阅读说明：本技术 机床的切屑处理装置以及切屑处理方法 (Chip disposal device and chip disposal method for machine tool ) 是由 下池昌广 于 2020-03-13 设计创作，主要内容包括：本发明提供一种机床的切屑处理装置,其具备对清洗喷嘴进行控制的清洗喷嘴控制部(30),该清洗喷嘴向机械加工时飞散的切屑喷射清洗流体而将所述切屑引导至切屑回收部,所述清洗喷嘴控制部(30)具备：位置推定部(31),通过对被加工物的加工条件进行解析来推定产生的切屑的堆积位置；以及喷嘴姿势调整部(34),朝向由所述位置推定部(31)推定出的堆积位置而对所述清洗喷嘴的姿势进行调整。(The invention provides a chip processing device of a machine tool, which is provided with a cleaning nozzle control part (30) for controlling a cleaning nozzle, wherein the cleaning nozzle sprays cleaning fluid to chips scattered in machining so as to guide the chips to a chip recovery part, and the cleaning nozzle control part (30) is provided with: a position estimation unit (31) which estimates the position of the deposit of the generated chips by analyzing the machining conditions of the workpiece; and a nozzle posture adjustment unit (34) that adjusts the posture of the cleaning nozzle toward the deposition position estimated by the position estimation unit (31).)

1. A chip disposal device for a machine tool, comprising a cleaning nozzle control unit for controlling a cleaning nozzle for injecting a cleaning fluid to chips scattered during machining and guiding the chips to a chip collection unit,

the cleaning nozzle control unit includes:

a position estimating unit that estimates a deposition position of generated chips by analyzing a machining condition of the workpiece; and

and a nozzle posture adjusting unit that adjusts the posture of the cleaning nozzle toward the deposition position estimated by the position estimating unit.

2. The chip disposal device for a machine tool according to claim 1,

the position estimation unit includes:

an FEM analysis unit that calculates a trajectory of scattering of generated chips by performing FEM analysis based on machining parameters included in an NC program for machining; and

and a position calculation unit that calculates a chip accumulation position based on the scattering trajectory of the chips and the shape in the machine calculated by the FEM analysis unit.

3. The chip disposal device for a machine tool according to claim 1 or 2,

the position estimating unit includes a machine learning device that outputs a chip accumulation position when a machining parameter and an internal shape included in an NC program for machining are input.

4. The chip disposal device for a machine tool according to claim 3,

the machine learning device is a primary learning completion device that performs primary learning in advance, using the machining parameters and the internal shape as input data, and using a chip accumulation position calculated based on a chip scattering trajectory obtained by FEM analysis based on the machining parameters and the internal shape as teaching data.

5. The chip disposal device for a machine tool according to claim 4,

the machine learning device is a secondary learning completion device in which the primary learning completion device performs secondary learning by using, as teaching data, the chip accumulation positions obtained from the in-machine images before and after machining.

6. The chip disposal device of a machine tool according to any one of claims 2 to 5,

the present invention is configured to include a machining control unit that moves a tool and a workpiece relative to each other, and to execute the process of estimating the deposition position by the position estimating unit in parallel with execution of the NC program by the machining control unit, wherein the nozzle posture adjusting unit adjusts the posture of the cleaning nozzle toward the deposition position.

7. A chip disposal method for a machine tool, comprising a cleaning nozzle control step of controlling a cleaning nozzle for injecting a cleaning fluid to chips scattered during machining and guiding the chips to a chip collection unit,

the cleaning nozzle controlling step includes:

a position estimating step of estimating a deposition position of generated chips by analyzing a machining condition of the workpiece; and

a nozzle posture adjusting step of adjusting the posture of the cleaning nozzle toward the deposition position estimated by the position estimating step.

8. The chip disposal method for a machine tool according to claim 7,

the position estimating step includes:

an FEM analysis step of performing FEM analysis based on machining parameters included in an NC program for machining and calculating a scattering trajectory of generated chips; and

a position calculation step of calculating a chip accumulation position from the scattering trajectory of the chips calculated in the FEM analysis step and the shape in the machine.

9. The chip disposal method for a machine tool according to claim 7,

the position estimating step is a step of inputting machining parameters and an in-machine shape included in an NC program for machining to a machine learning device, and outputting a chip deposition position.

10. The chip disposal method for a machine tool according to claim 9,

the machine learning device is a primary learning completion device that performs primary learning in advance using the machining parameters and the internal shape as input data, and using a chip accumulation position calculated based on a chip scattering trajectory obtained by FEM analysis based on the machining parameters and the internal shape as teaching data.

11. The chip disposal method for a machine tool according to claim 10,

the machine learning device is a secondary learning completion device in which the primary learning completion device performs secondary learning by using, as teaching data, the chip accumulation positions obtained from the in-machine images before and after machining.

Technical Field

The present invention relates to a chip treatment apparatus and a chip treatment method for a machine tool provided with a cleaning nozzle control unit that controls a cleaning nozzle that guides chips to a chip recovery unit by spraying a cleaning fluid onto chips scattered during machining.

Background

Patent document 1 discloses a machine tool including a nozzle for supplying a coolant to a portion to be machined of a workpiece by a tool, and configured to cool the workpiece and the tool and remove chips generated by machining.

Further, patent document 2 discloses a chip disposal apparatus that collects chips together with coolant into a coolant tank provided below a machining site and sends the chips out of the machine by a chip conveyor disposed at the bottom of the coolant tank.

Prior art documents

Patent document

Patent document 1: japanese Kohyo publication No. 2010-520071

Patent document 2: japanese laid-open patent publication No. 2002-96237

Disclosure of the invention

Technical problem to be solved by the invention

In the machine tool disclosed in patent document 1, a coolant discharge mechanism having a nozzle that is rotatable with respect to a tool holder mounting portion of a tool rest is provided, and chips scattered toward a tray or a table can be cleaned by appropriately rotating the nozzle toward a chip accumulation portion in the machine and discharging the coolant.

However, the worker who visually checks the accumulation state of the chips scattered on the tray or the table needs to adjust the posture of the nozzle by operating the operation panel, which is an inefficient work that requires labor and time.

In view of the above problems, an object of the present invention is to provide a chip disposal apparatus and a chip disposal method for a machine tool, which can automatically adjust the posture of a nozzle so as to discharge a cleaning fluid toward a chip deposition portion.

Means for solving the technical problem

In order to achieve the object, a chip disposal device for a machine tool according to the present invention includes a cleaning nozzle control unit that controls a cleaning nozzle that sprays a cleaning fluid to chips scattered during machining to guide the chips to a chip collection unit, the chip disposal device including: a position estimating unit that estimates a position of accumulation of generated chips by analyzing a machining condition of the workpiece; and a nozzle posture adjustment unit that adjusts the posture of the cleaning nozzle toward the deposition position estimated by the position estimation unit.

Further, a chip disposal method for a machine tool according to the present invention includes a cleaning nozzle control step of controlling a cleaning nozzle that sprays a cleaning fluid to chips scattered during machining to guide the chips to a chip recovery unit, the method including: a position estimation step of estimating a deposition position of generated chips by analyzing a machining condition of the workpiece; and a nozzle posture adjustment step of adjusting the posture of the cleaning nozzle toward the deposition position estimated by the position estimation step.

Effects of the invention

According to the present invention, it is possible to provide a chip disposal apparatus and a chip disposal method for a machine tool, which can automatically adjust the posture of a nozzle so as to discharge a cleaning fluid toward a chip deposition portion.

While the novel features of the present invention are set forth in the appended claims, the present invention will be further understood, both as to its structure and its content, from the following detailed description taken in conjunction with the accompanying drawings, together with other objects and features of the invention.

Drawings

Fig. 1 is an explanatory view of a machine tool incorporating a chip disposal device, in which (a) is an explanatory view showing a basic structure of the machine tool as viewed from a side surface, and (b) is an explanatory view showing a basic structure of the machine tool as viewed from a front surface.

Fig. 2 is an explanatory view of a main part of the chip disposal device.

Fig. 3 is an explanatory diagram of functional blocks constituting a control system of the machine tool.

Fig. 4 is a flowchart showing steps of a machining process performed by the machine tool.

Fig. 5 is a flowchart showing the steps of chip disposal performed by a chip disposal device incorporated in a machine tool.

Fig. 6 is an explanatory diagram of a neural network, showing an example of the machine learning device.

Fig. 7(a) is an explanatory view of a machine learning device in which primary learning is completed, which is provided in the position estimating unit, and (b) is an explanatory view of a machine learning device in which secondary learning is completed, which is provided in the position estimating unit.

Detailed Description

(basic embodiment of chip-processing apparatus for machine tool)

A chip disposal device for a machine tool according to the present invention includes a cleaning nozzle control unit that controls a cleaning nozzle that sprays a cleaning fluid to chips scattered during machining to guide the chips to a chip collection unit, the cleaning nozzle control unit including: a position estimating unit that estimates a position of accumulation of generated chips by analyzing a machining condition of the workpiece; and a nozzle posture adjusting section that adjusts the posture of the cleaning nozzle toward the deposition position estimated by the position estimating section.

That is, the position estimating unit analyzes the machining conditions of the workpiece, estimates the chip deposition position generated by machining, and the nozzle position adjusting unit adjusts the position of the cleaning nozzle so as to spray the cleaning fluid to the chip deposition position.

As one aspect, the position estimating unit preferably includes: an FEM analysis unit that calculates a trajectory of scattering of generated chips by performing FEM analysis based on machining parameters included in an NC program used for machining; and a position calculation unit that calculates a chip accumulation position based on the scattering trajectory of the chips calculated by the FEM analysis unit and the shape in the machine.

That is, in order to analyze the machining conditions of the workpiece, the FEM analysis unit is provided in the position estimation unit, and the FEM analysis unit executes FEM analysis based on the machining parameters included in the NC program, thereby calculating the scattering trajectory of the chips generated by the machining. Then, the position calculation unit provided in the position estimation unit calculates the chip accumulation position based on the chip scattering trajectory and a preset internal shape.

As another aspect, the position estimating unit preferably includes a machine learning device that outputs the position of the chip deposit when the machining parameters and the internal shape included in the NC program for machining are input.

That is, when the machining parameters and the internal shape are input to the machine learning device, the chip deposition position corresponding to the machining parameters and the internal shape is output.

Specifically, the machine learning device is preferably a primary learning completion device that performs primary learning in advance by using the machining parameters and the in-machine shape as input data and using the chip accumulation positions calculated based on the chip scattering trajectory and the in-machine shape obtained by FEM analysis based on the machining parameters as teaching data, and is excellent in that the primary learning can be performed quickly without requiring teaching data based on prior verification of the accumulation positions in the actual equipment.

The machine learning device is more preferably a secondary learning completion device in which the primary learning completion device performs secondary learning by using the accumulation position of chips obtained from the in-machine captured images before and after machining as teaching data, and can realize a machine learning device with higher accuracy by accumulating learning using teaching data based on verification of the accumulation position in the actual equipment.

Preferably, the robot cleaning apparatus includes a machining control unit that moves the tool and the object to be machined relative to each other, and the nozzle posture adjustment unit adjusts the posture of the cleaning nozzle toward the deposition position by executing the estimation process of the deposition position by the position estimation unit in parallel with execution of the NC program by the machining control unit.

In this way, when the machining process and the cleaning process using the cleaning nozzle are executed in parallel in the order of execution of the NC program that is the aggregate of the unit machining processes, the storage unit can be used more efficiently, and the waiting time until the machining process is started due to the estimation process of the accumulation position in advance can be shortened, compared to a case where the estimation process of the accumulation position of the chips corresponding to all the unit machining processes included in the NC program and a case where a large amount of calculation data necessary for the posture adjustment of the cleaning nozzle based on the estimation result are stored in the storage unit before the execution of the NC program.

[ basic embodiment of a method for processing chips of a machine tool ]

A chip disposal method for a machine tool according to the present invention includes a cleaning nozzle control step of controlling a cleaning nozzle that sprays a cleaning fluid to chips scattered during machining to guide the chips to a chip recovery unit, the method including: a position estimation step of estimating a deposition position of generated chips by analyzing a machining condition of the workpiece; and a nozzle posture adjustment step of adjusting the posture of the cleaning nozzle toward the deposition position estimated by the position estimation step.

The position estimation step includes: an FEM analysis step of performing FEM analysis based on a machining parameter included in an NC program for machining and calculating a scattering trajectory of generated chips; and a position calculation step of calculating a chip accumulation position based on the scattering trajectory of the chips and the shape in the machine calculated by the FEM analysis step.

The position estimating step is a step of inputting machining parameters and an in-machine shape included in an NC program for machining to a machine learning device, and outputting a chip deposition position.

The machine learning device is preferably a primary learning completion device that performs primary learning in advance using, as input data, a chip scattering trajectory and an in-machine shape obtained by FEM analysis based on machining parameters included in an NC program, and using, as teaching data, a chip accumulation position calculated based on the chip scattering trajectory and the in-machine shape.

The machine learning device is preferably a secondary learning device in which the primary learning device performs secondary learning using the chip accumulation position obtained from the in-machine image before and after machining as teaching data.

[ detailed embodiments of the chip disposal device for machine tool ]

Fig. 1 (a) and (b) show a machine tool 300 including a machine tool 100 incorporating the chip disposal device of the present invention and a control system 200 for controlling the machine tool 100 based on a preset NC program.

The machine tool 100 is a vertical machining center including a bed 1, a slide carriage 2 moving in the Y-axis direction along a guide surface on the bed 1, a table 3 moving in the X-axis direction along the guide surface of the slide carriage 2, a column 4 vertically provided on the bed 1, and a spindle head 5 moving in the Z-axis direction along the guide surface of the column 4, and is covered with a cover having an openable and closable door body, although not shown in the drawing, and an operation panel constituting a control system 200 is provided outside the cover.

When the servomotor MY is driven, the carriage 2 moves on the bed 1 along the linear drive axis in the Y-axis direction, when the servomotor MX is driven, the table 3 moves on the saddle 2 along the linear drive axis in the X-axis direction, and when the servomotor MZ is driven, the spindle head 5 moves on the column 4 along the linear drive axis in the Z-axis direction.

The tool 7 is held by a tool holder 6 provided in the spindle head 5, and when the servo motor MS1 is driven, the tool 7 rotates around the vertical axis. The table 3 is formed in a front view "コ" shape in which a pair of vertical walls 3W are arranged to face each other, and workpiece holders 3H holding workpieces 10 to be processed are provided on the vertical walls 3W, respectively, and when the servo motor MS2 is driven, the workpieces 10 held by the workpiece holders 3H rotate about horizontal axes along the X axis. That is, the table 3 serves as a workpiece holding portion. For example, when the side surface machining, the groove machining, or the like of the workpiece 10 is intended, an end mill having cutting edges on the outer peripheral surface and the end surface is used as the tool 7.

By driving the above-described servomotors via the servo control unit based on a preset NC program, the workpiece 10 and the tool 7 are moved relative to each other, and the workpiece 10 is machined into a desired shape.

A coolant tank 8 for recovering coolant, which is a fluid supplied for cooling and cleaning, is provided below the shoe 2, and chips generated by machining are recovered to the coolant tank 8 together with the coolant. A chip conveyor 9 is disposed at the bottom of the coolant tank 8, and chips collected in the coolant tank 8 are sent out of the apparatus by the chip conveyor 9 and collected in a collection container.

The machine tool 100 is provided with a chip treatment device 50 which, when a workpiece 10 is machined by a tool 7, sprays a coolant as a cleaning fluid onto a deposition position of chips 12 so as to prevent the chips 12 scattered into the machine from being deposited on the workpiece 10, the table 3, the saddle 2, and the like, and guides the chips 12 together with the coolant to the coolant tank 8. This is because the chips become high temperature due to heat generated during cutting, and the workpiece 10, the table 3, the saddle 2, and the like are thermally displaced, which may reduce machining accuracy, and it is very difficult to clean and remove a large amount of accumulated chips later.

As shown in fig. 2, the chip disposal device 50 includes a cleaning nozzle mechanism 51 and a cleaning nozzle control unit 30 that controls the cleaning nozzle mechanism 51.

The cleaning nozzle mechanism 51 includes: an annular body 52 rotatably attached to the outer periphery of one of the work holders 3H via a bearing; and a cleaning nozzle 53 attached to the annular body 52. The cleaning nozzle 53 is attached to the side surface of the ring body 52 so as to protrude toward the other vertical wall 3W, and the tip of the cleaning nozzle 53 is configured to be rotatable about an axis P2 perpendicular to the rotation axis P1 of the workpiece 10 by a motor M2 provided in the cleaning nozzle 53.

The ring body 52 is configured to be rotatable around the rotation axis P1 of the workpiece 10 coaxially therewith via a motor M1 attached to the vertical wall 3W and a gear mechanism for driving transmission. The rotation angle of the annular body 52 is adjusted by the motor M1, and the inclination angle of the cleaning nozzle 53 is adjusted by the motor M2, whereby the cooling liquid can be ejected from the cleaning nozzle 53 in any direction. The annular body 52 is provided with a fluid flow path for guiding the coolant to the cleaning nozzle 53, and the coolant collected in the coolant tank 8 is circulated and supplied to the fluid flow path through a dust removing filter and a fluid transfer pipe, for example.

The cleaning nozzle control unit 30 includes: a position estimating unit 31 for estimating a position of chip deposition caused by machining; and a nozzle posture adjustment unit 34 that controls the rotation of the motors M1 and M2 so as to adjust the posture of the cleaning nozzle 53 toward the deposition position estimated by the position estimation unit 31. The motors M1 and M2 are provided with encoders for detecting the rotational position of the drive shaft, and the nozzle posture adjustment unit 34 controls the rotational position of the annular body 52 and the inclination angle of the cleaning nozzle 53 so that the target rotational position and inclination angle are obtained based on the outputs of the encoders.

As shown in fig. 3, the control system 200 includes a system control unit 20, a servo control unit 40 as an example of a machining control unit for controlling the servo motors, and a cleaning nozzle control unit 30 for controlling the cleaning nozzle mechanism 51. Each of the control units 20, 30, and 40 includes: a main board having a CPU, a ROM, and a RAM; an I/O board that exchanges various control data between the main board and the machine tool 100 or an operator; and hardware such as a communication board that exchanges necessary information between the respective motherboards.

Various programs and data such as a system control program, an NC program, tool parameters, and internal shape data are stored in the ROM of the system control unit 20, a servo control program for controlling various servo motors is stored in the ROM of the servo control unit 40, and a cleaning nozzle control program is stored in the ROM of the cleaning nozzle control unit 30.

The servo control unit 40 realizes the respective functional blocks of the positioning control unit 41 and the velocity control unit 42 by the CPU executing a servo control program, and the cleaning nozzle control unit 30 realizes the respective functional blocks of the position estimating unit 31 and the nozzle posture adjusting unit 34 by the CPU executing a cleaning nozzle control program.

The position estimating unit 31 executes FEM analysis based on machining parameters included in an NC program for machining, and embodies the functional blocks of the FEM analyzing unit 32 that calculates the scattering locus of generated chips, and the position calculating unit 33 that calculates the accumulation position of chips in the machine based on the scattering locus of chips and the shape in the machine calculated by the FEM analyzing unit 32.

When a start switch provided in the operation panel is operated by the operator, the system control unit 20 reads the NC program and transmits a necessary control command to the servo control unit 40. Upon receiving a control command from the system control unit 20, the servo control unit 40 controls each servo motor via the positioning control unit 41 and the speed control unit 42 so that the tool and the workpiece move to a predetermined relative position at a predetermined speed. The form of the control command is not particularly limited, and may be a pulse train signal capable of specifying a feed amount, a feed speed, and the like necessary for servo control, for example. The control command includes the position of the tool and the workpiece, the rotation speed of the spindle, the feed speed, and the like.

In parallel with the execution of the NC program by the servo control unit 40, the system control unit 20 transmits the machining conditions, the tool parameters, and the internal shape data based on the NC program to the position estimating unit 31 provided in the cleaning nozzle control unit 30. The FEM analysis unit 32 included in the position estimation unit 31 calculates a scattering trajectory of chips generated by machining by performing FEM analysis on the machining state of the workpiece 10 based on the machining conditions of the workpiece 10 and the tool parameters included in the NC program, and the position calculation unit 33 calculates a deposition position of chips based on the scattering trajectory of chips calculated by the FEM analysis unit 32 and the internal shape.

Then, the nozzle posture adjustment unit 34 controls the motors M1 and M2 to adjust the posture of the cleaning nozzle 53 so that the coolant is ejected to the chip deposition position calculated by the position calculation unit 33.

That is, the chip disposal device of the machine tool includes the servo control unit 40 that machines the workpiece into a desired shape by relatively moving the spindle head 5 that holds the tool 7 via the tool holder 6 and the workpiece holding unit (table 3) that holds the workpiece 10 via the workpiece holder 3H based on the NC program, executes the estimation processing of the deposition position by the position estimation unit 31 in parallel with the execution of the NC program by the servo control unit 40, and the nozzle posture adjustment unit 34 adjusts the posture of the cleaning nozzle 53 toward the deposition position.

[ detailed embodiments of the method for processing chips of a machine tool ]

Fig. 4 shows the steps of the machining process executed by the system control unit 20 and the servo control unit 40.

When the start switch is operated (SA1), the system controller 20 starts the chip conveyor 9(SA2), reads the NC program stored in the storage unit (SA3), analyzes the machining command incorporated in the NC program, and transmits a control command to the servo controller 40 (SA 4).

The servo control unit 40 that has received the control command controls the feed speed and feed amount of the relevant servo motor so that the tool 7 and the workpiece 10 are positioned at a predetermined relative position, drives the main spindle to which the tool 7 is attached at a predetermined rotational speed via the Z-axis rotary motor MS1, and drives the workpiece 10 at a predetermined rotational speed via the X-axis rotary motor MS2 (SA 5).

The processing of steps SA3 to SA5 is repeated until the execution of all the commands included in the NC program is completed, whereby the workpiece 10 is machined into a desired shape (SA6, N), and when the execution of all the commands is completed (SA6, Y), the chip conveyor 9 is stopped, and the machining is completed.

Fig. 5 shows a procedure of the cleaning nozzle control process executed by the system control unit 20 and the cleaning nozzle control unit 30.

At the initial stage, the system control unit 20 transmits the internal shape data to the position calculation unit 33, and stores the internal shape data in the storage unit provided in the cleaning nozzle control unit 30 (SB 1).

Next, the system control unit 20 reads the NC program stored in the storage unit (SB2), analyzes the machining instruction assembled in the NC program, and transmits the analysis instruction and the tool parameter to the FEM analysis unit 32 (SB 3). When the FEM analysis unit 32 receives the analysis command and the tool parameters (SB4), a predetermined FEM analysis program is executed to calculate the scattering trajectory of the chips generated by the machining (SB 5).

The position calculating unit 33 calculates the chip accumulation position based on the scattering trajectory of the chips calculated by the FEM analyzing unit 32 and the in-machine shape data (SB 6). The nozzle posture adjustment unit 34 controls the motors M1 and M2 to spray the coolant to the chip deposition position calculated by the position calculation unit 33 to adjust the posture of the cleaning nozzle 53 (SB7), opens a valve provided in the coolant supply path, and sprays the coolant from the cleaning nozzle 53 until the execution of the machining command is completed (SB 8). When the execution of the machining command is completed (SB9, Y), the valve provided in the coolant supply path is closed to stop the supply of the coolant (SB 10).

By repeating the processing from step SB2 to step SB10 until the execution of all the commands included in the NC program is completed, the chips generated until the workpiece 10 is machined into the desired shape are collected in the coolant tank 8 without being accumulated on the table 3 and the shoe 2. When the execution of all the instructions is finished (SB11, Y), the control process of cleaning the nozzles is finished.

That is, the cleaning nozzle control step is constituted by a position estimation step (SB5, BB6) for estimating the deposition position of the generated chips by analyzing the machining conditions of the workpiece as the workpiece, and a nozzle posture adjustment step (SB7) for adjusting the posture of the cleaning nozzle toward the deposition position estimated by the position estimation step.

Further, the position estimation step is configured by an FEM analysis step (SB5) of calculating a scattering trajectory of the generated chips by performing FEM analysis based on the machining parameters included in the NC program for machining and a position calculation step (SB6) of calculating a deposition position of the chips based on the scattering trajectory of the chips calculated by the FEM analysis step and the shape in the machine.

[ details of FEM analysis ]

The FEM analyzer 32 is an arithmetic processing unit that calculates the generation and scattering direction of chips generated by cutting or the like using AvantEdge (registered trademark) which is software dedicated to chip generation simulation, for example, and analyzes the elastoplastic deformation and the thermal conduction by numerical arithmetic processing using a Finite Element Method (Finite Element Method) with an analysis command including the shape, material characteristics, cutting conditions, and the like of the workpiece, the tool, and tool parameters as inputs, and outputs the cutting resistance, the chip shape, the temperature, the stress distribution, the scattering direction, the scattering speed, and the like. The software for chip generation simulation is not limited to AvantEdge (registered trademark), and other software may be used.

In the finite element method, a structure is divided into triangular elements each composed of 3 nodes, and the relationship between an external force { f } acting on a node and a displacement { V } of the node is determined by the equation { f } - [ K ] { V } through a rigid matrix [ K ], and analysis from elastic deformation to plastic deformation is performed. The external force { f }, the rigidity matrix [ K ], and the like are set based on machining parameters such as tool parameters and analysis commands.

The tool parameters include tool materials such as tool steel and cemented carbide, tool types such as a drill, a milling cutter, and an end mill, tool characteristics such as the number of edges, the shape of a bottom edge, and a helix angle in the case of a drill, tool characteristics such as the number of edges, the shape of a bottom edge, and a helix angle in the case of an end mill, and material characteristics of a workpiece (workpiece) such as stainless steel and aluminum. The tool parameters are stored in advance in a storage unit provided in the system control unit 20, and based on tool specifying information and the like defined by the NC program, tool parameters necessary for analysis are extracted and supplied to the FEM analysis unit 32.

The analysis command includes a machining path for the workpiece, the rotational speed of the tool, the feed rate per 1 blade, the depth of cut, and the like, and these values are grasped by the NC program. Further, as a tool load, drive power of a motor that drives the spindle as needed, and the like are supplied from the servo control unit 40 to the FEM analysis unit 32 via the system control unit 20. That is, the machining parameters required for the FEM analysis are configured by any one or a combination of the tool load, the tool parameters, and the analysis command.

When the estimation processing of the deposition position by the position estimation unit 31 is executed in parallel with the execution of the NC program by the servo control unit 40, if the arithmetic processing capability by the FEM analysis unit 32 is sufficiently ensured, the execution of the NC program by the servo control unit 40 and the estimation processing of the deposition position by the position estimation unit 31 can be executed in real time in synchronization.

However, when the arithmetic processing capability by the FEM analyzing unit 32 is insufficient, a time difference may be provided between the execution of the estimation processing of the deposition position by the position estimating unit 31 and the execution of the NC program by the servo control unit 40, and the execution of the NC program by the servo control unit 40 may be performed in parallel after the estimation processing of the deposition position by the position estimating unit 31 is advanced.

That is, since the NC program is constituted by an aggregate of a plurality of machining processes, if the estimation processing of the deposition position by the position estimation unit 31 is executed before the execution timing of the servo control unit 40 for each machining process, the coolant can be injected in real time to the deposition position of the chips at the execution timing of the servo control unit 40.

In addition to the execution of the NC program by the servo control unit 40 and the estimation processing of the deposition position by the position estimation unit 31 being performed in parallel, the NC program by the servo control unit 40 may be executed after all the estimation processing of the deposition position by the position estimation unit 31 is completed.

[ first mode of position estimation processing ]

The position calculator 33 can be configured to calculate the falling position and the falling amount of the chips that change in time series with machining, based on the scattering trajectory of the chips calculated from the scattering direction and the scattering speed of the chips obtained by the numerical operation processing described above, and the in-machine shape data indicating the three-dimensional shapes of the workpiece 10, the saddle 2, the table 3, and the like.

[ Another mode of position estimation processing ]

In the above example, the mode in which the position estimating unit 31 is configured by the FEM analyzing unit 32 and the position calculating unit 33 has been described, but the position estimating unit may be configured not to include the FEM analyzing unit 32 and the position calculating unit 33, and may include a machine learning device that outputs chip accumulation information including at least the accumulation position of chips when the machining parameters and the internal shape are input. The chip accumulation information preferably further includes an accumulation amount and an accumulation time.

A neural network suitable as such a machine learning device is shown in fig. 6.

Each node constituting the input layer and each node constituting the intermediate layer are coupled to each other with predetermined coupling coefficients Wi, n (n is the product of the number of nodes of the input layer and the number of nodes of the intermediate layer), and each node constituting the intermediate layer and each node constituting the output layer are coupled to each other with predetermined coupling coefficients Wo, m (m is the product of the number of nodes of the intermediate layer and the number of nodes of the output layer).

One section of the storage unit, which is a unit of arithmetic processing, corresponds to each node. For example, if the unit of arithmetic processing is 16 bits, the value of each node is expressed by 16-bit data.

The value input to each node of the input layer is weighted-added based on the coupling coefficients Wi, n and the activation function and input to each node of the intermediate layer, and the value input to each node of the intermediate layer is weighted-added based on the coupling coefficients Wi, m and the activation function and input to each node of the output layer. That is, when the machining parameters and the internal shape are input to each node constituting the input layer, the deposition position, deposition amount, and deposition timing of the chips in the machine are output from the output layer. As the activation function, a step function, a sigmoid function, or the like is used.

As the machining parameters, machining parameters such as the tool material, the tool type, the tool characteristics, the material of the workpiece, the machining path, the rotational speed of the tool, the feed speed per 1-pass, and the depth of cut, and the internal shape are input.

For example, the tool material, the tool type, the tool characteristics, and the material of the workpiece are configured such that each node of the input layer is assigned a separate option, 1 is input to the selected node, and zero is input to the non-selected node.

In addition, with regard to the characteristics of the same amount of the rotational speed of the tool, the feed rate per 1 blade, and the depth of cut, a plurality of numerical ranges are allocated to each node of the input layer in advance, 1 is input to the corresponding node, and zero is input to the node that does not match.

The internal shape is configured such that the internal shape is divided into a plurality of square regions in a plan view, each divided region is assigned to a node, and the height of the square region normalized in a range of zero to 1 is input to the corresponding node.

The output layer includes a node indicating the amount of chip accumulation and a node indicating the accumulation time for each of the square regions. The node indicating the amount of chip accumulation outputs the amount of accumulation in the range from zero to 1, and the node indicating the accumulation time outputs the elapsed time indicated by the range from zero to 1 normalized in the time required from the start to the end of machining. When zero is output to a node indicating the accumulation amount, it can be determined that no chips are accumulated in the square region specified by the node, and when a value other than zero is output, it can be determined that a larger amount of chips are accumulated as the value approaches 1.

Such a neural network is learned in advance based on teaching data so that the coupling coefficients Wi, n and Wo, m become optimal values.

This will be specifically explained. The chip accumulation amount and the accumulation time for each square region obtained by performing FEM analysis on the machining parameters and the internal shape are prepared as teaching data for each standard unit machining process using an FEM analysis device.

Thereafter, the operation of adjusting the coupling coefficients Wi, n and Wo, m is repeated so that the difference between the teaching data and the data output from the output layer becomes minimum when the machining parameters and the in-machine shape corresponding to each standard unit machining process are input to the input layer. As such a learning algorithm, an error propagation method is preferably used.

That is, as shown in fig. 7(a), the machine learning device is configured by a primary learning completion device that takes machining parameters and the internal shape as input data, and performs primary learning in advance using, as teaching data, the chip accumulation position calculated based on the chip scattering trajectory and the internal shape obtained by FEM analysis based on the machining parameters.

[ still another mode of the position estimation processing ]

In the neural network shown in fig. 6, a case where a primary learning completion device that learns using analysis results obtained by an FEM analysis device for teaching data is used as the machine learning device has been described, but as shown in fig. 7(b), it is more preferable that the neural network be configured as a secondary learning completion device that performs secondary learning by using the accumulated position of chips obtained from the in-machine captured images before and after machining as teaching data.

In an actual machine tool, a captured image in an overhead view showing an in-machine state of a chip accumulation state generated when an NC program defining a unit machining process specified by predetermined machining parameters is executed and a captured image in the machine before the chip is scattered are compared by image processing, teaching data showing which square region of the machine divided into a plurality of square regions in the overhead view has the chip accumulated to what extent is generated, and the teaching data is applied to a primary learning completion device and subjected to secondary learning, whereby more appropriate coupling coefficients Wi, n and Wo, m can be adjusted.

The neural network is an example, and the number of input layers, intermediate layers, nodes constituting an output layer, the type and format of input data input to the input layers, and the type and format of output data output from the output layers can be appropriately set, and are not limited to this example.

The learning algorithm may use an algorithm other than the error propagation method, or may use a neural network configured to increase the number of intermediate layers and perform deep learning.

[ other embodiments ]

Although not particularly described in the above embodiment, other cleaning nozzles for cleaning and removing chips scattered in the machine, as well as other cleaning nozzles for supplying a coolant for the purpose of lubrication, cooling, chip removal, and the like, may be provided at the contact position between the tool and the workpiece.

The fluid ejected from the cleaning nozzle is not limited to the coolant, and other fluids such as compressed gas may be used as the cleaning fluid to remove the chips scattered into the machine.

In the above-described embodiment, the example in which the machine tool 100 is configured by the vertical machining center has been described, but the machine tool 100 to which the present invention is applied is not limited to the vertical machining center, and can be applied to various machining centers, and can also be applied to a machine tool such as a lathe that does not include a spindle head that holds a tool.

The specific structure and mounting position of the cleaning nozzle mechanism 51 are not limited to those in the above-described embodiments, and the cleaning nozzle mechanism may be mounted to another member such as a spindle head, or the number of the cleaning nozzle mechanisms 51 may be plural.

While the embodiments and embodiments of the present invention have been described above, the disclosure may be changed in details of the structure, and combinations of elements in the embodiments and changes in the order thereof may be made without departing from the scope and spirit of the present invention as claimed.

Industrial applicability of the invention

As described above, according to the present invention, it is possible to realize a machine tool including a chip disposal device capable of automatically adjusting the posture of a nozzle and ejecting a cleaning fluid to a chip deposition portion without depending on an operation of an operator.

Description of the symbols

1: lathe bed

2: slide carriage seat

3: working table

3W: vertical wall

3H: workpiece holder

4: upright post

5: spindle head

6: tool rack

7: cutting tool

8: cooling liquid tank

9: chip conveyor

10: processed object (workpiece)

12: cutting scrap

20: system control unit

30: cleaning nozzle control part

31: position estimating unit

32: FEM analysis unit

33: position calculating part

34: nozzle posture adjusting part

40: servo control unit

41: positioning control part

42: speed control unit

50: chip disposal device

100: machine tool

200: control system

300: working system