the dynamic rigidity exhibited by the spindle assembly 320, tailstock assembly 330, and support assembly 340, which constitute the workpiece support device of the lathe 302, when supporting the workpiece W. The workpiece support dynamic rigidity is defined by the attenuation coefficient Cw and the spring constant Kw. The attenuation coefficient Cw is a value representing the relationship between the relative velocity of the workpiece W with respect to the reference position of the spindle assembly 320, tailstock assembly 330, and support assembly 340 and the external force acting on the workpiece W. The spring constant Kw is a value representing the relationship between the relative position of the workpiece W with respect to the reference position of the spindle assembly 320, tailstock assembly 330, and support assembly 340 and the external force acting on the workpiece W.

[0154] The tool support dynamic stiffness is the support-related dynamic stiffness of the tool table 350, and it is the dynamic stiffness exhibited by the tool table 350, which constitutes the tool support device of the lathe 302, when supporting the cutting tool T2. The tool support dynamic stiffness is defined by the attenuation coefficient Ct and the spring constant Kt. The attenuation coefficient Ct is a value representing the relationship between the relative velocity of the cutting tool T2 with respect to the reference position of the tool table 350 and the external force acting on the cutting tool T2. The spring constant Kt is a value representing the relationship between the relative position of the cutting tool T2 with respect to the reference position of the tool table 350 and the external force acting on the cutting tool T2.

[0155] The processing inference system 301 of this embodiment achieves the same effect as the processing inference system 1 of embodiment 1.

[0156] (other)

[0157] In the above embodiments, grinding operations using a grinding machine 2 and cutting operations using a lathe 302 were described as examples. However, the same principles can be applied to cutting operations using a machining center.

Claims

1. A processing inference device (3, 5, 303), wherein, In a machining apparatus (2, 302) that processes a workpiece (W) using tools (T, T2), dynamic contact rigidity data (Ci, Ki) between the workpiece and the tool, as exhibited during machining due to their contact, are used to infer at least one of the following: the state of the workpiece or the tool during machining; the shape of the workpiece; the shape of the tool; and the mechanical state of the machining apparatus. The mechanical state includes at least one of a vibration state and a temperature state. The aforementioned contact dynamic stiffness data are defined by the attenuation coefficient Ci and the spring constant Ki. The attenuation coefficient Ci represents the relationship between the relative velocity of the workpiece and the tool and the external force acting on the workpiece or the tool. The spring constant Ki represents the relationship between the relative position of the workpiece and the tool and the external force acting on the workpiece or the tool. The aforementioned contact dynamic rigidity data is generated from the relationship between the vibration force and the displacement of the workpiece while processing the workpiece with the aforementioned tool and applying a vibration force to the workpiece.

2. The processing inference device (3, 5) according to claim 1, wherein, The tool mentioned above is a grinding wheel (T). The aforementioned contact dynamic rigidity data is data caused by the elastic deformation of the aforementioned grinding wheel, and is data corresponding to the contact arc length (L) of the aforementioned grinding wheel and the aforementioned workpiece during grinding.

3. The processing inference device (3, 5, 303) according to claim 1, wherein, The aforementioned contact dynamic rigidity data is data caused by the elastic deformation of the aforementioned tool or workpiece, and corresponds to the processing conditions during processing.

4. The processing inference apparatus (3, 303) according to any one of claims 1 to 3, comprising: The contact dynamic rigidity table storage unit (103) pre-stores processing conditions including at least one of the relative positions of the workpiece and the tool, the rotational speed of the tool as a rotating body, and the rotational speed of the workpiece, and the correspondence between these conditions and the contact dynamic rigidity data; and The inference unit (102) uses the processing conditions at the time of inference and the corresponding relationship stored in the contact dynamic rigidity table storage unit to infer at least one of the state of the workpiece or the tool, the shape of the workpiece, the shape of the tool, and the mechanical state of the processing device at the time of processing.

5. The processing inference device (3, 303) according to claim 4, wherein, The above-mentioned contact dynamic rigidity table storage unit The correspondence between the aforementioned processing conditions, i.e. the first processing conditions, and the aforementioned contact dynamic rigidity data measured under the aforementioned first processing conditions is stored in advance. Furthermore, by interpolating the contact dynamic rigidity data for the first processing condition, additional contact dynamic rigidity data for the second processing condition, which is different from the first processing condition, is stored.

6. The processing inference apparatus (3, 5, 303) according to any one of claims 1 to 3, wherein, Using the above contact dynamic rigidity data, The dynamic rigidity data (Cw, Kw) of the workpiece support device when the workpiece is supported by the workpiece support device constituting the above-described processing apparatus, and The tool support dynamic rigidity data (Ct, Kt) of the tool support device when the tool is supported by the tool support device constituting the above-mentioned processing apparatus. To infer at least one of the following during processing: the state of the workpiece or tool, the shape of the workpiece, the shape of the tool, and the mechanical state of the processing device. The aforementioned dynamic stiffness data of the workpiece support is defined by the attenuation coefficient Cw and the spring constant Kw. The attenuation coefficient Cw represents the relationship between the relative velocity of the workpiece relative to the reference position of the workpiece support device and the external force acting on the workpiece. The spring constant Kw represents the relationship between the relative position of the workpiece relative to the reference position of the workpiece support device and the external force acting on the workpiece. The dynamic stiffness data of the tool support mentioned above are defined by the attenuation coefficient Ct and the spring constant Kt. The attenuation coefficient Ct is a value representing the relationship between the relative velocity of the tool with respect to the reference position of the tool support device and the external force acting on the tool. The spring constant Kt is a value representing the relationship between the relative position of the tool with respect to the reference position of the tool support device body and the external force acting on the tool.

7. The processing inference apparatus (3, 303) according to claim 6, wherein, The aforementioned contact dynamic rigidity data, the aforementioned workpiece support dynamic rigidity data, and the aforementioned tool support dynamic rigidity data are data that are determined individually.

8. The processing inference apparatus (5, 303) according to claim 6, wherein, The aforementioned contact dynamic rigidity data, the aforementioned workpiece support dynamic rigidity data, and the aforementioned tool support dynamic rigidity data are determined to be aggregated into a single total dynamic rigidity data.

9. The processing inference apparatus (3, 5) according to claim 6, wherein, The aforementioned processing apparatus (2) is a cylindrical grinding machine that uses a grinding wheel as the aforementioned tool to grind the outer circumferential surface of the cylindrical workpiece. The workpiece support device consists of a spindle device (30) that supports one axial end of the workpiece and drives its rotation, and a tailstock center (41) that supports the other axial end of the workpiece.

10. The processing inference apparatus (3, 5) according to claim 9, wherein, The aforementioned tailstock center is configured to control the pressing force relative to the workpiece along the axial direction of the workpiece. The aforementioned dynamic rigidity data of the workpiece support is data that changes along with the change in the contact state between the tailstock center and the workpiece. The contact state between the tailstock center and the workpiece changes due to the change in the pressing force of the tailstock center.

Images