Cutting force coefficient prediction device, cutting force coefficient prediction method, and cutting force coefficient prediction program

Abstract

The invention provides a cutting force coefficient prediction device, a cutting force coefficient prediction method, and a cutting force coefficient prediction program, which can easily and appropriately estimate a cutting force coefficient. In a conversion computer (10), a storage resource (14) stores a plurality of correspondence relationships between values of a plurality of parameters of a tool related to a cutting process and a cutting force coefficient of the tool in the cutting process using the values of the plurality of parameters, and a CPU (11) generates a plurality of decision tree models for estimating the cutting force coefficient on the basis of the plurality of correspondence relationships. Based on an error between a cutting force coefficient estimated by using a plurality of decision tree models and a correct cutting force coefficient, a use parameter, which is a parameter to be used, is determined from a plurality of parameters, and a use decision tree model is generated based on the use parameter. Specified parameter values, which are values of a plurality of types of parameters relating to a cutting process to which a cutting force coefficient is to be estimated, are received, and the cutting force coefficient at the specified parameter values is estimated using a decision tree model.

Description

Technical Field

[0001] The present invention relates to a technique for estimating a cutting force coefficient of a tool used in a machining center performing a cutting process. Background Art

[0002] In recent years, there has been a case where a workpiece (hereinafter sometimes referred to as a workpiece) is machined by inputting an NC program into an NC cutting machine.

[0003] For example, in machining using a rotating tool, machining errors may occur due to deflection of the rotating tool, and a technique is known for correcting the tool's movement path in the NC program to offset the machining errors in advance.

[0004] For example, Patent Document 1 discloses a technique for predicting machining errors of a workpiece caused by the action of cutting forces on a rotating tool. In Patent Document 1, a cutting force coefficient used to estimate the cutting forces is input in advance.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2016-218641 Summary of the Invention

[0008] Problems to be solved by the invention

[0009] For example, the technology of Patent Document 1 requires the knowledge of the cutting force coefficient in advance in order to estimate the cutting force. To know the cutting force coefficient, for example, it is necessary to actually perform cutting using a rotating tool, measure various values, and perform calculations.

[0010] As described above, in order to know the cutting force coefficient, it is necessary to actually perform cutting and measure various values, so the measurement takes a long time. In addition, there is a problem that it is necessary to measure various values ​​and to have a measuring device required for the measurement.

[0011] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technique capable of easily and appropriately estimating a cutting force coefficient.

[0012] Means for solving problems

[0013] A cutting force coefficient estimation device according to one viewpoint is a cutting force coefficient prediction device, which includes a processor and a storage unit, and is capable of estimating the cutting force coefficient of a tool used in a processing machine performing a cutting process, wherein the storage unit stores values ​​of multiple parameters of the tool related to the cutting process and multiple correspondences with the cutting force coefficient of the tool in the cutting process using the values ​​of the multiple parameters, the processor changes the combination of some parameters of the multiple parameters based on the multiple correspondences, generates multiple decision tree models for estimating the cutting force coefficient, determines the parameters to be used, i.e., the usage parameters, from multiple parameters based on the errors between the cutting force coefficients estimated using the multiple decision tree models and the correct cutting force coefficients, generates a decision tree model used to estimate the cutting force coefficient based on the usage parameters, i.e., the usage decision tree model, receives values ​​of multiple parameters related to the cutting process as the object for estimating the cutting force coefficient, i.e., designated parameter values, and uses the usage decision tree model to estimate the cutting force coefficient when the designated parameter values ​​are set.

[0014] Effects of the Invention

[0015] According to the present invention, the cutting force coefficient can be easily and appropriately estimated. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a diagram illustrating side processing.

[0017] Figure 2 This is a diagram illustrating groove processing.

[0018] Figure 3 This is a diagram illustrating areas of a tool used for side machining and groove machining.

[0019] Figure 4 It is an overall structural diagram of a processing system according to an embodiment.

[0020] Figure 5 This is a structural diagram of a conversion computer according to one embodiment.

[0021] Figure 6 is a diagram illustrating the structure of cutting force coefficient information according to an embodiment.

[0022] Figure 7 This is a diagram showing an example of a cutting input screen according to one embodiment.

[0023] Figure 8 1 is a diagram illustrating the axial cutting depth and the radial cutting depth according to one embodiment.

[0024] Figure 9 is a flowchart of a path correction process according to one embodiment.

[0025] Figure 10This is a flowchart of a cutting force coefficient calculation process according to one embodiment.

[0026] Figure 11 This is a structural diagram of an example of a decision tree model according to one embodiment.

[0027] Figure 12 This is a diagram illustrating feature quantity selection according to one embodiment. DETAILED DESCRIPTION

[0028] The embodiments will be described with reference to the accompanying drawings. The embodiments described below do not limit the scope of the invention, and not all elements and combinations thereof described in the embodiments are essential to the solution provided by the invention.

[0029] An outline of one embodiment will be described. First, side surface processing using the tool TL will be described.

[0030] Figure 1 Here, the X-axis indicates the moving direction of the tool TL, the Z-axis indicates the rotation axis direction of the tool TL, and the Y-axis indicates the direction perpendicular to the X-axis on a plane perpendicular to the rotation axis.

[0031] Side processing such as Figure 1 As shown, the side surface of a workpiece (workpiece) W is cut by the tool TL by rotating the tool TL and moving it in the X-axis direction.

[0032] The cutting force in side machining can be expressed as shown in the following formula (1).

[0033] [Formula 1]

[0034]

[0035] Here, if Figure 1 As shown, Ft is the cutting force in the tool's tangential direction, Fr is the cutting force in the tool's radial direction, and Fa is the cutting force in the tool's axial direction. Ktc, Krc, and Kac are the cutting force coefficients for the cutting force generated by shearing of the workpiece. Ktc is the cutting force coefficient in the tool's tangential direction, Krc is the cutting force coefficient in the tool's radial direction, and Kac is the cutting force coefficient in the tool's axial direction. fz is the feed per edge (feed per edge), φ is the tool's rotation angle, and dz represents the thickness of the tool's axial micro-element. Kte, Kre, and Kae are the cutting force coefficients generated by friction between the tool and the workpiece. Kte is the cutting force coefficient in the tool's tangential direction, Kre is the cutting force coefficient in the tool's radial direction, and Kae is the cutting force coefficient in the tool's axial direction.

[0036] Next, groove processing using the tool TL will be described.

[0037] Figure 2 This is a diagram illustrating groove processing.

[0038] Groove processing such as Figure 2 As shown, the tool TL is rotated and moved in the X-axis direction to cut the workpiece W with the tool TL to form a groove.

[0039] The cutting force during groove machining can also be expressed as shown in formula (1). In addition, when distinguishing between the cutting force coefficients for side machining and groove machining, after the symbol of each cutting force coefficient (Kte, etc.), if it is side machining, _S is added, for example, Kte_S, and if it is groove machining, _G is added, for example, Kte_G.

[0040] Here, the difference between side surface processing and groove processing is explained.

[0041] Figure 3 This is a diagram illustrating areas of a tool used for side machining and groove machining.

[0042] like Figure 3 As shown, the area of ​​the tool TL that is frequently used during side machining is the area on the positive side of the tool TL's Y axis, while the area frequently used during groove machining is the area on the negative side of the tool TL's Z axis. Thus, the areas frequently used differ between side machining and groove machining, resulting in different cutting force coefficients for side machining and groove machining.

[0043] System Structure

[0044] Figure 4 It is an overall structural diagram of a processing system according to an embodiment.

[0045] The processing system 1 includes a conversion computer 10, which is an example of a cutting force coefficient prediction device, a plurality of NC cutting machines 20 (an example of a processing machine), and a plurality of on-site computers 30. The conversion computer 10, the plurality of NC cutting machines 20, and the plurality of on-site computers 30 are connected via a network 40. The network 40 can be a wired network or a wireless network. In this embodiment, the NC cutting machines 20 and the on-site computers 30 are respectively arranged at locations A and B, and the conversion computer 10 is arranged at location C. In addition, the conversion computer 10 can also be arranged at either location A or location B. In addition, the plurality of NC cutting machines 20 and the plurality of on-site computers 30 can also be arranged at the same location.

[0046] The conversion computer 10 performs processing to convert an NC program for one NC cutting machine 20 (the conversion source NC program: the pre-corrected NC program) into an NC program for another NC cutting machine 20 (the conversion target NC program: the corrected NC program). Details of the conversion computer 10 will be described later.

[0047] The on-site computer 30 is a computer operated by an on-site operator, and is composed of, for example, a PC (Personal Computer) having a processor, storage resources, etc. Figure 4 A typical example is a place (e.g., inside a factory, a building, a floor, etc.) where the NC cutting machine 20 is installed. However, the on-site computer 30 can also be used in places other than the place where the NC cutting machine 20 is installed as long as it is used to convert the screen display of the computer 10.

[0048] In addition, in the following description, the on-site computer 30 is responsible for the download processing and screen display of the converted NC program, the screen display of the conversion input screen, etc., and the actual conversion processing is handled by the conversion computer 10. However, although the convenience is somewhat reduced, the functions (including some functions) that each computer is responsible for can be exchanged or merged with each other. In addition, the conversion computer 10 can also be composed of multiple computers. Therefore, in the following description, the term "conversion system" is sometimes used. The system includes one or more computers (on-site computers 30 or conversion computers 10), and is a system that performs the processing described below by the conversion computer 10 and the on-site computer 30. In addition, part of the processing implemented by the on-site computer 30 can also be omitted.

[0049] The NC cutting machine 20 is, for example, a machining center, having a main body 22 that performs machining (e.g., cutting), an NC controller 21 that controls the machining of the main body 22, and a tool library 25 that is an example of a storage unit for tools TL that can store one or more tool groups used in the main body 22.

[0050] The tool magazine 25 has a plurality of slots (SL: 25 a , 25 b , 25 c ) each capable of accommodating one tool TL.

[0051] The NC controller 21 controls the machining process of the main body 22 and the tool replacement process according to the NC program stored therein.

[0052] The main body 22 includes a processing head 23, a workbench 24 and a tool changing unit 26. The processing head 23 has a spindle that can be mounted with a tool TL and can rotate. In addition, the processing head 23 can also be the spindle itself. The workbench 24 can carry and move the workpiece W that is the object of processing. The tool changing unit 26 removes the tool TL from the processing head 23 and stores it in an empty slot of the tool magazine 25. In addition, the tool changing unit 26 takes the tool TL out of the slot of the tool magazine 25 and installs it on the processing head 23. An example of the tool changing unit 26 is the changing arm (also called the ATC arm) of the automatic tool changing device (ATC). In addition, the above-mentioned tool magazine 25 is also a component of the automatic tool changing device. The NC program can internally record a series of commands (called codes in the terminology of the NC program, or words with parameters added to the codes) that represent tool changing commands, and the tool changing command includes a slot number indicating the position of a slot (the meaning of which will be described later) in the tool magazine 25. The tool changing unit 26 takes out the tool TL from the slot specified by the slot number included in the parameter of the tool changing command and attaches it to the processing head 23 in accordance with the instruction of the NC controller 21 that has read the tool changing command.

[0053] In the NC cutting machine 20 , the number of tools TL that can be stored in the tool magazine 25 is limited. By preparing one or more tool sets 50 in advance and replacing the tool sets stored in the tool magazine 25 according to the machining process to be performed, various machining processes can be handled.

[0054] In this embodiment, the tool TL includes a tool part TLa such as an end mill, drill, or turning tool for cutting the workpiece W, and a bracket TLb for mounting the tool part TLa on the processing head 23. However, for example, if the tool part TLa can be directly mounted on the processing head 23, the bracket TLb may not be included, and at least the tool part TLa may be included.

[0055] In the following description, sometimes a thing including at least a processing machine that uses an NC program to be transformed (i.e., a source NC program for transformation) for processing and a tool set corresponding to the processing machine is referred to as a "source environment". In addition, sometimes a thing including at least a processing machine that is scheduled to use a transformed NC program (i.e., a target NC program for transformation) for processing and a tool set corresponding to the processing machine is referred to as a "target environment". In addition, the source environment and the target environment may include physical or logical things contained in each location (such as the temperature of the location, or a temperature sensor, humidity, a humidity sensor, or the floor where the processing machine is installed in the location, or a building constituting the location). In addition, the "tool set corresponding to the processing machine" includes not only the tool set already stored in the tool library of the processing machine, but also the tool set that has the possibility of being stored in the tool library and used in the future. Typically, the tool set corresponding to the processing machine is located in the same location as the processing machine.

[0056] Next, the conversion computer 10 will be described in detail.

[0057] Figure 5 This is a structural diagram of a conversion computer according to one embodiment.

[0058] <<Hardware>>

[0059] The conversion computer 10 is, for example, a personal computer or a general-purpose computer. It includes a CPU 11, a network interface 12 (abbreviated as "Net I / F" in the figure), a user interface 13 (abbreviated as "User I / F" in the figure), a storage resource 14 as an example of a storage unit, and an internal network connecting these components.

[0060] The CPU 11 is an example of a processor that can execute programs stored in the storage resource 14. As a processor, a GPU (Graphics Processing Unit) can also be used, and other semiconductor devices can also be used as long as they are the main body that executes the specified processing. The storage resource 14 stores the program that is the execution target of the CPU 11, various information used in the program, the NC program used by the NC cutting machine 20, etc. The storage resource 14 can be, for example, a semiconductor memory, a flash memory, an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc., and can be a volatile memory or a non-volatile memory.

[0061] The network interface 12 is an interface for communicating with external devices (for example, the field computer 30 , the NC controller 21 of the NC cutting machine 20 , etc.) via the network 40 .

[0062] The user interface 13 is, for example, a touch panel, a display, a keyboard, a mouse, etc., and may be any other device as long as it can accept an operation by an operator (user) and display information. The user interface 13 may be composed of a plurality of devices.

[0063] <<Data, etc.>>

[0064] The storage resource 14 stores a conversion program 141, which is an example of a cutting force coefficient prediction program; a structure information acquisition program 142; machine structure information 143; tool set information 144; individual tool information 145; a conversion source NC program 146; a conversion target NC program 147; conversion history information 148; and cutting force coefficient information 149. The storage resource 14 may also store other information. Details of each data and program will be described in the next paragraph. Furthermore, individual information or portions of each information item may be omitted.

[0065] * Processing machine configuration information 143. The processing machine configuration information 143 is configured as, for example, a table storing information related to each NC cutting machine 20. The processing machine configuration information 143 includes the following information for each NC cutting machine 20.

[0066] (a1) Identifier (machine ID) of the NC cutting machine 20 : As the machine ID, the identifier of the NC controller 21 or the network address of the NC controller 21 may be used instead.

[0067] (a2) Model of the NC cutting machine 20 .

[0068] (a3) Installation location of the NC cutting machine 20.

[0069] (a4) Actual usage status of the NC cutting machine 20, such as usage time.

[0070] (a5) Temperature of a predetermined portion of the NC cutting machine 20. The predetermined portion may be the spindle of the NC cutting machine 20 or the table 24.

[0071] (a6) Information on the rigidity of a predetermined portion of the NC cutting machine 20 (e.g., Young's modulus, deflection, etc.) The predetermined portion may be the spindle of the processing head 23 or the table 24 of the NC cutting machine 20 .

[0072] (a7) Shape of a Predetermined Part of the NC Cutting Machine 20. The shape of the predetermined part may be the length of the spindle of the NC cutting machine 20 or the length of the table 24.

[0073] (a8) The maximum number of tools that can be stored in the tool magazine 25, that is, the number of slots.

[0074] (a9) Offset value set to accommodate changes over time and the installation environment. This offset value is used to make minor corrections to the coordinates of the tool movement in the NC program, for example, to correct for subtle tilts of the worktable due to aging.

[0075] (a10) Manufacturer, model, etc. of the NC controller 21. The description format of the NC program may differ slightly depending on the manufacturer and model of the NC controller 21. This information is used to determine such a situation.

[0076] (a11) The looseness of the spindle, worktable and other components, the movement accuracy (such as the backlash of the worktable), straightness, flatness, parallel movement, amplitude and vibration frequency during device operation.

[0077] In this embodiment, information regarding (a1), (a2), (a4), (a5), (a8), (a9), and (a10) is obtained, for example, from the NC controller 21 of the NC cutting machine 20, while information regarding (a3), (a6), (a7), and (a11) is obtained from information input by the operator. The method of obtaining information is not limited to this. At least a portion of (a1), (a2), (a4), (a5), (a8), (a9), and (a10) can be obtained from information input by the operator via the user interface 13, and information regarding (a3), (a6), (a7), and (a11) that can be obtained from the NC controller 21 can also be obtained from the NC controller 21. Furthermore, information obtained from the NC controller 21 can also be obtained from an alternative device (e.g., another computer or the sensor itself).

[0078] * Tool information (tool group information 144 and individual tool information 145)

[0079] The tool group information 144 is information for managing a group (set) composed of one or more tools TL. The tool group information 144 is a set of identification information (tool group ID) of the tool group and identifiers or model numbers of one or more tools TL constituting the group.

[0080] The individual tool information 145 is information about each tool and includes the following information.

[0081] (b1) Identifier of Tool TL (Tool ID: e.g., serial number). The identifier of Tool TL may be the value of an individual ID assigned to the cutter T1a or holder TLb, or may be automatically assigned by the CPU 11 executing the configuration information acquisition program 142 if no individual ID is assigned.

[0082] (b2) Model of tool TL (an example of tool identification information). For example, the model numbers of the tool unit TLa and the holder TLb that make up the tool TL. If the tool TL consists solely of the tool unit TLa, the model number of the tool unit TLa alone may be used. If the tool unit TLa consists of multiple components, the model numbers of all or a portion of the components may be used.

[0083] (b3) Material, shape, rigidity (Young's modulus, deflection, etc.), usage history, temperature, etc. of the tool TL (e.g., the tool portion TLa and the holder TLb). Since rigidity varies depending on the material and shape of the tool TL, this information also relates to rigidity. Furthermore, unless otherwise specified, "shape" includes, in addition to the three-dimensional shape and cross-sectional shape generally shown in the drawings and CAD data, representative values ​​derived from the shape, such as length, the length of the tool portion TLa protruding from the holder TLb (tool extension length), the thickness of the tool portion TLa, and the straightness of the tool portion TLa.

[0084] (b4) Information (position information, slot number) on the arrangement position (slot) of the tool magazine 25 where the tools are to be stored.

[0085] In the present embodiment, information regarding (b1) to (b4) is obtained from information input by an operator via the user interface 13 , for example. However, information that can be obtained from the NC controller 21 may also be obtained from the NC controller 21 .

[0086] *The source NC program 146 is an NC program used for processing in the source NC cutting machine 20 (referred to as the source NC cutting machine 20). The source NC program 146 may be tuned according to the characteristics and status of the source NC cutting machine 20 to maintain a predetermined precision in the processing accuracy of the target object obtained by the processing performed by the source NC cutting machine 20.

[0087] *The conversion target NC program 147 is an NC program obtained by converting the conversion source NC program 146 to match the conversion target NC cutting machine 20 (referred to as the conversion target NC cutting machine 20). If no conversion processing has been performed on any of the conversion source NC programs 146, no conversion target NC program 147 exists.

[0088] * The conversion history information 148 is information for managing the conversion process history when converting the conversion source NC program 146 into the conversion target NC program 147. The conversion history information 148 is information that associates, for example, identification information for identifying the conversion process with various information used in the conversion process (input information, etc.).

[0089] Figure 6 is a diagram illustrating the structure of cutting force coefficient information according to an embodiment.

[0090] *Cutting force coefficient information 149 includes multiple entries indicating the correspondence between the values ​​of various parameters (feature quantities) related to the cutting process (cutting process) performed by the tool and the tool's cutting force coefficient. Entries of cutting force coefficient information 149 include tool ID 149a, tool type 149b, tool diameter 149c, number of flutes 149d, helix angle 149e, flute length 149f, coating 149g, overhang length 149h, shank diameter 149i, cutting force coefficient for side machining 149j, and cutting force coefficient for groove machining 149k.

[0091] In the tool ID 149a, the identifier of the tool TL corresponding to the entry (tool ID: for example, a serial number, etc.) is stored. In the tool type 149b, the type of the tool TL corresponding to the entry (tool type) is stored. As tool types, there are, for example, flat-end cutters, ball-end cutters, and rounded corner cutters. In the tool diameter 149c, the diameter of the tool TL is stored. In the number of blades 149d, the number of blades of the tool TL is stored. In the helix angle 149e, the helix angle representing the inclination angle of the peripheral blade when the axial direction of the tool TL corresponding to the entry is used as a reference is stored. In the blade length 149f, the blade length of the tool TL corresponding to the entry is stored. In the coating 149g, whether the tool TL corresponding to the entry has a coating is stored. In the present embodiment, "1" is stored when the tool TL has a coating, and "0" is stored when it does not have a coating. In the overhang length 149h, the overhang length of the tool TL corresponding to the entry is stored. In the shank diameter 149i, the shank diameter of the tool TL corresponding to the entry is stored.

[0092] The cutting force coefficient for side machining 149j stores the cutting force coefficient for side machining of the tool corresponding to the entry (first cutting force coefficient). Here, the six cutting force coefficients corresponding to side machining are Ktc_S, Krc_S, Kac_S, Kte_S, Kre_S, and Kae_S. The cutting force coefficient for groove machining 149k stores the cutting force coefficient for groove machining of the tool corresponding to the entry (second cutting force coefficient). Here, the six cutting force coefficients corresponding to groove machining are Ktc_G, Krc_G, Kac_G, Kte_G, Kre_G, and Kae_G.

[0093] In addition, the following information may be stored in the storage resource 14 .

[0094] * Workpiece W information. This information includes, for example, the workpiece W's pre-machining shape data, material, rigidity, and target machining shape data. The target machining shape data represents the target shape when machining using the NC program. It indicates that the error will be zero when machining the workpiece W to the target shape.

[0095] * Information on the pre-conversion environment or the conversion destination environment other than the processing machine configuration information 143, tool group information 144, and individual tool information 145. To clarify, this information may be referred to as "other pre-conversion environment information" or "other pre-conversion environment information."

[0096] <Programs running on the conversion computer>

[0097] <<Conversion Program 141>>

[0098] The conversion program 141 performs the following processing when executed by the CPU 11. Here, the CPU 11 executes the conversion program 141, thereby constituting a conversion unit.

[0099] * When the conversion start instruction is given via the conversion input screen, the conversion program 141 reflects the various information inputted via the conversion input screen to the machining machine configuration information 143, tool group information 144, and individual tool information 145. Figure 7 ) receives parameter values ​​and conditions related to the cutting process. Furthermore, based on the various information inputted into the conversion input screen and the cutting input screen 200, as well as the information on the conversion target environment or the conversion source environment included in the processing machine configuration information 143, the tool group information 144, and the individual tool information 145, the conversion program 141 executes a conversion process to convert the conversion source NC program 146, which is the target of the conversion, into the conversion target NC program 147. The resulting conversion target NC program 147 is then stored in the storage resource 14.

[0100] During the conversion process of converting the source NC program 146 into the target NC program 147, the conversion program 141 generates data containing commands that modify or add to the source NC program 146, based on information about the rigidity of the target NC cutting machine 20 or the rigidity of the tool TL of the tool set 50 used in the target NC cutting machine 20, as the target NC program 147. Furthermore, by using tool diameter correction, tool length correction, tool wear correction, feed rate, or cutting speed as the added or modified commands, significant changes in the machining operation, such as an increase in the number of times the tool TL processes the workpiece W, can be avoided. However, commands that increase the number of times the workpiece W is processed (e.g., commands corresponding to trial cutting) can also be added.

[0101] For example, in the conversion process of converting the conversion source NC program 146 into the conversion target NC program 147, the conversion program 141 executes a cutting force coefficient calculation process (refer to Figure 10 ).

[0102] Furthermore, when at least a portion of the description format of the NC program differs between the NC controller 21 of the source NC cutting machine 20 and the NC controller 21 of the destination NC cutting machine 20, the conversion program 141 converts the portion of the description of the source NC program that differs in the description format to the description format of the NC controller 21 of the destination NC cutting machine 20. This allows smooth processing in the NC controller 21 of the destination NC cutting machine 20.

[0103] *After the conversion process, the conversion program 141 displays a download confirmation screen for confirming the download. When the download instruction is given, the conversion target NC program 147 is sent to the NC controller 21 of the conversion target NC cutting machine 20 or the on-site computer 30 at the location where the conversion target NC cutting machine 20 is located.

[0104] <<Structure information acquisition program 142>>

[0105] The configuration information acquisition program 142 is executed by the CPU 11 to perform the following processing.

[0106] * The configuration information acquisition program 142 acquires various information about the NC cutting machine 20 from the NC controller 21. The acquired information includes the above-mentioned information (a1), (a2), (a4), (a5), (a8), (a9), and (a10).

[0107] *The structural information acquisition program 142 causes the user interface 13 to display a transformation input screen, and obtains various information from the operator through the transformation input screen (information about the NC cutting machine 20 obtained from the operator ((a3), (a6), (a7) and (a11)), and information about the tool group 50 ((b1) to (b4)).

[0108] The structural information acquisition program 142 causes the user interface 13 to display the cutting input screen 200 , and acquires various information from the operator via the cutting input screen 200 .

[0109] Figure 7 This is a diagram showing an example of a cutting input screen according to one embodiment. Figure 8 1 is a diagram illustrating the axial cutting depth and the radial cutting depth according to one embodiment.

[0110] The cutting input screen 200 includes a tool parameter input area 201 and a cutting condition input area 202. The tool parameter input area 201 is an area for accepting input of various cutting parameters (specified parameters), such as values ​​(specified parameter values) related to tool shape. Cutting parameters are managed using the cutting force coefficient information 149 and include, for example, tool type, helix angle, tool diameter, coating presence, number of cutting edges, and workpiece material.

[0111] The processing condition input area 202 is an area for receiving the input of the processing conditions for the workpiece w in the cutting process. The processing conditions include the axial cutting depth and the radial cutting depth in the cutting process. The axial cutting depth is as follows: Figure 8 As shown, it represents the axial width (Z-axis direction) of the portion of the workpiece w cut by the tool TL, and the radial cutting depth is as shown in FIG. Figure 8 denoted by , represents the width in the radial direction (Y-axis direction) of the portion of the workpiece w cut by the tool TL.

[0112] Next, the processing operations performed by the conversion computer 10 will be described.

[0113] (Processing 1) The configuration information acquisition program 142 (strictly speaking, the CPU 11 executing the configuration information acquisition program 142) acquires various available information (e.g., (a1), (a2), (a4), (a5), (a8), (a9), and (a10)) about each NC cutting machine 20 from the NC controller 21 of each NC cutting machine 20 connected via the network 40. Note that this processing does not need to be performed every time the processing described below, including processing 2 and subsequent processing, is performed.

[0114] (Process 2) Next, the structure information acquisition program 142 displays the conversion input screen and accepts the following designations via the conversion input screen.

[0115] * The conversion target, ie, the conversion source, is specified by the NC program 146.

[0116] *Specify the information (machine ID) of the NC cutting machine 20 (conversion source NC cutting machine) that has processed the workpiece W using the conversion source NC program 146 .

[0117] * Specifying information (tool set ID) of a tool set used in a machining process performed using the conversion source NC program 146.

[0118] * The information (machine ID) of the NC cutting machine (the conversion destination NC cutting machine 20 ) that newly performs cutting of the workpiece W using the conversion destination NC program 147 obtained by converting the conversion source NC program 146 is determined.

[0119] * Specifying the information (tool set ID) of the tool set used in the conversion destination NC cutting machine 20.

[0120] At the same time, the structural information acquisition program 142 accepts input (direct input or selective input) of various information about the transformation source NC cutting machine 20 and the transformation target NC cutting machine 20 ((a3), (a6), (a7) and (a11)), and information about the tool group 50 used in the transformation source NC cutting machine 20 and the tool group 50 used in the transformation target NC cutting machine 20 (information (b1) to (b4)).

[0121] In addition, the structural information acquisition program 142 displays a cutting input screen 200, and accepts cutting processing parameters such as tool type, helix angle, tool diameter, whether it has coating, number of blades, material of the workpiece, and axial cutting depth and radial cutting depth through the cutting input screen 200.

[0122] (Process 3) Upon receiving a conversion start instruction from the user, the structural information acquisition program 142 sends a conversion start instruction to the conversion program 141. Here, the conversion start instruction includes various information input (direct input or selective input) into the conversion input screen and various information input into the cutting input screen 200.

[0123] (Processing 4) When the transformation program 141 receives a transformation start instruction, it reads the specified transformation source NC program 146 (NC program before correction), and based on the information contained in the transformation start instruction (at least information about the rigidity of the transformation target NC cutting machine 20 or the tool group used in the transformation target NC cutting machine 20), transforms the transformation source NC program 146 into the transformation target NC program 147 (NC program after correction), and saves the transformed transformation target NC program 147 in the storage resource 14.

[0124] (Process 5) Next, the conversion program 141 displays a download confirmation screen. Alternatively, instead of automatically displaying the download confirmation screen after completing Process 4, the download confirmation screen may be displayed in response to an operation performed by the user of the on-site computer 30. Upon receiving the download instruction, the conversion program 141 transmits the conversion target NC program 147 to the NC controller 21 of the conversion target NC cutting machine 20 or to the on-site computer 30 at the location where the conversion target NC cutting machine 20 is located.

[0125] For example, when the target NC program 147 is sent to the NC controller 21, the NC controller 21 can store the received target NC program 147 and execute it during subsequent processing. On the other hand, when the target NC program 147 is sent to the on-site computer 30, the on-site computer 30 stores it. Subsequently, by storing the target NC program 147 from the on-site computer 30 in the NC controller 21 via the network 40 or a recording medium, the NC controller 21 can execute the target NC program 147.

[0126] <Specific example of conversion processing performed by the conversion program>

[0127] Next, a specific example of the processing operation performed by the conversion computer 10 will be described.

[0128] Figure 9 This is a flowchart of a path correction process according to one embodiment. The path correction process is performed during the conversion process.

[0129] First, the transformation program 141 receives all program segments (blocks), material shapes, and tool shape information of the transformation source NC program 146 for the processing object (S1). Here, the program segment refers to the description part of the processing performed by the transformation source NC program 146, which includes one command (address) that can be indicated to the NC cutting machine 20. The program segment includes more than one command (address) that can be indicated at the same time. As an address, for example, a code indicating the type of command and parameters about the content of the command are included. In addition, if the capacity of the transformation source NC program 146 is large and all program blocks cannot be called into the working area of ​​the memory, the program segment to be read can be switched according to the progress of the processing.

[0130] Next, the conversion program 141 performs the following processing (steps S2 to S5 ) on each of the blocks that perform cutting processing on the workpiece W (referred to as a target block) among the read blocks.

[0131] The conversion program 141 performs a cutting force coefficient calculation process for estimating the cutting force coefficient of the tool TL in the cutting process for the target program segment (refer to Figure 10 )(S2).

[0132] Next, the conversion program 141 calculates the cutting force applied to the tool TL during the cutting process performed by the target program segment (S3). Specifically, the conversion program 141 calculates the cutting force applied to the tool TL based on the shape of the workpiece W, the shape of the tool, the cutting force coefficient, and the shape of the machined portion.

[0133] Next, the conversion program 141 determines a correction amount for the path (movement trajectory) of the tool TL based on the cutting force (S4). Here, the correction amount is, for example, the amount of deflection generated at the tip of the tool TL.

[0134] Next, the conversion program 141 updates each target block to a block reflecting the determined correction amount, and stores the conversion target NC program 147 including the updated block in the memory of the storage resource 14 ( S5 ).

[0135] By this path correction processing, it is possible to generate the conversion target NC program 147 in which the program segments of the cutting process are corrected to an appropriate correction amount.

[0136] Next, the cutting force coefficient calculation process in step S2 will be described.

[0137] Figure 10 This is a flowchart showing a cutting force coefficient calculation process according to one embodiment.

[0138] The conversion program 141 accesses and obtains the cutting force coefficient information 149 ( S11 ).

[0139] Next, the conversion program 141 executes the processing of steps S12 to S17 for each of the six cutting force coefficients used in side machining (Ktc_S, Krc_S, Kac_S, Kte_S, Kre_S, Kae_S) and the six cutting force coefficients used in groove machining (Ktc_G, Krc_G, Kac_G, Kte_G, Kre_G, Kae_G). The cutting force coefficients used as targets are referred to herein as target cutting force coefficients.

[0140] In step S12 , the conversion program 141 stores the number of parameters (feature quantities) of the cutting force coefficient information 149 (the number of feature quantities) in the variable i_num, and sets 0 to the variable i ( S12 ).

[0141] Next, the conversion program 141 generates a decision tree model for estimating the target cutting force coefficient based on the cutting force coefficient information and a combination of multiple feature quantities other than feature quantity i (the i-th feature quantity). The method of generating a decision tree model based on multiple feature quantities can be implemented using known techniques.

[0142] Figure 11 This is a structural diagram of an example of a decision tree model according to one embodiment.

[0143] The object cutting force coefficient is the decision tree model of KTC, for example Figure 11 The decision tree model shown in FIG. Based on this decision tree model, KTC can be estimated according to the values ​​of multiple feature quantities such as tool diameter, helix angle, number of cutting edges, and presence or absence of coating.

[0144] return Figure 10 As described above, the conversion program 141 uses the generated decision tree model to estimate the target cutting force coefficient for predetermined test data (a set of values ​​of multiple parameters having the correct cutting force coefficient), thereby calculating a value of the error between the target cutting force coefficient estimated by the decision tree model for feature quantity i (for the case where feature quantity i is removed) and the correct cutting force coefficient (for example, an average accuracy reduction rate indicating a decrease in accuracy when estimating based on feature quantities excluding feature quantity i). Here, the average accuracy reduction rate can be calculated using a known technique.

[0145] Next, the conversion program 141 determines whether i = i_num (S15). If i = i_num (S15: No), this indicates that decision tree models have not been generated for all combinations of multiple feature quantities excluding each feature quantity. Therefore, the conversion program 141 proceeds to step S13 and performs the same process.

[0146] On the other hand, when i = i_num (S15: Yes), this indicates that decision tree models have been generated for all combinations of multiple feature quantities excluding each feature quantity. Therefore, the conversion program 141 determines the feature quantity (usage feature quantity: usage parameter) used to generate the decision tree model from all feature quantities based on the average accuracy reduction rate for each feature quantity, and generates a decision tree model (usage decision tree model) based on the usage feature quantity of the cutting force coefficient information 149 (S17). Here, the usage feature quantity may be, for example, a predetermined number of feature quantities with the highest average accuracy reduction rate among the multiple feature quantities, or may be all feature quantities with an average accuracy reduction rate above a predetermined threshold. This makes it possible to generate a decision tree model with high prediction accuracy by eliminating feature quantities that have little or no effect on the estimation of the cutting force coefficient.

[0147] Figure 12 This is a diagram illustrating selection of feature quantities according to one embodiment. Figure 12 In the figure, the horizontal axis represents the amount of features removed, and the vertical axis represents the corresponding average accuracy reduction rate.

[0148] The average accuracy reduction rate of the feature quantity i calculated in step S14 (the average accuracy reduction rate when the feature quantity i is removed) can be expressed as follows, for example: Figure 12 In this case, for example, the top-ranked feature quantity with a larger average accuracy reduction rate can be expressed as follows. Figure 12 In the example of , for example, three feature quantities, feature quantity 0, feature quantity 1, and feature quantity 2, are determined as the feature quantities to be used.

[0149] return Figure 10As described above, by executing the processing of steps S12 to S17 for each of the six cutting force coefficients in side machining and the six cutting force coefficients in groove machining, a usage decision tree model for estimating each cutting force coefficient is generated.

[0150] Next, the conversion program 141 estimates (calculates) six cutting force coefficients in side machining and six cutting force coefficients in groove machining using each decision tree model based on the parameters input via the cutting input screen 200 ( S18 ).

[0151] Next, the conversion program 141 calculates the cutting force coefficient (each of Ktc_R, Krc_R, Kac_R, Kte_R, Kre_R, and Kae_R) corresponding to the target cutting process based on the input ratio of the radial cutting depth to the axial cutting depth (S19). Specifically, the cutting force coefficient (each of Ktc_R, Krc_R, Kac_R, Kte_R, Kre_R, and Kae_R) is calculated according to the following formula (2).

[0152] Cutting force coefficient = (cutting force coefficient of side machining) × (axial cutting depth / (axial cutting depth + radial cutting depth)) +

[0153] (Cutting force coefficient for grooving) × (radial cutting depth / (axial cutting depth + radial cutting depth))

[0154] ……(2)

[0155] For example, Ktc_R is expressed as the following equation (3).

[0156] Ktc_R=Ktc_S×(axial cutting depth / (axial cutting depth+radial cutting depth))+

[0157] Ktc_G×(radial cutting depth / (axial cutting depth + radial cutting depth))

[0158] ……(3)

[0159] In this way, based on the ratio of the radial cutting depth to the axial cutting depth, the cutting force coefficient corresponding to the target cutting process is calculated according to the cutting force coefficient of the side processing and the cutting force coefficient of the groove processing, so the cutting force coefficient corresponding to the target cutting process can be calculated with good accuracy.

[0160] <Function / Effect>

[0161] In this manner, according to the cutting force coefficient calculation process, it is possible to generate an appropriate decision tree model based on the pre-stored cutting force coefficient information 149 and estimate the cutting force coefficient without actually using the tool TL to perform various measurements.

[0162] <Transformation>

[0163] The present invention is not limited to the above-described embodiment, and can be implemented with appropriate modifications without departing from the spirit of the present invention.

[0164] <<Other ways to use the on-site computer>>

[0165] In the above embodiment, an example is described in which the user interface 13 of the conversion computer 10 displays a conversion input screen, a download confirmation screen, and a cutting input screen to accept input. However, the present invention is not limited to this. Alternatively, any on-site computer 30 may be caused to display the conversion input screen, the download confirmation screen, and the cutting input screen and accept input. For example, the on-site computer 30 at the location of the conversion target NC cutting machine 20 may be caused to display and accept input. Alternatively, the on-site computer 30 at the location of the conversion source NC cutting machine 20 may be caused to display a portion of the conversion input screen and accept input, while the on-site computer 30 at the location of the conversion target NC cutting machine 20 may be caused to display the remaining portion of the conversion input screen and accept input.

[0166] <<Other>>

[0167] In addition, in the above embodiment, a part or all of the processing performed by the CPU 11 can be performed by hardware circuits. In addition, the program in the above embodiment can be installed from a program source. The program source can be a program distribution server or a non-volatile recording medium (e.g., a removable recording medium).

[0168] The conversion source NC program can be an NC program generated from the target shape data using a CAM program and before cutting by a machining center. Furthermore, the tool set at this time can be input with the tool data from when the CAM program was generated. Furthermore, the toolpath correction amount can be determined based on the rigidity of the workpiece W and the amount of thermal expansion during cutting (or, in other words, the amount of thermal contraction after cutting), in addition to the spindle rigidity or tool rigidity mentioned above.

[0169] In the above embodiment, two types of cutting force coefficients, namely the cutting force coefficient for side machining and the cutting force coefficient for groove machining, are stored as cutting force coefficient information. However, it is also possible to store only one type. In this case, the cutting force coefficient obtained from the generated decision tree model can be directly used as the cutting force coefficient for the target cutting process without receiving machining conditions from the operator.

[0170] Furthermore, in the above-mentioned embodiment, the calculated cutting force coefficient may be stored in a storage resource, or the calculated cutting force coefficient may be displayed and output.

[0171] Furthermore, regarding the multiple parameters related to the cutting process performed by the tool, not all of the multiple parameters included in the cutting force coefficient information 149 are required. For example, at least some of the parameters may be required, and other parameters may be added.

[0172] Furthermore, in the above embodiment, the values ​​of a plurality of parameters included in the cutting force coefficient information 149 regarding the cutting process are accepted via the cutting input screen 200. However, for example, only the values ​​of the usage parameters used in the decision tree model may be accepted.

[0173] In the above description, a machining center is mainly used as an example of a processing machine, but other processing machines may also be used as long as they can be NC controlled.

[0174] While the above description partially omits the data transmission and reception between the field computer and the conversion computer, data transmission and reception between the field computer and the conversion computer is of course performed. For example, when the conversion program 141 is executed by the conversion computer and the field computer displays a user interface, displays information related to the operation, or inputs information, the field computer executes a program responsible for a portion of the processing performed by the configuration information acquisition program in the field computer. This program then transmits the input information to the conversion computer, or receives display information sent from the conversion computer, thereby displaying the user interface.

[0175] Description of Reference Numerals

[0176] 1 processing system, 10 conversion computer, 11 CPU, 12 network interface, 13 user interface, 14 storage resource, 20 NC cutting machine, 21 NC controller, 30 on-site computer, 50 tool set, W workpiece, TL tool.

Claims

1. A cutting force coefficient prediction device comprising a processor and a storage unit, capable of estimating a cutting force coefficient of a tool used in a processing machine performing a cutting process, wherein: The storage unit stores a plurality of correspondences between values ​​of various parameters of a tool related to the cutting process and cutting force coefficients of the tool in the cutting process using the values ​​of the various parameters. The processor changes a combination of some parameters of the plurality of parameters based on the plurality of correspondences to generate a plurality of decision tree models for estimating the cutting force coefficient. Based on the error between the cutting force coefficients estimated by the plurality of decision tree models and the correct cutting force coefficients, a parameter to be used, i.e., a use parameter, is determined from a plurality of parameters. A decision tree model is generated based on the usage parameters to estimate the cutting force coefficient, that is, a usage decision tree model. Accepting the values ​​of various parameters related to the cutting process for which the cutting force coefficient is to be estimated, i.e., designated parameter values, The decision tree model is used to infer the cutting force coefficient under the specified parameter value.

2. The cutting force coefficient prediction device according to claim 1, wherein: The storage unit includes a first cutting force coefficient in side machining and a second cutting force coefficient in groove machining as the cutting force coefficients, The processor generates a plurality of decision tree models for each of the first cutting force coefficient and the second cutting force coefficient, determines the usage parameters for each of the first cutting force coefficient and the second cutting force coefficient, and generates the usage decision tree model. The cutting force coefficients in side machining and groove machining under the specified parameter values ​​are estimated.

3. The cutting force coefficient prediction device according to claim 2, wherein: The processor receives a radial cutting depth and an axial cutting depth in a cutting process as a target for estimating a cutting force coefficient, The cutting force coefficient in the cutting process is estimated based on the estimated cutting force coefficient in the groove machining and the estimated cutting force coefficient in the side machining, and the ratio of the radial cutting depth to the axial cutting depth.

4. The cutting force coefficient prediction device according to claim 1, wherein: The processor determines the use parameter from the plurality of parameters based on an average accuracy reduction rate of a decision tree model that does not use each parameter.

5. The cutting force coefficient prediction device according to claim 1, wherein: The processor may obtain multiple parameters of the tool related to the cutting process, including the type of tool, the diameter of the tool, the number of tool edges, the helix angle of the tool, whether the tool has a coating, and the material of the workpiece to be cut.

6. The cutting force coefficient prediction device according to any one of claims 1 to 5, characterized in that: The processor generates an NC program for performing the cutting process based on the estimated cutting force coefficient.

7. A cutting force coefficient prediction method of a cutting force coefficient prediction device for estimating the cutting force coefficient of a tool used in a processing machine performing a cutting process, characterized in that: The cutting force coefficient prediction device: storing a plurality of correspondences between values ​​of various parameters of a tool related to the cutting process and cutting force coefficients of the tool in the cutting process using the values ​​of the various parameters, Based on the multiple corresponding relationships, a combination of some parameters of the multiple parameters is changed to generate multiple decision tree models for estimating the cutting force coefficient. Based on the error between the cutting force coefficients estimated by the plurality of decision tree models and the correct cutting force coefficients, a parameter to be used, i.e., a use parameter, is determined from a plurality of parameters. A decision tree model is generated based on the usage parameters to estimate the cutting force coefficient, that is, a usage decision tree model. Accepting the values ​​of various parameters related to the cutting process for which the cutting force coefficient is to be estimated, i.e., designated parameter values, The decision tree model is used to infer the cutting force coefficient under the specified parameter value.

8. A cutting force coefficient prediction program for causing a computer to estimate a cutting force coefficient of a tool used in a processing machine performing a cutting process, characterized in that: Make the computer: Based on the values ​​of various parameters of the tool related to the cutting process and the various correspondences between the cutting force coefficients of the tool in the cutting process using the values ​​of the various parameters, a plurality of decision tree models for estimating the cutting force coefficients are generated by changing the combination of some of the various parameters. determining a parameter to be used, i.e., a usage parameter, from a plurality of parameters based on an error between the cutting force coefficients estimated using the plurality of decision tree models and the correct cutting force coefficient; generating a decision tree model for estimating the cutting force coefficient based on the usage parameters, that is, using a decision tree model; receiving designated parameter values, which are values ​​of various parameters related to the cutting process for which the cutting force coefficient is to be estimated; The decision tree model is used to infer the cutting force coefficient under the specified parameter value.

Images