MACHINING CONDITION SETTING DEVICE AND 3D LASER MACHINING SYSTEM

• US 20200130107A1
• Filed: 10/07/2019
• Published: 04/30/2020
• Est. Priority Date: 10/30/2018
• Status: Active Grant

• Current Assignee: Fanuc Corporation
• Original Assignee: Fanuc Corporation
• Inventors: MOCHIZUKI, Tatsuya, TAKIGAWA, Hiroshi
• Granted Patent: US 11,167,377 B2
• Time in Patent Office: Days
• CPC Class Codes: B23K 26/0626 Energy control of the laser beam B23K26/0622 takes precedence
• CPC Class Codes: B23K 26/0884 in at least in three axial directions, e.g. manipulators, robots
• CPC Class Codes: B23K 26/21 by welding
• CPC Class Codes: B23K 26/36 Removing material B23K26/55, B23K26/57 take precedence
• CPC Class Codes: B23K 26/38 by boring or cutting
• CPC Class Codes: B23K 26/702 Auxiliary equipment
• CPC Class Codes: B23K 26/705 Beam measuring device
• CPC Class Codes: B25J 9/1671 characterised by simulation, either to verify existing program or to create and verify new program, CAD/CAM oriented, graphic oriented programming systems
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion
• CPC Class Codes: G05B 13/042 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 19/4097 characterised by using design data to control NC machines, e.g. CAD/CAM G05B19/4093 takes precedence
• CPC Class Codes: G05B 2219/35134 3-D cad-cam
• CPC Class Codes: G05B 2219/40318 Simulation of reaction force and moment, force simulation
• CPC Class Codes: G05B 2219/45165 Laser machining
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A 3D laser machining system comprises: a move state simulation unit that simulates a move state of a machining head using 3D CAD data about a workpiece containing material information defining thermophysical properties and 3D CAD data about a machining head under a condition of moving the machining head relative to the workpiece while the machining head is maintained at a predetermined angle a predetermined distance along a machining line in virtual space; a thermal fluid simulation unit that conducts non-stationary thermal fluid simulation for obtaining a temperature distribution in a region covering the workpiece to be changed by the move of the machining head outputting a laser beam; and a machining condition setting unit that sets a laser machining condition containing a relative move condition for the machining head and a laser beam output condition before laser machining on the basis of results of the simulations.

Claims

1. A machining condition setting device comprising:

a move state simulation unit that simulates a move state of a machining head containing at least a relative position or a state of temporal change in a relative move speed of the machining head relative to a workpiece using 3D CAD data about the workpiece containing material information defining thermophysical properties including at least heat conductivity, specific heat, and density, and 3D CAD data about the machining head containing information at least about an outer shape of the machining head and about a relative position of the optical axis of a laser beam output from the machining head relative to the outer shape of the machining head, with an intersection point of a machining surface of the workpiece defined by the 3D CAD data about the workpiece and the optical axis of the machining head determined to be a machining point in virtual space, the simulation being conducted under a condition of moving the machining head relative to the workpiece along a machining line set as a path of the machining point determined when the machining head moves relative to the workpiece in the virtual space while the optical axis of the machining head is maintained at a predetermined angle from each of a normal vertical to the machining surface on which the machining point passes through and the machining line, and while a distance between an end surface of the machining head on a laser beam output side and the machining point is maintained at a predetermined distance;

a thermal fluid simulation unit that conducts non-stationary thermal fluid simulation for obtaining a temperature distribution in a region covering at least a part of the workpiece to be changed by the move of the machining head outputting a laser beam; and

a machining condition setting unit that sets a laser machining condition in advance containing at least a relative move condition for the machining head and a laser beam output condition before implementation of actual laser machining on the basis of simulation results obtained by the move state simulation unit and the thermal fluid simulation unit.

2. A 3D laser machining system comprising:

the machining condition setting device according to claim 1;

a laser device comprising at least one laser oscillator, a power supply unit that supplies the laser oscillator with power for laser oscillation, a laser optical system for propagation of a laser beam output from the laser oscillator to the machining head, at least one photodetection means capable of detecting at least one of a laser output beam from the laser oscillator propagated in the laser optical system and a returning beam propagated in the laser optical system in an opposite direction to the laser output beam, and a control unit that receives output signals from at least the machining condition setting unit and the photodetection means and outputs a control signal containing at least a power output command to the power supply unit; and

a driver that receives the control signal from the control unit and moves the machining head relative to the workpiece.

3. The 3D laser machining system according to claim 2, comprising the at least one photodetection means capable of detecting the returning beam, wherein

if the control unit determines that the quantity of the returning beam is required to be reduced on the basis of a result of detection of the returning beam obtained by the photodetection means, the control unit outputs the control signal to the power supply unit and the driver so as to satisfy a condition allowing a temperature at the machining point or a temperature at the workpiece in the vicinity of the machining point to be maintained at a temperature close to an intended temperature by changing the laser machining condition set in advance by the machining condition setting unit at least temporarily and changing the laser machining condition containing both the laser beam output condition and the relative move speed of the machining head relative to the workpiece.

4. The 3D laser machining system according to claim 2, further comprising ranging means capable of measuring a distance between the end surface of the machining head on the laser beam output side and the machining point, wherein

on the basis of a measurement result obtained by the ranging means during implementation of actual laser machining, the control unit fulfills a function of compensating for at least the relative move condition for the machining head of the laser machining condition set in advance on the basis of a result of the machining head move simulation.

5. The 3D laser machining system according to claim 2, further comprising temperature detection means capable of detecting at least one of a temperature at the machining point moving along the machining line during laser machining and a surface temperature of the workpiece in the vicinity of the machining point, wherein

if a temperature difference occurs between a detected temperature detected by the temperature detection means at a certain time point in laser machining on the workpiece performed under the laser machining condition set by the machining condition setting device and a predicted temperature predicted at a position of a time point corresponding to the certain time point contained in the simulation result obtained under the laser machining condition set by the machining condition setting device, the control unit fulfills a function of compensating for at least a part of the laser machining condition set by the machining condition setting device in response to the temperature difference, and

the laser device continues laser machining under the compensated laser machining condition.

6. The 3D laser machining system according to claim 2, wherein the thermal fluid simulation unit limits effect of irradiation of the workpiece with a laser beam from the machining head to temperature increase at the workpiece achieved by heat input to the workpiece, and conducts the non-stationary thermal fluid simulation on the assumption that a phase transition of the workpiece does not occur.

7. The 3D laser machining system according to claim 6, wherein, for at least a part of a machining zone along the machining line, the machining condition setting device sets a laser machining condition to obtain a simulation result that at least one of a temperature at the machining point and a temperature in the vicinity of the machining point is at a predetermined temperature at any relative move speed of the machining head moving relative to the workpiece.

8. The 3D laser machining system according to claim 7, wherein the thermal fluid simulation unit contains correlation data indicating correlation between the temperature at the machining point or in the vicinity of the machining point contained in the simulation result obtained under the laser machining condition set by the machining condition setting unit and a temperature at a corresponding position measured during implementation of actual laser machining under the laser machining condition, and

the thermal fluid simulation unit makes reference to the correlation data to determine a predetermined replacement temperature intended in the non-stationary thermal fluid simulation as a replacement for the predetermined temperature intended during implementation of actual laser machining, and then conducts the non-stationary thermal fluid simulation.

9. The 3D laser machining system according to claim 7, wherein the machining condition setting device further comprises a machining condition storage unit,

the machining condition storage unit contains a substantially optimum acquired laser machining condition for a workpiece condition prepared for each type of laser machining and containing at least a material and a thickness about the workpiece having a plate-like shape, the substantially optimum acquired laser machining condition containing laser machining conditions including at least laser beam output, the relative move speed of the machining point relative to the plate-like workpiece, and a distance between the end surface of the machining head on the laser beam output side and the machining point, and

regarding at least a partial particular zone of a zone in which the machining line set at the workpiece has a linear shape and the workpiece has a constant thickness along the machining line, the machining condition setting unit sets, from among the optimum laser machining conditions stored in the machining condition storage unit, an approximately optimum laser machining condition with conforming or approximately conforming to a workpiece condition, and regarding at least a partial zone of the zone along the machining line other than the particular zone and at least adjacent to the particular zone, the machining condition setting unit sets a laser machining condition under which, with a temperature at the machining point or in the vicinity of the machining point obtained as the simulation result conducted under the approximately optimum laser machining condition set for the particular zone determined to be the predetermined temperature, a temperature at the machining point or in the vicinity of the machining point obtained by the non-stationary thermal fluid simulation becomes the same predetermined temperature as in the particular zone.

10. The 3D laser machining system according to claim 6, further comprising temperature detection means capable of detecting at least one of a temperature at the machining point moving along the machining line during laser machining and a surface temperature of the workpiece in the vicinity of the machining point, wherein

if a temperature difference occurs between a detected temperature detected by the temperature detection means at a certain time point in laser machining on the workpiece performed under the laser machining condition set by the machining condition setting device and a predicted temperature predicted at a position of a time point corresponding to the certain time point contained in the simulation result obtained under the laser machining condition set by the machining condition setting device, the control unit fulfills a function of compensating for at least a part of the laser machining condition set by the machining condition setting device in response to the temperature difference, and

the laser device continues laser machining under the compensated laser machining condition.

11. The 3D laser machining system according to claim 6, further comprising temperature detection means and a first machine learning device, wherein

the temperature detection means is capable of detecting at least one of a temperature at the machining point moving along the machining line during laser machining and a surface temperature of the workpiece in the vicinity of the machining point,

the first machine learning device comprises:

a first state observation unit that receives state data about the 3D laser machining system as input data containing at least the laser machining condition set by the machining condition setting unit and the simulation result obtained under the set laser machining condition;

a label acquisition unit that acquires temporal change data about the temperature detected by the temperature detection means as a label corresponding to answer data; and

a first learning unit that receives a large number of pairs of the input data and the label to learn a relationship between the input data and the label,

the first learning unit predicts the temporal change data about the temperature detected by the temperature detection means relative to input data newly input using a result of the learning, and

the machining condition setting device refers to the temporal change data about the temperature predicted by the first learning unit to adjust the laser machining condition.

12. The 3D laser machining system according to claim 11, wherein if a temperature difference between a detected temperature detected by the temperature detection means at a certain time point in laser machining on the workpiece performed under the laser machining condition adjusted with reference to the temporal change date about the temperature predicted the first learning unit by the machining condition setting device and a predicted temperature predicted at a position of a time point corresponding to the certain time point contained in the simulation result obtained under the adjusted laser machining condition exceeds than a predetermined temperature difference, the control unit fulfills a function of compensating for at least a part of the laser machining condition adjusted by the machining condition setting device in response to the temperature difference, and

the laser device continues laser machining under the compensated laser machining condition.

13. The 3D laser machining system according to claim 11, further comprising a higher-order computer, a higher-order network, and a lower-order network, wherein

the first machine learning device and a plurality of the laser devices are connected through the lower-order network to form a manufacturing cell,

a plurality of the manufacturing cells and the higher-order computer are connected through the higher-order network, the higher-order computer functions at least as the thermal fluid simulation unit of the machining condition setting device,

the first state observation unit of the first machine learning device receives at least the laser machining condition as a simulation condition for the non-stationary thermal fluid simulation by the thermal fluid simulation unit and at least temporal change data about a temperature expected to be detected by the temperature detection means as a part of a simulation result of the non-stationary thermal fluid simulation input through the higher-order network and the lower-order network,

the temporal change data about the temperature detected by the temperature detection means is input to the label acquisition unit of the first machine learning device through the lower-order network, and

a learning model is shared between a plurality of the first learning units through the higher-order network and the lower-order network.

14. The 3D laser machining system according to claim 2, further comprising at least one in-line image monitoring device and a second machine learning device, wherein

the in-line image monitoring device is capable of acquiring image data indicating a 2D shape or a 3D shape at a surface of the workpiece or inside the workpiece in at least either a part of the workpiece being subjected to laser machining or a part of the workpiece immediately after the laser machining;

the second machine learning device comprises;

a second state observation unit that observes a state inside the 3D laser machining system and a state outside the 3D laser machining system containing at least the laser machining condition set by the machining condition setting unit and shape data about the workpiece corresponding to the image data as the simulation result obtained under the set laser machining condition, and outputs the observed states as state data;

a determination data acquisition unit that acquires time-series image data obtained by the in-line image monitoring device, and outputs a result of comparison between the acquired image data and sample image data given a score stored in advance as determination data;

a second learning unit that receives the outputs from the second state observation unit and the determination data acquisition unit; and

a decision making unit that determines a laser machining condition to be output from the second learning unit on the basis of a value function as a result of learning by the second learning unit,

the second learning unit comprises;

a reward calculation unit that calculates a reward from the determination data; and

a value function update unit that updates the value functions sequentially on the basis of the calculated reward, and

the second learning unit associates the laser machining condition output from the decision making unit with the state data input from the second state observation unit and the determination data input from the determination data acquisition unit, and learns an optimum laser machining condition in each state in a trial-and-error method on the basis of the determination data.

15. The 3D laser machining system according to claim 14, wherein, when the determination data acquisition unit compares the image data obtained by imaging by the in-line image monitoring device with the sample image data given a score and outputs determination data indicating that the score of the image data obtained by imaging by the in-line image monitoring device is lower than a predetermined point during laser machining on the workpiece under the laser machining condition output from the second machine learning device, the control unit fulfills a function of compensating for at least a part of the laser machining condition originally output from the second machine learning device, and

the laser device continues laser machining under the compensated laser machining condition.

16. The 3D laser machining system according to claim 14, further comprising a higher-order computer, a higher-order network, and a lower-order network, wherein

the second machine learning device and a plurality of the laser devices are connected through the lower-order network to form a manufacturing cell,

a plurality of the manufacturing cells and the higher-order computer are connected through the higher-order network, the higher-order computer functions at least as the thermal fluid simulation unit of the machining condition setting device,

the second state observation unit of the second machine learning device receives at least the laser machining condition as a simulation condition for the non-stationary thermal fluid simulation by the thermal fluid simulation unit and at least the shape data about the workpiece corresponding to the image data as a part of a simulation result of the non-stationary thermal fluid simulation input through the higher-order network and the lower-order network,

the image data obtained by imaging by the in-line image monitoring device is input to the determination data acquisition unit of the second machine learning device through the lower-order network, and

the value function is shared between a plurality of the second learning units through the higher-order network and the lower-order network.