WO2023140015A1 - 積層造形物の製造方法及び製造装置、制御支援装置、並びにプログラム - Google Patents
積層造形物の製造方法及び製造装置、制御支援装置、並びにプログラム Download PDFInfo
- Publication number
- WO2023140015A1 WO2023140015A1 PCT/JP2022/046977 JP2022046977W WO2023140015A1 WO 2023140015 A1 WO2023140015 A1 WO 2023140015A1 JP 2022046977 W JP2022046977 W JP 2022046977W WO 2023140015 A1 WO2023140015 A1 WO 2023140015A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- laminate
- state transition
- shape
- model
- shape index
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B33—ADDITIVE MANUFACTURING TECHNOLOGY
- B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- B33Y30/00—Apparatus for additive manufacturing; Details thereof or accessories therefor
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/34—Laser welding for purposes other than joining
- B23K26/342—Build-up welding
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22F—WORKING METALLIC POWDER; MANUFACTURE OF ARTICLES FROM METALLIC POWDER; MAKING METALLIC POWDER; APPARATUS OR DEVICES SPECIALLY ADAPTED FOR METALLIC POWDER
- B22F10/00—Additive manufacturing of workpieces or articles from metallic powder
- B22F10/20—Direct sintering or melting
- B22F10/25—Direct deposition of metal particles, e.g. direct metal deposition [DMD] or laser engineered net shaping [LENS]
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22F—WORKING METALLIC POWDER; MANUFACTURE OF ARTICLES FROM METALLIC POWDER; MAKING METALLIC POWDER; APPARATUS OR DEVICES SPECIALLY ADAPTED FOR METALLIC POWDER
- B22F10/00—Additive manufacturing of workpieces or articles from metallic powder
- B22F10/30—Process control
- B22F10/36—Process control of energy beam parameters
- B22F10/366—Scanning parameters, e.g. hatch distance or scanning strategy
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22F—WORKING METALLIC POWDER; MANUFACTURE OF ARTICLES FROM METALLIC POWDER; MAKING METALLIC POWDER; APPARATUS OR DEVICES SPECIALLY ADAPTED FOR METALLIC POWDER
- B22F10/00—Additive manufacturing of workpieces or articles from metallic powder
- B22F10/80—Data acquisition or data processing
- B22F10/85—Data acquisition or data processing for controlling or regulating additive manufacturing processes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22F—WORKING METALLIC POWDER; MANUFACTURE OF ARTICLES FROM METALLIC POWDER; MAKING METALLIC POWDER; APPARATUS OR DEVICES SPECIALLY ADAPTED FOR METALLIC POWDER
- B22F12/00—Apparatus or devices specially adapted for additive manufacturing; Auxiliary means for additive manufacturing; Combinations of additive manufacturing apparatus or devices with other processing apparatus or devices
- B22F12/90—Means for process control, e.g. cameras or sensors
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B33—ADDITIVE MANUFACTURING TECHNOLOGY
- B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- B33Y10/00—Processes of additive manufacturing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B33—ADDITIVE MANUFACTURING TECHNOLOGY
- B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- B33Y50/00—Data acquisition or data processing for additive manufacturing
- B33Y50/02—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/4097—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by using design data to control NC machines, e.g. CAD/CAM
- G05B19/4099—Surface or curve machining, making 3D objects, e.g. desktop manufacturing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K9/00—Arc welding or cutting
- B23K9/04—Welding for other purposes than joining, e.g. built-up welding
- B23K9/042—Built-up welding on planar surfaces
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/49—Nc machine tool, till multiple
- G05B2219/49015—Wire, strang laying, deposit fluid, welding, adhesive, hardening, solidification, fuse
Definitions
- the present invention relates to a method and apparatus for manufacturing a laminate-molded article, a control support apparatus, and a program.
- a 3D printer that laminates metal materials uses a heat source such as a laser or an arc, for example, to melt and solidify metal powder or metal wires and laminate weld metal (weld beads) to create a desired shape.
- a heat source such as a laser or an arc
- weld metal laminate weld metal
- Patent Document 1 a laminate-molded device that measures the height at the measurement position of the laminate-molded object already formed on the workpiece, and controls the processing conditions for newly laminating the weld metal according to the height of the measurement position.
- the shape index such as the layered height from the viewpoint of maintaining the quality of the modeled product and controlling the device.
- a laser sensor can be used for height control, but when molding a part with a complicated shape or a part composed of a plurality of objects, height control for each part is required, which complicates the process.
- the modeled shape may locally deviate from the expected shape. In that case, it is required to compensate for the shape deviation.
- the object of the present invention is to provide a method and apparatus for manufacturing a layered product, a control support device, and a program that enable the manufacture of a highly accurate layered product by integrating the measurement results obtained from the measuring instrument and the state transition extracted from the trajectory plan, obtaining a more accurate shape index and updating the trajectory plan, or by repeating the lamination of weld beads while constantly managing the shape index.
- the present invention consists of the following configurations.
- (1) A manufacturing method for a laminate-molded product in which weld beads formed by melting and solidifying a filler material by a molding device are repeatedly laminated based on a predetermined trajectory plan, A step of setting an observation model obtained from measurement information collected from at least one or more measuring instruments of the modeling apparatus, and a state transition model representing a state transition of the shape index of the laminate-molded object extracted from the trajectory plan; extracting the observed physical quantity of the shape index from the observation model and extracting the state physical quantity of the shape index from the state transition model; obtaining a state transition estimated value of the shape index by integrating the observed physical quantity and the state physical quantity; updating the weld bead forming condition defined by the trajectory plan according to the state transition estimated value;
- a method of manufacturing a laminate-molded article comprising: (2) A manufacturing apparatus for a laminate-molded product that repeatedly laminates weld beads formed by melting and solidifying a filler material based on a predetermined trajectory plan, a model setting unit that sets
- a control support device that supports control of the manufacturing apparatus for a laminate-molded product according to (2),
- the observation model includes point cloud data collected from a sensor that measures a shape index of the laminate-manufactured article and representing the measured shape index of the laminate-manufactured article;
- the state transition model includes planned shape data representing a shape index of the laminate-manufactured object based on the trajectory plan, a data acquisition unit that acquires the point cloud data from the observation model and acquires the planned shape data from the state transition model; a calculation unit that divides the obtained point cloud data into a plurality of unit blocks, divides the obtained plan shape data into regions corresponding to the unit blocks, and calculates a difference value for each unit block between the point cloud data and the plan shape data; an extraction unit for extracting a unitary domain for which the difference value is greater than a predetermined threshold among the plurality of unitary domains; an information output unit that outputs information on the extracted unitary domain and the difference value corresponding to the unitary domain;
- the present invention by integrating the measurement results obtained from the measuring instrument and the state transition extracted from the trajectory plan, it is possible to obtain a more accurate shape index and update the trajectory plan, or to manufacture a highly accurate additive manufacturing product by repeating the lamination of weld beads while constantly managing the shape index.
- FIG. 1 is a schematic overall configuration diagram of a modeling apparatus.
- FIG. 2 is a schematic block diagram of the control unit.
- FIG. 3 is a schematic diagram showing a welding torch, a laser sensor, and a camera.
- FIG. 4 is an image captured by a camera of a molten pool formed at the tip of the filler material.
- FIG. 5 is an explanatory diagram showing an example of a trajectory of a welding torch when forming a weld bead.
- FIG. 6 is a flowchart showing a procedure executed by the control unit to estimate a shape index from information measured by a plurality of measuring instruments and update the trajectory plan.
- FIG. 7 is an explanatory diagram showing the contents of model setting and integration processing.
- FIG. 1 is a schematic overall configuration diagram of a modeling apparatus.
- FIG. 2 is a schematic block diagram of the control unit.
- FIG. 3 is a schematic diagram showing a welding torch, a laser sensor, and a camera.
- FIG. 4 is an
- FIG. 8A is a diagram showing the relationship between the advancing direction of the welding torch and the arrangement direction of the measuring instrument on the welding torch, and is a schematic diagram when the advancing direction and the arrangement direction match.
- FIG. 8B is a diagram showing the relationship between the traveling direction of the welding torch and the arrangement direction of the measuring instrument on the welding torch, and is a schematic diagram when the traveling direction and the arrangement direction intersect.
- FIG. 9 is an explanatory diagram schematically showing changes in bead height of a weld bead.
- FIG. 10A is a diagram showing a case where the welding torch is subjected to weaving operation, and is an explanatory diagram showing a trajectory when the weaving vibration period is long.
- FIG. 10B is a diagram showing a case where the welding torch is weaving, and is an explanatory diagram showing a trajectory when the vibration period is short.
- FIG. 11 is a flow chart showing a procedure for feedback-controlling the stacking height to correct the deviation of the modeled shape from the design shape.
- FIG. 12 is a schematic explanatory diagram showing how the amount of deviation between the height measurement result and the planned height is mapped.
- 13A and 13B are explanatory diagrams showing changes in the layer-by-layer layer height of the weld bead by controlling the layer height.
- FIG. 14 is a schematic configuration diagram showing a modeling system in which the welding robot of the modeling apparatus shown in FIG. 1 is mounted on a slider device and a layered product is manufactured using a positioner device.
- FIG. 14 is a schematic configuration diagram showing a modeling system in which the welding robot of the modeling apparatus shown in FIG. 1 is mounted on a slider device and a layered product is manufactured using a positioner device.
- FIG. 15 is a schematic explanatory diagram showing a state in which the outer peripheral surface of the bead layer is developed into a plane.
- FIG. 16 is an explanatory diagram showing height information of the outer peripheral surface of the bead layer and its processing result.
- FIG. 17 is a block configuration diagram of a control support device.
- FIG. 18 is an explanatory diagram showing how the wall is formed.
- FIG. 19A is a graph showing height measurement results and height distribution from a set horizontal plane.
- FIG. 19B is a graph showing height measurement results, showing a height distribution from a set horizontal plane.
- FIG. 19C is a graph showing height measurement results and height distribution from a set horizontal plane.
- FIG. 20A is a graph showing welding speed settings for feedback control of stack height.
- FIG. 20B is a graph showing welding speed settings for feedback control of the stack height.
- FIG. 20C is a graph showing welding speed settings for feedback control of the stack height.
- FIG. 21 is a graph showing the measurement results of the height of each layer, showing the height distribution from the set horizontal plane.
- FIG. 22 is a graph showing the measurement results of the height of each layer, showing changes in the standard deviation of the height measurement values at a plurality of locations within the layer.
- FIG. 1 is a schematic overall configuration diagram of a laminate-molded article manufacturing apparatus 100 .
- a laminate-molded article manufacturing apparatus 100 (hereinafter also referred to as a modeling apparatus) includes a modeling section 11 that laminates weld beads B to manufacture a laminate-molded article W, and a control section 13 that controls each part of the modeling section 11 .
- the modeling apparatus 100 includes a welding robot 17 having a welding torch 15 on its tip axis, a robot driving section 19 that drives the welding robot 17, a filler material supply section 21 that supplies a filler material (welding wire) M to the welding torch 15, and a welding power supply section 23 that supplies welding current and welding voltage to the welding torch 15.
- the modeling apparatus 10 also includes a plurality of measuring instruments.
- measuring instruments include a laser sensor 25A and a camera 25B which are provided on the tip axis of the welding robot 17 and whose measurement area is near the tip of the welding torch 15, and a current and voltage meter 25C which measures the welding current and welding voltage in the welding power source 23.
- other measuring instruments such as an encoder and a position sensor (not shown) which are provided in the welding robot 17 and measure the position and moving speed of the welding torch 15 may be used.
- the number of measuring instruments is not limited to a plurality, and in some cases a single measuring instrument may be used.
- the laser sensor 25A, the camera 25B, and the current/voltmeter 25C are also collectively referred to as a measuring instrument 25.
- the welding torch 15 is a gas metal arc welding torch that has a shield nozzle (not shown) and is supplied with a shield gas from the shield nozzle.
- the arc welding method may be a consumable electrode method such as coated arc welding or carbon dioxide gas arc welding, or a non-consumable electrode method such as TIG welding or plasma arc welding, and is appropriately selected according to the laminate model W to be manufactured.
- a contact tip is arranged inside the shield nozzle, and the contact tip holds the filler material M to which the melting current is supplied.
- the welding torch 15 holds the filler material M and generates an arc from the tip of the filler material M in a shield gas atmosphere.
- the welding robot 17 is a multi-joint robot, and the continuously supplied filler material M is supported at the tip of the welding torch 15 attached to the tip shaft of the robot arm.
- the position and posture of the welding torch 15 can be arbitrarily three-dimensionally set within the range of degrees of freedom of the robot arm according to commands from the robot driving section 19 .
- the filler material supply unit 21 includes a reel 21a around which the filler material M is wound.
- the filler material M is sent from the filler material supply unit 21 to a feeding mechanism (not shown) attached to a robot arm or the like, and fed to the welding torch 15 while being fed forward and reverse by the feeding mechanism as necessary.
- any commercially available welding wire can be used as the filler material M.
- MAG welding and MIG welding solid wires JIS Z 3312
- arc welding flux-cored wires for mild steel
- high-strength steel and low-temperature steel JIS Z 3313
- Welding wires specified by the standard can be used.
- filler metals M such as aluminum, aluminum alloys, nickel, nickel-based alloys, etc. can be used depending on the desired properties.
- the robot drive unit 19 drives the welding robot 17 to move the welding torch 15, and melts the continuously supplied filler material M with an arc generated by the welding current and welding voltage from the welding power source unit 23.
- a modeling program based on the trajectory plan of the laminate-molded article W to be manufactured is transmitted from the control unit 13 to the robot driving unit 19 .
- the modeling program consists of a large number of instruction codes, and is created based on an appropriate algorithm according to various conditions such as the shape data (CAD data, etc.) of the layered product, the material, and the amount of heat input.
- the welding torch 15 or the laser sensor 25A and camera 25B provided on the tip side of the robot arm are driven by the welding robot 17 to move together with the welding torch 15, and the measurement direction, measurement range, etc. are set.
- the robot driving unit 19 executes the received trajectory plan, drives each unit such as the welding robot 17, the filler material supply unit 21 and the welding power supply unit 23, and forms the weld bead B according to the trajectory plan. That is, the robot driving unit 19 drives the welding robot 17 to move the welding torch 15 along the trajectory (bead forming trajectory) of the welding torch 15 set in the trajectory plan. At the same time, the filler material supply unit 21 and the welding power supply unit 23 are driven according to the set welding conditions, and the filler material M at the tip of the welding torch 15 is melted and solidified by the arc. Thereby, a weld bead B is formed on the base plate 27 along the trajectory of the welding torch 15 . The weld beads B are adjacent to each other to form a weld bead layer, and the weld bead layer of the next layer is laminated on the weld bead layer, thereby forming the desired three-dimensional shaped product W.
- FIG. 2 is a schematic block diagram of the control unit 13.
- the control unit 13 is a computer device including a processor 31 such as a CPU or MPU, a memory 33 such as a ROM or RAM, a storage unit 35 having a storage such as an SSD (Solid State Drive) or hard disk, and an input/output interface 37 .
- the input/output interface 37 is connected to the plurality of measuring instruments 25 , the filler material supply section 21 , the robot driving section 19 and the welding power supply section 23 described above.
- the control unit 13 includes a trajectory planning unit 39, a model setting unit 41, a physical quantity extraction unit 43, an integration processing unit 45, and a control condition updating unit 47, details of which will be described later.
- the trajectory planning section 39 creates a modeling program for the laminate-molded article W to be produced. That is, when data on the material, shape, and welding conditions of the laminate-molded article W to be manufactured is input to the trajectory planning unit 39, the trajectory planning unit 39 first divides the shape model of the laminate-molded article W into layers according to the input data, and determines the bead formation procedure such as the bead formation order of the weld beads and the welding conditions so that the obtained layers are filled with the weld beads. This bead formation procedure is made into a build program. There are various algorithms for determining the above-described trajectory plan, and the determination method is not limited.
- the created modeling program is stored in the storage unit 35 and output to the robot driving unit 19 when an output request is received from the robot driving unit 19 .
- trajectory planning unit 39, the model setting unit 41, the physical quantity extraction unit 43, the integration processing unit 45, and the control condition update unit 47 of the control unit 13 may be provided in another computer device separated from the modeling unit 11. In that case, it may be configured to connect to the modeling apparatus 100 from a remote location via communication means such as a network.
- FIG. 3 is a schematic diagram showing the welding torch 15, the laser sensor 25A and the camera 25B.
- a commercially available laser displacement sensor or the like can be used as the laser sensor 25A, and the reflected light of the irradiated laser light is acquired as height data.
- the laser sensor 25A can measure the shape of the existing weld bead B and the shape of the base before the bead is formed. The details of the shape measurement principle are well known and therefore omitted.
- the camera 25B is an imaging device for three-dimensional shape measurement, and can measure the shape of the weld bead B and the base from the captured image, similar to the laser sensor 25A.
- FIG. 4 is a captured image of the molten pool formed at the tip of the filler material M captured by the camera 25B.
- the image 51 captured by the camera 25B includes the arc center 53, the filler material tip 55, the molten pool tip 57, the left end of the molten pool and the right end of the molten pool (indicated by lines 59L and 59R in FIG. 4), and the position and shape of each part can be extracted by appropriate image processing. For example, information such as arc width, arc shape, weld pool position and shape can be extracted.
- the camera 25B may be a camera for photographing the state of bead formation such as a molten pool, or a camera for temperature measurement such as a thermography.
- FIG. 5 is an explanatory diagram showing an example of the trajectory of the welding torch 15 when forming the weld bead B.
- the trajectory PS1 shown in FIG. 5 is a trajectory in which the welding torch 15 advances from a position P0 on the base plate 27 toward an obstacle R such as an upright wall, passes through the position P1, then curves and reverses at the position of the obstacle R, passes through a position P3, and returns to the position P0.
- an obstacle R such as an upright wall
- the measuring instrument 25 When the welding torch 15 is moved along the trajectory PS1, the measuring instrument 25 is arranged ahead of the welding torch 15 in the moving direction on the forward path on the left side and the return path on the right side in FIG.
- the laser sensor 25A, the camera 25B, and the current/voltage meter 25C can be used for measurement without any problem. That is, from the laser sensor 25A and the camera 25B, the shape of the base on which the weld bead is to be formed can be measured satisfactorily. Further, the current and voltage meter 25C can well measure the welding current and welding voltage of the weld bead being formed.
- the moving direction ahead is out of the measurement range of the laser sensor 25A. Therefore, the laser sensor 25A cannot measure the shape ahead of the welding torch 15 in the moving direction.
- the camera 25B measurement is possible within an area in which the moving direction ahead is within the imaging range, but measurement is impossible in an area outside the imaging range.
- the welding current and the welding voltage are likely to generate noise as the welding direction changes, and the measurement accuracy decreases. Even if the directions of the measurement directions of the laser sensor 25A and the camera 25B are adjusted, the trajectory PS1 cannot be completely followed geometrically, and a situation in which measurement becomes impossible or a situation in which measurement accuracy deteriorates cannot be avoided.
- the laser sensor 25A and the camera 25B are arranged behind the welding torch 15 in the moving direction at a position P3 on the way back to the position P0 after passing through the obstacle R. Therefore, the shape ahead in the movement direction cannot be measured from the laser sensor 25A. In addition, in the camera 25B, the area outside the imaging range increases, and the measurement accuracy is lowered or the measurement becomes impossible.
- the above results are the same for the trajectory PS2 on the opposite side of the obstacle R (upper in FIG. 5), and depending on the moving position of the welding torch 15, measurement by each measuring instrument 25 may be difficult.
- the welding torch 15 may be placed in various postures during manufacturing, and weld beads may be formed from various directions, so advanced measurement management and control are required. Since the measured information is used to correct the trajectory plan for forming the weld bead, accuracy of the measured information is required.
- the shape index is estimated by a predetermined prediction method using an observation model composed of shape indexes (various physical quantities representing bead shape etc.) measured by each measuring instrument and a state transition model composed of shape indexes (physical quantities such as design values) extracted from a predetermined trajectory plan.
- FIG. 6 is a flowchart showing a procedure executed by the control unit 13 from estimating a shape index from information measured by a plurality of measuring instruments 25 to updating the trajectory plan.
- FIG. 7 is an explanatory diagram showing details of model setting and integration processing. First, with reference to FIGS. 6 and 7, outlines of the following procedures (1) to (4) will be described in order.
- the control unit 13 determines a trajectory plan for modeling a laminate-molded article to be manufactured (S1). Specifically, a drive program based on the trajectory plan is selected.
- the model setting unit 41 (Fig. 2) of the control unit 13 sets a model corresponding to the determined trajectory plan (driving program) and the device configuration of the modeling unit 11 (S2).
- the model here is an observation model set from measurement information collected from a plurality of measuring instruments 25 of the modeling apparatus 100, and a state transition model set from state transition information of the shape index of the laminate-molded object extracted from the trajectory plan.
- the measuring instrument 25 may include a simulation model that is set by obtaining information that is difficult to directly measure by simulation.
- a model is set according to each measurement target by the laser sensor 25A, camera 25B, and current/voltmeter 25C shown in FIG.
- a model may be set using the encoder, position sensor, thermo camera, etc. of the welding robot 17 .
- the observation model here is a shape index of the weld bead B to be formed, for example, parameters such as the base surface height before bead lamination, bead height, bead width, bead profile, and weld cross-sectional area, and parameters such as the arc center, arc width, arc shape, filler material tip position, and weld pool end position (these parameters are also referred to as observation physical quantities) obtained from the captured image of the molten pool.
- a state transition model is a model created based on the information contained in the trajectory plan, and the shape index is related by the time (timing), position, or weld bead formation conditions.
- the trajectory plan includes parameters such as the trajectory of the welding torch 15, the torch movement speed (welding speed), the torch tip position (target position), the torch posture, the welding current, the welding voltage, the filler material supply speed, the number of trajectories (passes), the length of the passes, and the intervals between passes, as well as shape indices (also referred to as state physical quantities) such as the planned height of the weld bead, the planned bead width, and the planned welding amount. From the trajectory plan information, it is possible to determine the transition of the relationship between various shape indexes and weld bead formation conditions at a specific position or at a specific timing after the start of molding, and a model is created based on the obtained transition information.
- a simulation model is a model that is created by analytically obtaining information that is difficult to measure directly with a measuring instrument, such as the amount of overlap between beads and parameters (also called analytical physical quantities) such as the temperature of the molten pool (molten metal) during molding. Specifically, calculations such as thermoelastic analysis and thermoelastic-plastic analysis are performed from the trajectory plan, the physical properties of the bead material, the temperature at the time of molding, etc., and the above-described parameters that are difficult to directly measure are predicted. Using a simulation model can be expected to improve the accuracy of estimating a shape index, which will be described later. For the simulation, commercially available software having a function of reproducing the state of the laminate-molded article such as heat transfer analysis or deformation analysis may be used.
- control unit 13 executes the molding program to drive each part of the molding unit 11, thereby starting layered manufacturing (S3).
- S3 layered manufacturing
- measurement information is input from each measuring device 25 to the control unit 13 .
- This measurement information is sent to the physical quantity extraction unit 43 (FIG. 2).
- the physical quantity extraction unit 43 of the control unit 13 extracts the observed physical quantity based on the observation model and the state physical quantity based on the state transition model (S4). Then, as shown in FIG. 7, the integrated processing unit 45 obtains the state transition estimated value of the shape index by integrating the extracted observed physical quantity and state physical quantity (S5). At this time, the physical quantities analyzed by the simulation model can also be integrated. Known methods such as Kalman filter, Bayes filter, machine learning, and the like are used as a method of calculating and integrating physical quantities from each model to estimate a shape index.
- the Kalman filter is a method of estimating more accurate information using multiple pieces of uncertain information, and uses Kalman gain to control how much importance is placed on which piece of information among the pieces of information.
- Bayesian filtering is a technique based on Basis' theorem, which computes probabilities derived from conditional probabilities and multiplication theorems. From the prior probability, when a certain result is observed, the probability of which cause of the prior probability is the cause can be known, so the state transition estimated value can be derived.
- Machine learning is a method of predicting the future (calculating a state transition estimated value) by finding rules from a state transition model and an observation model.
- a learning model may be configured with input as observation model parameters and state transition model parameters, and output as a shape index.
- the obtained state transition estimated values of the shape index are, for example, the values of the height of the weld bead, the width of the weld bead, the welding current, the welding voltage, etc., and are substitute values for the values measured by the laser sensor, camera, and current/voltmeter shown in FIG. 5 above.
- the control condition update unit 47 of the control unit 13 checks the difference between the obtained state transition estimated value of the shape index and the design shape and design conditions based on the trajectory plan. If the difference is equal to or greater than a predetermined threshold value, the weld bead formation conditions are updated. For example, the modeling program that is being executed or the modeling program scheduled to be executed next time is changed. In this way, the trajectory plan is updated so that the laminate-molded article has the planned design shape and design conditions (S6). Then, the steps S2 to S6 described above are repeated until the modeling is completed (S7).
- the shape index includes, for example, the bead height, bead width, etc. of the weld bead, as well as welding conditions and the like.
- An outline of the integration process is as follows. First, the shape index h of the laminate-molded article to be manufactured is determined, and a trajectory plan is created based on the determined shape index h. Then, the welding bead B is formed by the modeling apparatus 100 based on the created trajectory plan. In the process of forming the weld bead B, the measuring instrument 25 measures the shape index of the weld bead B to set an observation model. On the other hand, a state transition model is set from the shape indices extracted from the trajectory plan. Then, the observation model and the state transition model are associated using a predetermined prediction method to estimate the actual shape index (the actual shape of the weld bead, the performed welding conditions, etc.).
- the trajectory plan is updated so as not to deviate from the initial trajectory plan according to the estimated shape index.
- a weld bead B is formed based on the trajectory plan thus updated. By repeating this in time series to narrow the width of the normal distribution of the shape error (deviation), for example, the weld bead B can be brought closer to the designed bead height.
- Equation (1) the shape index h k at a certain point k during the manufacturing of the laminate is expressed by Equation (1).
- hk +1 hk + Vk (1)
- the shape index h k+1 at the next time k+1 is obtained by adding the noise component V k of the state transition model to the shape index h k at time k.
- a known error model may be applied to the noise component Vk .
- Equation (2) the shape index h k at a certain time point k is expressed by Equation (2).
- zk hk + Wk (2)
- zk is the measured value obtained from the meter 25;
- the noise component Wk of the observation model is added to the measured value zk at a certain time point k.
- a known error model may be applied to the noise component Wk .
- Observation models are distinguished for each measuring instrument. For example, an observation model using the bead width measurement result (z k, L ) by the laser sensor 25A is represented by Equation (3), and an observation model using the bead width measurement result (z k, C ) by the camera 25B is represented by Equation (4).
- Wk ,L is the noise component of the laser sensor 25A
- Wk ,c is the noise component of the camera 25B.
- the predicted shape index is obtained from each of the multiple observation models (measurement instruments 25), and it is necessary to collect these multiple shape indices. Therefore, we integrate these independent observation models. It should be noted that the state transition model also needs to be summarized, but the integration procedure can be performed basically in the same manner as the observation model integration procedure described below.
- each observation model is linear
- the constant determined by the observation means i is denoted by C i .
- Equation (5) The relationship between the observed value zi and the desired shape index h can be expressed by Equation (5).
- z i C i h + W i (5)
- W i is a noise component possessed by the observation means i, for example, an error based on a normal distribution (variance of W i : ⁇ i 2 ).
- the shape index h can be obtained directly from one observation means. However, if this is not the case, it can be estimated from multiple observation models as follows.
- the weighting coefficient ai can be expressed by Equation (7).
- a i C i 2 ⁇ i ⁇ 2 /( ⁇ j C j 2 ⁇ j ⁇ 2 ) (7)
- the nonlinear case can be solved in the same way as the linear case by, for example, Taylor expansion of the model and linear approximation.
- the magnitude relationship of the weighting coefficients ai may be determined based on the bead formation direction and formation position of the weld bead, or the rod operation and rod operation attitude of the welding torch 15 .
- FIG. 8A and 8B are diagrams showing the relationship between the traveling direction WD of the welding torch 15 and the arrangement direction of the measuring device (laser sensor 25A) 25 on the welding torch 15.
- FIG. 8A is a schematic diagram when the traveling direction WD and the arrangement direction match.
- FIG. 8B is a schematic diagram when the traveling direction WD and the arrangement direction intersect.
- the welding torch 15 takes various postures depending on obstacles around the torch, restrictions such as the torch trajectory, and the like.
- the laser sensor 25A provided in the welding torch 15 may be arranged in front of the traveling direction WD shown in FIG.
- the measurement range of the laser sensor 25A does not include the entire substrate on which the bead is to be formed. Therefore, in the case shown in FIG. 8B, there is a high possibility that the measurement accuracy will deteriorate, so the weighting coefficients ai are made smaller than in the case shown in FIG. 8A.
- the weighting coefficient ai of the observation model is decreased as the angle ⁇ formed between the traveling direction WD and the arrangement direction of the measuring device 25 increases.
- FIG. 9 is an explanatory diagram schematically showing changes in the bead height of the weld bead B.
- the bead height is more likely to change between the start and end of the bead formation compared to the stable portion between the start and end due to changes in the arc.
- the bead height at the leading end is greater than that at the stable portion, and the bead height at the trailing end is smaller than that at the stable portion. Therefore, since the measured values tend to be unstable at the start and end of bead formation, the weighting coefficient ai is set smaller than that at the stable portion.
- the shape at the start and end portions of the weld bead B does not become stationary like the stable portion, but even if such a curved change in shape occurs, a model that conforms to the actual bead shape of the weld bead B can be set by appropriately adjusting the weighting coefficients ai . It is also possible to distinguish between the linearly formed area and the curvedly formed area of the weld bead B, and set the weighting coefficients ai appropriately for these areas.
- a model in which the bead shape changes steadily and a model in which the bead shape changes curvilinearly can be combined on the same trajectory (same path).
- FIGS. 10A and 10B are diagrams showing a case where the welding torch 15 is subjected to weaving operation
- FIG. 10A is an explanatory diagram showing a trajectory when the weaving vibration period is long.
- FIG. 10B is an explanatory diagram showing a trajectory when the vibration period is short.
- the noise variance ⁇ i 2 of each model may be set as a reference (initial value).
- the state transition estimated value h Est of the shape index can be expressed by equation (8).
- the weighting factor a i set based on the above-described noise variance ⁇ i 2 may be used as an initial value, and the weighting factor a i may be adjusted as the modeling progresses.
- the variance ⁇ i 2 of noise is used in this way, it is possible to consider the unique characteristics of each measuring instrument 25 and the state transition model and reflect them in the setting of integration conditions.
- Equation (9) can be referred to.
- h Est,k+1 h Est,k + K k+1 ⁇ z i,k+1 ⁇ C i (h Est,k ) ⁇ (9) where Kk is the Kalman gain.
- the state transition estimated value of the shape index is updated by the amount obtained by adjusting the difference between the actual observed value and the predicted observed value using the Kalman gain.
- the state transition estimated value h Est,k once output is saved as information possessed by the trajectory plan, and by using the saved state transition estimated value each time the state transition estimated value is updated, the latest state transition estimated value and the transition of the state transition estimated value can be always grasped during modeling.
- the storage unit 35 may be used to store the state transition estimated value.
- the control condition updating unit 47 shown in FIG. 2 described above performs the shape control of the laminate-molded article based on the state transition estimated value h Est of the shape index obtained by the integrated processing in this way. According to this, by comparing the state transition estimated value and the design value and adjusting the welding bead formation conditions in accordance with the difference, it is possible to perform modeling while correcting in real time. Alternatively, the trajectory plan may be updated to change, add, or delete trajectories (also referred to as paths), and corresponding control instructions (driving programs) may be generated. If there is a large difference between the state transition estimated value and the design value, the bead formation conditions may be changed significantly and the control may become unstable.
- the design values such as the target height, target width, and target welding volume of the weld bead may be changed.
- the difference between the state transition estimated value and the design value may be compared with a predetermined threshold, and if the difference exceeds the threshold, the design value may be corrected.
- a state transition model derived from the trajectory plan and an observation model based on the measurement information of the measuring instrument 25 it is preferable to use a state transition model derived from the trajectory plan and an observation model based on the measurement information of the measuring instrument 25 .
- the observation model and the state transition model it becomes possible to always accurately estimate the shape index such as the bead height of the weld bead, which may be difficult to measure stably.
- FIG. 11 is a flow chart showing a procedure for feedback-controlling the stacking height to correct the deviation of the modeled shape from the design shape.
- the molding apparatus 100 shown in FIG. 1 reads a drive program based on a set trajectory plan, drives the welding robot 17 and the like according to the drive program, and stacks the weld beads B (S11).
- the laser sensor 25A scans and measures the stacked height of the modeled object. Then, information on the height distribution of the bead layer collected by this measurement is acquired as point cloud data (S12).
- the scanning of the laser sensor 25A may be performed at the same time as the welding bead B is formed by driving the welding robot 17, but it is preferable to perform the scanning after the welding bead B is formed. In other words, it is preferable to measure the shape of the modeled object based on a trajectory created separately from the trajectory (path) for forming the weld bead B.
- This point cloud data is height information of a large number of measurement points measured by the laser sensor 25A at arbitrary sampling intervals.
- the drive cycle for driving the welding robot 17 and the measurement cycle for the laser sensor 25A are different, it is preferable to interpolate the driving of the welding robot 17 in accordance with the measurement timing for the laser sensor 25A.
- the planned height (lamination plan data) corresponding to each measurement position of the point cloud data obtained by measuring the height of the bead layer is obtained from the trajectory plan. Then, using the information of the point cloud data (observation model) and the planned height (state transition model), the difference between the measured height (actual height) and the planned height at each measurement position (hereinafter also referred to as the amount of deviation) is calculated (S13).
- the measured height at each measurement position is the average value of the heights of the unitary divisions (for example, rectangular, polygonal, circular or elliptical fixed range) on the height measurement surface of the bead layer.
- the size of this unitary domain is arbitrary.
- a square with a side of 5 mm obtained by dividing the height measurement surface into a lattice is defined as a unitary domain, and the average value of the heights in the unitary domain is represented as point cloud data as the height of a representative point at the center of the square.
- the unitary domain is preferably divided along the direction of weld bead formation.
- the height information output from the laser sensor 25A is output in the coordinate system of the laser sensor 25A, it is preferable to convert the output height information into the robot coordinate system of the welding robot 17. Further, when a positioner device and a slider device, which will be described later, are provided, it is desirable to convert the operation coordinate system of each device into the robot coordinate system of the welding robot 17 and unify them into one of the coordinate systems. This makes it easier to grasp the tendency of each information.
- the layered plan data described above is divided into a plurality of areas corresponding to the unitary sections, and the average value of the planned height within each area, that is, within the unitary section is obtained for each unitary section.
- FIG. 12 is a schematic explanatory diagram showing how the amount of deviation between the height measurement result and the planned height is mapped.
- the amount of deviation between the obtained average value of the measured height in each unitary section UC and the average value of the planned height of the unitary section UC is calculated.
- a deviation amount map is created in which the deviation amount is registered for each representative point Pk, with the central position of each square that is the unitary domain UC as the representative point Pk.
- a unitary domain UC whose deviation amount is greater than a predetermined threshold value is extracted (S14).
- the formation conditions (information included in the trajectory plan) of the weld bead to be laminated next are updated (S15). In this way, since the trajectory plan is selectively updated and the shape is adjusted for the unitary domain UC with a large amount of deviation, the control can be limited only to the locations where the height control is required, and the control itself can be kept stable.
- FIG. 13 is an explanatory diagram showing the transition of the lamination height of the weld bead for each layer by controlling the lamination height. For example, if the measured height (actual height) of the bead layer of the mth layer (m is an integer) is lower than the planned height, the target height of the next layer (m+1th layer) may be set higher than the planned height of the next layer. However, if the difference between the actual height and the planned height exceeds a predetermined limit value, the planned height may not be reached even if the target height is adjusted for only one next layer. In that case, the target height may be adjusted over multiple layers to be formed later.
- the target height is set to a height that is increased by ⁇ H1 from the actual height
- the m+2nd layer it is set to a height that is increased by an even greater ⁇ H2 from the actual height for the m+1st layer.
- the actual height on the m+2 layer plane easily reaches the planned height, and the difference ⁇ between the actual height and the planned height can be greatly reduced.
- the procedure of feedback control of the stacking height shown here is an example, and the present invention is not limited to this.
- FIG. 14 is a schematic configuration diagram showing a modeling system 200 that mounts the welding robot 17 of the modeling apparatus 100 shown in FIG.
- a modeling system 200 includes the modeling apparatus 100 shown in FIG.
- the slider device 60 includes a moving table 61 on which the welding robot 17 is placed, and a horizontal moving mechanism 63 that moves the moving table 61 along the horizontal XY plane.
- the horizontal movement mechanism 63 includes an X-axis rail 65 and a Y-axis rail 67, and moves the moving table 61 in the X-axis and Y-axis directions by a drive source such as a motor (not shown).
- the positioner device 70 includes a base 71 and a cylindrical rotating body 73 whose both end shafts are supported by the base 71 .
- the rotating body 73 is driven to rotate about its axis while being supported by the base 71 .
- a base material (not shown) is fixed to the surface of the rotating body 73, and a laminate-molded article W is formed by laminating a weld bead on the base material.
- the modeling system 200 shown in FIG. 14 was used to model a cylindrical shaped object, and the effect of feedback control of the layer height was verified.
- Formation of the weld bead and scanning by the laser sensor 25A were performed while the positioner device 70 was being rotated.
- the scanning of the laser sensor 25A was performed while synchronizing with the measurement timing of the laser sensor 25A itself. Therefore, the measurement data, which is the measurement result, includes point cloud data representing the radial distance r (height information) from the center of the rotating body 73, robot coordinates (x, y, z) at the measurement position, and information such as the rotation angle ( ⁇ ) of the positioner device 70.
- the point cloud data of the measured height information is converted into the robot coordinate system using the robot coordinates and the rotation angle information of the positioner device 70, and the height distribution of the cylindrical modeled object with a central angle of 360° is converted into developed view information. That is, the process of synchronizing the timing of driving the robot with the measurement period of the laser sensor 25A and the process of unifying the drive command signal of each part in the robot coordinate system of the welding robot 17 are performed.
- the welding torch 15 is positioned at a desired position by driving the slider device 60 and the welding robot 17 shown in FIG. Then, by moving the welding torch 15 in the Y direction by driving the welding robot 17, a ring-shaped layered product W that is layered in an annular or spiral shape is obtained. Further, the welding torch 15 or the laser sensor 25A provided near the torch is positioned in the same manner as described above, and the laser sensor 25A scans the outer peripheral surface (surface of the bead layer) of the laminate-molded article W while rotationally driving the rotating body 73 of the positioner device 70. Thereby, the height distribution of the formed bead layer is obtained.
- FIG. 15 is a schematic explanatory diagram showing a state in which the outer peripheral surface of the bead layer is developed into a plane.
- the developed view information 80 obtained by converting the 360° surface shape of the outer peripheral surface of the formed bead layer into the robot coordinate system will be used.
- FIG. 16 is an explanatory diagram showing height information of the outer peripheral surface of the bead layer and its processing result.
- the development view information 80A shown in A of FIG. 16 is a contour map showing the height distribution (point cloud data) of the outer peripheral surface of the bead layer, and the longer (higher) or shorter (lower) radial distance r from the central axis of the modeled object is expressed by the level of density.
- a region with a higher concentration has a shorter radial distance r (lower height), and a region with a higher concentration has a longer radial distance r (higher height).
- the representative points Pk of the aforementioned unitary domains are shown as dots. According to FIG. 16A, the areas Ar1 and Ar2 of the developed view information 80A are particularly lower in height than the surroundings.
- the development view information 80B shown in B of FIG. 16 is the result of obtaining the amount of deviation between the value Hm1 obtained by averaging the height distribution (point cloud data) shown in A for each unitary section UC centered on each representative point Pk and the value Hm2 obtained by averaging the planned height corresponding to each unitary section UC for each unitary section UC.
- a unitary domain UC with a lower deviation amount than the surroundings is recognized.
- the developed view information 80C shown in FIG. 16C is the result of extracting the unitary domain UC whose deviation amount is larger than the predetermined threshold value from the unitary domain UCs shown in FIG. 16B.
- a unitary domain UC having a deviation amount of 2mm or more in a square unitary domain UC of 5 mm on a side is extracted.
- the model setting unit 41 sets the laser sensor 25A as an observation model, and sets the welding bead formation conditions according to the drive program selected as the state transition model.
- the physical quantity extraction unit 43 extracts output data (point group data) from the laser sensor 25A, and extracts the aforementioned planned height from the state transition model.
- the integrated processing unit 45 obtains the aforementioned deviation amount as the state transition estimated value of the shape index from the point cloud data and the planned height information. Furthermore, the unitary domain UC with a large amount of deviation is extracted.
- the control condition updating unit 47 updates the welding bead formation condition based on the deviation amount of the unitary domain extracted by the integration processing unit 45 .
- the deviation amount is an estimated state transition value, and at the same time, it is an index that considers the difference from the design shape and design conditions based on the trajectory plan.
- the trajectory planning section 39 then generates a driving program based on the updated trajectory planning. By executing the drive program updated in this way to form the next bead layer, the deviation from the planned height can be reduced.
- control operation described above may be performed by a control support device that controls the control unit 13 from the outside.
- the control support device 300 acquires information such as the trajectory plan, the point cloud data representing the shape, the planned shape, etc. from the control unit 13, and outputs the result of finding the region where the deviation from the plan is particularly large (the difference value between the extracted unitary domain and the unitary domain) to the control unit 13.
- the control unit 13 can update the welding bead forming condition at the position corresponding to the extracted unitary domain.
- FIG. 17 is a block configuration diagram of the control support device 300. As shown in FIG.
- the control support device 300 includes a data acquisition unit 81 , a calculation unit 83 , an extraction unit 85 , an information output unit 87 , and may further include a display unit 89 .
- the observation model set by the control unit 13 includes point cloud data collected from sensors that measure the shape indices of the laminate-manufactured article and representing the measured shape indices of the laminate-manufactured article. That is, the observation model includes point cloud data of height information measured by the laser sensor 25A.
- the state transition model also includes planned shape data representing shape indices of the laminate-manufactured object based on the trajectory plan. This planned shape data is data including a shape index planned by trajectory planning. That is, the state transition model includes, for example, information on the planned height of the weld bead.
- the data acquisition unit 81 acquires point cloud data from the observation model, and acquires stacking plan data from the state transition model.
- the calculation unit 83 divides the acquired point cloud data into a plurality of unit blocks, divides the acquired stacking plan data into regions corresponding to the unit blocks, and calculates the difference value for each unit block between the point cloud data and the stacking plan data.
- the difference value may be a deviation value obtained by averaging the values within the unitary domain.
- the extraction unit 85 extracts a unitary domain whose difference value is greater than a predetermined threshold value from among the plurality of unitary domains.
- the information output unit 87 also outputs information on the extracted unitary domain and the difference value corresponding to the unitary domain.
- the control unit 13 updates the welding bead formation conditions according to the output information.
- the display unit 89 such as a monitor
- the display unit 89 displays the output information, so that the operator can clearly grasp the area where the actual laminated shape index is significantly different from the planned shape, that is, the part to be corrected and controlled.
- FIG. 18 is an explanatory diagram showing how the wall is formed.
- modeling was started from a state in which the modeled object including the base material (base plate 27) was tilted. Also, the start and end of the weld bead B are out of control because local height fluctuations occur.
- the weld bead B of the first layer is formed without feedback control of the lamination height.
- feedback control of the stacking height was performed to observe changes in height.
- a total of 10 layers of weld beads B were laminated to form the wall.
- Feedback control of the lamination height here is control for increasing the welding speed in the region where the height is excessive and decreasing the welding speed in the region where the height is insufficient.
- FIGS. 19A to 19C are graphs showing height measurement results and height distribution from a set horizontal plane.
- 20A-20C are graphs showing welding speed settings for feedback control of stack height.
- FIG. 19A is the height distribution of the base material surface inclined from the horizontal plane.
- FIG. 19B shows the height distribution of the first-layer weld bead formed on the surface of the base material.
- FIG. 20A since the first layer does not perform height feedback control, it has an inclination substantially the same as the height distribution of the base material.
- FIG. 19C shows the height distribution of weld beads in the second layer.
- the second layer as shown in FIG. 20B, feedback control was performed to increase the welding speed in the region where the height was excessive and to decrease the welding speed in the region where the height was insufficient.
- feedback control of the lamination height of the third layer increases or decreases the welding speed so as to flatten the fine unevenness as shown in FIG. 20C.
- FIG. 21 is a graph showing the measurement results of the height of each layer and showing the height distribution from the set horizontal plane.
- the width from the maximum value to the minimum value of the stacking height is relatively large in the 0th layer and the 1st layer in which the feedback control of the stacking height is not performed, and is relatively small in the second and subsequent layers in which the feedback control is performed.
- FIG. 22 is a graph showing the height measurement results for each layer, showing changes in the standard deviation of the height measurement values at multiple locations within the layer.
- the standard deviation is relatively large in the 0th layer and the 1st layer in which the feedback control of the lamination height is not performed, but the standard deviation is greatly reduced in the 2nd layer and the subsequent layers in which the feedback control is performed.
- the layer height of the modeled object can be flattened.
- a manufacturing method for a laminate-molded product in which weld beads formed by melting and solidifying a filler material by a molding device are repeatedly laminated based on a predetermined trajectory plan, A step of setting an observation model obtained from measurement information collected from at least one or more measuring instruments of the modeling apparatus, and a state transition model representing a state transition of the shape index of the laminate-molded object extracted from the trajectory plan; extracting the observed physical quantity of the shape index from the observation model and extracting the state physical quantity of the shape index from the state transition model; obtaining a state transition estimated value of the shape index by integrating the observed physical quantity and the state physical quantity; updating the weld bead forming condition defined by the trajectory plan according to the state transition estimated value; A method of manufacturing a laminate-molded article comprising: According to this laminate-molded article manufacturing method, the shape index can be estimated by associating the state transition model and the observation model.
- the trajectory plan can be updated based on the more accurate state transition estimated value obtained by the integrated processing at any point of the laminate-molded article, and highly accurate molding control of the laminate-molded article can be performed.
- the estimation accuracy can be improved because the orbit planning information is taken into account compared to the case of simple observation model calculations.
- the shape index can be estimated with higher accuracy by using analytical physical quantities calculated by simulation of information such as the amount of overlap between weld beads that is difficult to directly observe with a measuring instrument, the temperature of the molten pool during lamination, and the like.
- the weighting factor can be determined in response to the generation of noise signals due to the movement of the rod or the change in the measurement environment of the measuring instrument due to the attitude of the rod, so that the shape index can be estimated stably and accurately at all times.
- this method for manufacturing a laminate-molded article it is possible to model the shape of the actually formed weld bead in accordance with the tendency thereof. For example, it is possible to model a linear part and a curved part separately, and control according to the tendency of state transition is possible.
- the observation model includes point cloud data collected from a sensor that measures the shape index of the laminate-manufactured article and representing the measured shape index of the laminate-manufactured article
- the state transition model includes planned shape data representing a shape index of the laminate-manufactured object based on the trajectory plan,
- the method of manufacturing a laminate-molded article according to (1) wherein the point cloud data is divided into a plurality of unit blocks, the planned shape data is divided into regions corresponding to the respective unit blocks, and a difference value is obtained by comparing the point cloud data and the planned shape data for each unit block, and the welding bead forming conditions are updated according to the difference value.
- this method for manufacturing a laminate-molded article it is possible to update the welding bead formation conditions for each unit section according to the difference value obtained by comparing the measured point cloud data and the planned shape data. As a result, shape deviation from the plan can be suppressed, and it is possible to manufacture laminate-molded objects with high accuracy.
- a manufacturing apparatus for a laminate-molded product that repeatedly laminates weld beads formed by melting and solidifying a filler material based on a predetermined trajectory plan, a model setting unit that sets an observation model obtained from measurement information collected from at least one or more measuring instruments, and a state transition model that represents the state transition of the shape index of the laminate-molded article extracted from the trajectory plan; a physical quantity extraction unit that extracts the observed physical quantity of the shape index from the observation model and extracts the state physical quantity of the shape index from the state transition model; an integration processing unit that integrates the observed physical quantity and the state physical quantity to obtain a state transition estimated value of the shape index; a control condition updating unit that updates the welding bead formation condition determined by the trajectory plan according to the state transition estimated value; A manufacturing apparatus for a laminate-molded product.
- the shape index can be estimated by associating the state transition model and the observation model.
- the trajectory plan can be updated based on the more accurate state transition estimated value obtained by the integrated processing at any point of the laminate-molded article, and highly accurate molding control of the laminate-molded article can be performed.
- the estimation accuracy can be improved because the orbit planning information is taken into account compared to the case of simple observation model calculations.
- a control support device that supports control of the manufacturing apparatus for a laminate-molded product according to (13),
- the observation model includes point cloud data collected from a sensor that measures a shape index of the laminate-manufactured article and representing the measured shape index of the laminate-manufactured article;
- the state transition model includes planned shape data representing a shape index of the laminate-manufactured object based on the trajectory plan, a data acquisition unit that acquires the point cloud data from the observation model and acquires the planned shape data from the state transition model; a calculation unit that divides the obtained point cloud data into a plurality of unit blocks, divides the obtained plan shape data into regions corresponding to the unit blocks, and calculates a difference value for each unit block between the point cloud data and the plan shape data; an extraction unit for extracting a unitary domain for which the difference value is greater than a predetermined threshold among the plurality of unitary domains; an information output unit that outputs information on the extracted unitary domain and the difference value corresponding to the unitary domain;
- control support device further comprising a display section for displaying output information from the information output section.
- the display on the display unit makes it easier for the operator to visually manage the parts to be controlled.
- the trajectory plan can be updated based on the more accurate state transition estimated value obtained by the integrated processing at any point of the laminate-molded article, and highly accurate molding control of the laminate-molded article can be performed.
- the estimation accuracy can be improved because the orbit planning information is taken into account compared to the case of simple observation model calculations.
- control unit 15 welding torch 17 welding robot 19 robot drive unit 21 filler material supply unit 21a reel 23 welding power supply unit 25 measuring instrument 25A laser sensor 25B camera 25C current-voltage meter 27 base plate 31 processor 33 memory 35 storage unit 37 input/output interface 39 trajectory planning unit 41 model setting unit 4 3 physical quantity extraction unit 45 integration processing unit 47 control condition update unit 51 captured image 53 arc center 55 filler metal tip 57 molten pool tip 59L molten pool left end line 59R molten pool right end line 60 slider device 61 movement table 63 horizontal movement mechanism 65 X-axis rail 67 Y-axis rail 70 positioner device 71 base 73 rotating body 80, 80A, 80B, 80C Development view 81 Data acquisition unit 83 Calculation unit 85 Extraction unit 87 Information output unit 89 Display unit 100 Modeling device (manufacturing device for layered product) 200 modeling system 300 control support device B welding bead M filler material R obstacle W layered product
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Materials Engineering (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Optics & Photonics (AREA)
- Plasma & Fusion (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Human Computer Interaction (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
Description
しかし、計測器が溶接トーチに搭載された構成では、積層造形時の溶接トーチの姿勢、溶接方向等によってはトーチ周囲に存在する障害物が計測の障害となり、計測器による計測が実施できなかったり、実施できたとしても計測データの信頼性が低下したりすることがある。
(1) 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造方法であって、
前記造形装置が有する少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定する工程と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する工程と、
前記観測物理量と前記状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める工程と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する工程と、
を含む積層造形物の製造方法。
(2) 溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造装置であって、
少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定するモデル設定部と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する物理量抽出部と、
前記観測物理量と前記状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める統合処理部と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する制御条件更新部と、
を備える積層造形物の製造装置。
(3) (2)に記載の積層造形物の製造装置の制御を支援する制御支援装置であって、
前記観測モデルは、前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含み、
前記状態遷移モデルは、前記軌道計画に基づく前記積層造形物の形状指標を表す計画形状データを含み、
前記観測モデルから前記点群データを取得し、前記状態遷移モデルから前記計画形状データを取得するデータ取得部と、
取得した前記点群データを複数の単位区画に分割し、取得した前記計画形状データを前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとの前記単位区画ごとの差分値を算出する演算部と、
複数の前記単位区画のうち前記差分値が予め定めた閾値より大きい単位区画を抽出する抽出部と、
抽出した前記単位区画及び当該単位区画に対応する前記差分値の情報を出力する情報出力部と、
を備える制御支援装置。
(4) 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造手順を、コンピュータに実行させるプログラムであって、
前記コンピュータに、
前記造形装置が有する少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定する機能と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する機能と、
前記観測物理量と状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める機能と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する機能と、
を実現させるプログラム。
(5) 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造手順を、コンピュータに実行させるプログラムであって、
前記コンピュータに、
前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含む観測モデルと、前記軌道計画に基づく前記積層造形物の前記形状指標を表す計画形状データを含み、前記形状指標の状態遷移を表す状態遷移モデルとを設定する機能と、
前記観測モデルから前記点群データを取得し、前記状態遷移モデルから前記計画形状データを取得する機能と、
取得した前記点群データを複数の単位区画に分割し、前記計画形状データを前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとの前記単位区画ごとの差分値を算出する機能と、
算出した前記単位区画及び当該単位区画に対応する前記差分値の情報を出力する機能と、
を実現させるプログラム。
<造形装置の構成>
図1は、積層造形物の製造装置100の概略的な全体構成図である。
積層造形物の製造装置100(以下、造形装置ともいう。)は、溶接ビードBを積層して積層造形物Wを製造する造形部11と、造形部11の各部を制御する制御部13とを備える。
制御部13は、CPU、MPU等のプロセッサ31と、ROM、RAM等のメモリ33と、SSD(Solid State Drive)、ハードディスク等のストレージを有する記憶部35と、入出力インターフェイス37とを備えるコンピュータ装置である。入出力インターフェイス37には、前述した複数の計測器25、溶加材供給部21、ロボット駆動部19、溶接電源部23が接続される。制御部13は、詳細を後述する軌道計画部39と、モデル設定部41と、物理量抽出部43と、統合処理部45と、制御条件更新部47とを備える。
図3は、溶接トーチ15とレーザセンサ25A及びカメラ25Bとを示す模式図である。
レーザセンサ25Aは、市販のレーザ変位センサ等を使用でき、照射したレーザ光の反射光を高さデータとして取得する。これにより、レーザセンサ25Aは、既設の溶接ビードBの形状、及びビード形成前の下地形状が計測可能となる。形状の計測原理についての詳細は公知であるため省略する。
カメラ25Bによる撮像画像51には、アーク中心53、溶加材先端55、溶融池先端57、溶融池左端及び溶融池右端(図4に線59L,線59Rで示す)が含まれており、適宜な画像処理により各部の位置、形状を抽出できる。例えば、アークの幅、アークの形状、溶融池の位置及び形状等の情報が抽出可能となる。カメラ25Bは、これ以外にも、溶融池等のビード形成の様子を撮影するカメラ、サーモグラフィ等の温度測定用のカメラであってもよい。
図5は、溶接ビードBを形成する際の溶接トーチ15の軌道の一例を示す説明図である。
図5に示す軌道PS1は、溶接トーチ15がベースプレート27上の位置P0から、起立した壁部等の障害物Rに向かって進行し、位置P1を通過した後、障害物Rの位置で軌道を湾曲させて反転し、位置P3を通過して位置P0に戻る、という軌道である。この軌道PS1に沿って溶接ビードを形成する場合を考える。
位置P0から障害物Rまでの間の位置P1では、レーザセンサ25Aとカメラ25Bと電流電圧計25Cによる計測は支障なく可能な状態となる。つまり、レーザセンサ25A及びカメラ25Bからは、これから溶接ビードを形成する下地の形状を良好に計測できる。また、電流電圧計25Cからは、形成中の溶接ビードの溶接電流及び溶接電圧を良好に計測できる。
次に、上記した形状指標を推定する手順について説明する。
図6は、制御部13が実行する、複数の計測器25による計測情報から形状指標を推定して、軌道計画を更新するまでの手順を示すフローチャートである。また、図7は、モデル設定と統合処理の内容を示す説明図である。まず、図6、図7を参照しつつ、以下の各手順(1)~(4)の概要を順に説明する。
制御部13は、製造しようとする積層造形物を造形するための軌道計画を決定する(S1)。具体的には、軌道計画に基づいた駆動プログラムを選定する。
次に、制御部13のモデル設定部41(図2)は、決定した軌道計画(駆動プログラム)と、造形部11の機器構成とに対応したモデルを設定する(S2)。ここでいうモデルとは、造形装置100が有する複数の計測器25から収集された計測情報から設定される観測モデルと、軌道計画から抽出される積層造形物の形状指標の状態遷移の情報から設定される状態遷移モデルである。さらに、計測器25では直接測定することが困難な情報をシミュレーションにより求めて設定されるシミュレーションモデルを含んでいてもよい。
制御部13の物理量抽出部43は、観測モデルによる観測物理量と、状態遷移モデルによる状態物理量とを抽出する(S4)。そして、統合処理部45は、図7に示すように、抽出した観測物理量と状態物理量とを、詳細を後述する統合処理により、形状指標の状態遷移推定値を求める(S5)。このとき、シミュレーションモデルによる解析物理量を併せて統合処理することもできる。各モデルによる物理量を計算統合して形状指標を推定する手法は、例えば、カルマンフィルタ、ベイズフィルタ、機械学習、等の公知の手法が用いられる。
ベイズフィルタは、ベースの定理に基づく手法であり、条件付き確率と乗法定理から導かれる確率を計算する。事前確率から、ある結果が観測されたときに、事前確率のどの原因によるものであるかの確率が分かるため、状態遷移推定値を導き出すことができる。
機械学習は、状態遷移モデルと観測モデルから法則性を見いだして将来を予測する(状態遷移推定値を算出する)手法であり、例えば入力を観測モデルのパラメータと状態遷移モデルのパラメータ、出力を形状指標として学習モデルを構成しても良い。
制御部13の制御条件更新部47は、得られた形状指標の状態遷移推定値と、軌道計画による設計形状、設計条件との差異を確認し、差異が所定の閾値以上であれば、溶接ビードの形成条件を更新する。例えば、実行途中の造形プログラム、又は次回実行予定の造形プログラムを変更する。このようにして、積層造形物が予定通りの設計形状、設計条件となるように軌道計画を更新する(S6)。そして、上記したステップS2~S6までの工程を、造形が完了するまで繰り返す(S7)。
次に、上記した積層造形物の形状指標を、複数の計測器からの観測モデルと、軌道計画からの状態遷移モデルとを用いて統合処理する工程を説明する。形状指標としては、例えば、溶接ビードのビード高さ、ビード幅等のほか、溶接条件等も含まれる。
まず、作製しようとする積層造形物の形状指標hを決定し、決定した形状指標hに基づく軌道計画を作成する。そして、作成した軌道計画に基づいて造形装置100により溶接ビードBを形成していく。溶接ビードBの形成過程において、計測器25により溶接ビードBの形状指標を計測して観測モデルを設定する。一方で、軌道計画から抽出される形状指標から状態遷移モデルを設定する。そして、所定の予測手法を用いて観測モデルと状態遷移モデルとを関連付けて、実際の形状指標(溶接ビードの実形状、実施した溶接条件等)を推定する。推定した形状指標に応じて、当初の軌道計画とのずれが生じないように軌道計画を更新する。このようにして更新された軌道計画に基づいて溶接ビードBを形成する。これを、時系列に繰り返して形状誤差(ずれ)の正規分布の幅を狭めていくことで、例えば、溶接ビードBを、より設計通りのビード高さに近付けることができる。
状態遷移モデルにおいては、積層物を造形する途中のある時点kにおける形状指標hkは、式(1)で表される。
hk+1=hk+Vk ・・・(1)
式(1)によれば、次の時点k+1における形状指標hk+1は、時点kの形状指標hkに状態遷移モデルのノイズ成分Vkを加えたものとなる。ノイズ成分Vkについては、公知の誤差モデルを適用してもよい。
zk=hk+Wk ・・・(2)
zkは、計測器25から得られる計測値である。式(2)によれば、ある時点kにおける計測値zkには、観測モデルのノイズ成分Wkが加わる。ノイズ成分Wkについては、公知の誤差モデルを適用してもよい。観測モデルは計測器ごとに区別され、例えばレーザセンサ25Aによるビード幅の計測結果(zk,L)を用いた観測モデルは式(3)で表され、カメラ25Bによるビード幅の計測結果(zk,C)を用いた観測モデルは式(4)で表される。
zk,L=hk+Wk,L ・・・(3)
zk,c=hk+Wk,c ・・・(4)
ここで、Wk,Lは、レーザセンサ25Aのノイズ成分であり、Wk,cはカメラ25Bのノイズ成分である。
zi=Cih+Wi ・・・(5)
ここで、Wiは観測手段iが有するノイズ成分であり、例えば正規分布に基づく誤差である(Wiの分散:σi 2)。
E[hEst]=ΣiaiCih=h ・・・(6)
ai=Ci 2σi -2/(ΣjCj 2σj -2) ・・・(7)
統合するモデルがいずれも線形の場合は、制約付き最小二乗問題として公知の手法で解くことができる。非線形の場合は、モデルを例えばTaylor展開し、線形近似すれば、線形の場合と同様に解くことができる。
上記の一例として、計測器(レーザセンサ25A,カメラ25B)25の計測範囲と、溶接トーチ15の軌道との関係から、計測器25が測定不能になる場所については、重み付け係数aiを零(ai=0)とし、計測精度が低下する場所については、前回計測した場所よりも今回計測した場所での重み付け係数aiを十分に小さく設定する(ai,k<<ai,k-1)、こと等が挙げられる。
図8A、図8Bは、溶接トーチ15の進行方向WDと、溶接トーチ15への計測器(レーザセンサ25Aを例示する)25の配置方向との関係を示す図で、図8Aは、進行方向WDと配置方向とが一致している場合の概略図である。図8Bは、進行方向WDと配置方向とが交差している場合の概略図である。
溶接ビードBを形成する際、ビード形成の始端部と終端部とでは、アークの変化のため、始端部と終端部との間の安定部と比較してビード高さが変化しやすい。始端部ではビード高さが安定部より増加して、終端部ではビード高さが安定部より減少する。そこで、ビード形成の始端部と終端部については計測値が不安定になりやすいことから、重み付け係数aiを安定部よりも小さくする。
溶接トーチ15による運棒動作の振動周期が短くなるほど、溶接電流、溶接電圧の計測にノイズが入りやすくなる。そこで、ウィービング動作の振動周期が短くなるにしたがって、計測が不安定になりやすいことから、重み付け係数aiを減少させる。
hEst,k+1=hEst,k+ Kk+1{zi,k+1-Ci(hEst,k)} ・・・(9)
ここでKkはカルマンゲインである。
式(9)では、実際の観測値と予測された観測値の差分をカルマンゲインで調整した量によって、形状指標の状態遷移推定値を更新している。一度出力された状態遷移推定値hEst,kは軌道計画が有する情報として保存し、状態遷移推定値を更新する度に保存された状態遷移推定値を利用することで、最新の状態遷移推定値と状態遷移推定値の推移とを造形中常に把握することができる。状態遷移推定値の保存には記憶部35を利用してよい。
積層造形中は造形物の品質及び装置制御の観点から、造形高さ等の形状指標を管理するのが望ましい。積層造形中に造形形状が局所的に設計形状から外れた場合には、その形状ずれを補完する処理が必要となる。そこで、積層造形中に形状指標を計測し、計測した形状指標と設計形状との偏差量に応じて溶接ビードの形成条件を更新することで、形状指標のフィードバック制御を行う。これにより、次層からの造形形状が設計形状に近づくように補正される。このような造形形状の補正手順を以下に説明する。ここでは形状指標として積層高さ(ビード高さ)を例に説明するが、ビード幅等の他の指標としてもよい。
まず、図1に示す造形装置100により、設定された軌道計画に基づく駆動プログラムを読み込み、駆動プログラムに従って溶接ロボット17等を駆動して溶接ビードBを積層する(S11)。この造形により形成されるビード層ごとに、レーザセンサ25Aにより造形物の積層高さを走査(スキャン)して計測する。そして、この計測により収集されたビード層の高さ分布の情報を点群データとして取得する(S12)。なお、ここでのレーザセンサ25Aの走査は、溶接ロボット17の駆動による溶接ビードBの形成時と同時でもよいが、溶接ビードBの形成後に走査することが好ましい。つまり、溶接ビードBを形成するための軌道(パス)とは別途に作成した軌跡に基づいて、造形物形状を計測することが好ましい。この点群データとは、レーザセンサ25Aが任意のサンプリング間隔で計測した多数の計測点の高さ情報である。
次に、溶接ビードの形状指標として積層高さを計測し、計測結果と計画高さとの差分(偏差量)に応じて次層の積層高さをフィードバック制御する制御例を説明する。
図14は、図1に示す造形装置100の溶接ロボット17をスライダ装置60に搭載し、ポジショナ装置70を用いて積層造形物Wを製造する造形システム200を示す概略構成図である。造形システム200は、図1に示す造型装置100を含んで構成される。
図16は、ビード層の外周面の高さ情報とその処理結果を示す説明図である。図16のAに示す展開図情報80Aは、ビード層の外周面の高さ分布(点群データ)を示す等高線図であり、造形物の中心軸からの半径距離rが長い(高い)、短い(低い)を濃度の高低で表している。濃度が濃い領域ほど半径距離rが短く(高さが低く)、濃度が高い領域ほど半径距離rが長い(高さが高い)ことを表している。また、前述した単位区画の代表点Pkをドットとして示している。図16のAによれば、展開図情報80Aの領域Ar1,Ar2が周囲よりも特に低い高さとなっている。
演算部83は、取得した点群データを複数の単位区画に分割し、取得した積層計画データを単位区画に対応する領域に分割して、点群データと積層計画データとの単位区画ごとの差分値を算出する。差分値は、単位区画内を平均化した値同士で求めた偏差値であってもよい。
図18は、壁体の造形の様子を示す説明図である。ここでは、壁体の高さ異常を模擬するため、母材(ベースプレート27)を含む造形物を傾斜させた状態から造形を開始した。また、溶接ビードBの始端及び終端に関しては、局所的な高さ変動が生じるため、制御対象外とした。
(1) 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造方法であって、
前記造形装置が有する少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定する工程と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する工程と、
前記観測物理量と前記状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める工程と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する工程と、
を含む積層造形物の製造方法。
この積層造形物の製造方法によれば、状態遷移モデルと観測モデルを関連付けて形状指標を推定できる。これにより、計測できない又は計測しにくい箇所が存在して、観測モデルだけでは正確な形状指標の推定が難しい場合であっても、観測モデルと状態遷移モデルとを統合処理することで、互いの推定性能の弱点を補完して、より正確な予測が可能となる。これにより、積層造形物のどの箇所においても、統合処理により得られたより正確な状態遷移推定値に基づいて軌道計画を更新でき、高精度な積層造形物の造形制御が行える。また、状態遷移モデルを用いることで、単純な観測モデルの計算の場合と比較して、軌道計画の情報が加味されるため、推定精度を向上できる。
この積層製造物の製造方法によれば、推定した形状指標と軌道計画との差分が抑制され、より設計形状に近い積層造形物を製造できる。
この積層造形物の製造方法によれば、計測器では直接観測することが困難な溶接ビード同士の重なり量、積層時の溶融池温度、等の情報をシミュレーションにより算出した解析物理量を用いることで、形状指標をより高精度に推定できる。
この積層造形物の製造方法によれば、統合に使用するモデルを適切に選択(統合割合を調整)でき、状態遷移推定値のロバスト性を担保できる。
この積層造形物の製造方法によれば、予め判明している計測器等のノイズ成分が判明している場合に、そのノイズ成分による誤差を反映することで、状態遷移推定値の正確性を向上できる。
この積層造形物の製造方法によれば、造形する場所毎に適切な重み付け係数を決定でき、狙い位置ごとに状態遷移推定値が変動することを最小化できる。
この積層造形物の製造方法によれば、運棒動作によるノイズ信号の発生、又は運棒姿勢による計測器の計測環境の変化に対応して、重み付け係数を決定できるため、常に安定して高精度な形状指標の推定が可能となる。
この積層造形物の製造方法によれば、実際に形成される溶接ビードの形状に傾向に即してモデル化できる。例えば、直線状である箇所と、曲線状である箇所とを区別したモデル化が可能となり、状態遷移の傾向に応じた制御が可能となる。
前記状態遷移モデルは、前記軌道計画に基づく前記積層造形物の形状指標を表す計画形状データを含み、
前記点群データを複数の単位区画に分割し、前記計画形状データをそれぞれの前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとを前記単位区画ごとに比較した差分値を求め
前記差分値に応じて前記溶接ビードの形成条件を更新する、(1)に記載の積層造形物の製造方法。
この積層造形物の製造方法によれば、計測した点群データと計画形状データとを比較した差分値に応じて、単位区画ごとに溶接ビードの形成条件を更新できる。これにより、計画からの形状ずれが抑制され、高精度に積層造形物を製造できる
この積層造形物の製造方法によれば、特に差分値が大きい単位区画について溶接ビードの形成条件を更新できるため、効率よく形状ずれを抑制できる。
この積層造形物の製造方法によれば、計測ノイズ、又はビード表面の微小な凹凸の影響を受けにくくなり、単位区画内の大まかな傾向に基づいて形状の補正を安定して実施することが可能となる。
前記形状指標を計測したビード層における前記差分値が予め定めた限界値以下の場合には、当該ビード層の次に積層するビード層における前記溶接ビードの形成条件を更新し、
前記差分値が予め定めた限界値を超える場合には、前記形状指標を計測したビード層の以降に積層する複数のビード層における前記溶接ビードの形成条件をそれぞれ更新する、
(9)に記載の積層造形物の製造方法。
この積層造形物の製造方法によれば、差分値が過大となる単位区画が存在する場合でも、複数のビード層に補正を分配するため、急激な制御を防止して、緩やかに補正することが可能となる。
少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定するモデル設定部と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する物理量抽出部と、
前記観測物理量と前記状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める統合処理部と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する制御条件更新部と、
を備える積層造形物の製造装置。
この積層造形制御装置によれば、状態遷移モデルと観測モデルを関連付けて形状指標を推定できる。これにより、計測できない又は計測しにくい箇所が存在して、観測モデルだけでは正確な形状指標の推定が難しい場合であっても、観測モデルと状態遷移モデルとを統合処理することで、互いの推定性能の弱点を補完して、より正確な予測が可能となる。これにより、積層造形物のどの箇所においても、統合処理により得られたより正確な状態遷移推定値に基づいて軌道計画を更新でき、高精度な積層造形物の造形制御が行える。また、状態遷移モデルを用いることで、単純な観測モデルの計算の場合と比較して、軌道計画の情報が加味されるため、推定精度を向上できる。
前記観測モデルは、前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含み、
前記状態遷移モデルは、前記軌道計画に基づく前記積層造形物の形状指標を表す計画形状データを含み、
前記観測モデルから前記点群データを取得し、前記状態遷移モデルから前記計画形状データを取得するデータ取得部と、
取得した前記点群データを複数の単位区画に分割し、取得した前記計画形状データを前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとの前記単位区画ごとの差分値を算出する演算部と、
複数の前記単位区画のうち前記差分値が予め定めた閾値より大きい単位区画を抽出する抽出部と、
抽出した前記単位区画及び当該単位区画に対応する前記差分値の情報を出力する情報出力部と、
を備える制御支援装置。
この制御支援装置によれば、計測された点群データと計画された計画形状データとの差分値が閾値より大きい単位区画の位置が明確となり、その位置について選択的に制御することが容易に行える。
この制御支援装置によれば、表示部の表示によって、制御すべき箇所を作業者が視覚的に管理しやすくなる。
この制御支援装置によれば、座標系を統一することで、高さ等の傾向を把握しやすくなる。
前記コンピュータに、
前記造形装置が有する少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定する機能と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する機能と、
前記観測物理量と状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める機能と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する機能と、
を実現させるプログラム。
このプログラムによれば、状態遷移モデルと観測モデルを関連付けて形状指標を推定できる。これにより、計測できない又は計測しにくい箇所が存在して、観測モデルだけでは正確な形状指標の推定が難しい場合であっても、観測モデルと状態遷移モデルとを統合処理することで、互いの推定性能の弱点を補完して、より正確な予測が可能となる。これにより、積層造形物のどの箇所においても、統合処理により得られたより正確な状態遷移推定値に基づいて軌道計画を更新でき、高精度な積層造形物の造形制御が行える。また、状態遷移モデルを用いることで、単純な観測モデルの計算の場合と比較して、軌道計画の情報が加味されるため、推定精度を向上できる。
前記コンピュータに、
前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含む観測モデルと、前記軌道計画に基づく前記積層造形物の前記形状指標を表す計画形状データを含み、前記形状指標の状態遷移を表す状態遷移モデルとを設定する機能と、
前記観測モデルから前記点群データを取得し、前記状態遷移モデルから前記計画形状データを取得する機能と、
取得した前記点群データを複数の単位区画に分割し、前記計画形状データを前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとの前記単位区画ごとの差分値を算出する機能と、
算出した前記単位区画及び当該単位区画に対応する前記差分値の情報を出力する機能と、
を実現させるプログラム。
このプログラムによれば、計測された点群データと計画された計画形状データとの差分値が閾値より大きい単位区画の位置が明確となり、その位置について選択的に制御することが容易に行える。
13 制御部
15 溶接トーチ
17 溶接ロボット
19 ロボット駆動部
21 溶加材供給部
21a リール
23 溶接電源部
25 計測器
25A レーザセンサ
25B カメラ
25C 電流電圧計
27 ベースプレート
31 プロセッサ
33 メモリ
35 記憶部
37 入出力インターフェイス
39 軌道計画部
41 モデル設定部
43 物理量抽出部
45 統合処理部
47 制御条件更新部
51 撮像画像
53 アーク中心
55 溶加材先端
57 溶融池先端
59L 溶融池左端の線
59R 溶融池右端の線
60 スライダ装置
61 移動テーブル
63 水平移動機構
65 X軸レール
67 Y軸レール
70 ポジショナ装置
71 基台
73 回転体
80,80A,80B,80C 展開図
81 データ取得部
83 演算部
85 抽出部
87 情報出力部
89 表示部
100 造形装置(積層造形物の製造装置)
200 造形システム
300 制御支援装置
B 溶接ビード
M 溶加材
R 障害物
W 積層造形物
Claims (20)
- 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造方法であって、
前記造形装置が有する少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定する工程と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する工程と、
前記観測物理量と前記状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める工程と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する工程と、
を含む積層造形物の製造方法。 - 前記軌道計画を更新する工程は、前記形状指標の前記状態遷移推定値を求める工程で推定された前記形状指標と、前記軌道計画で計画された前記形状指標との差分に応じて、前記溶接ビードの形成条件を調整する工程を含む、
請求項1に記載の積層造形物の製造方法。 - 前記統合処理は、前記軌道計画に基づいて前記溶接ビードを模擬形成するシミュレーションモデルから抽出される解析物理量を含めて統合する、
請求項1に記載の積層造形物の製造方法。 - 前記統合処理は、前記軌道計画に基づいて前記溶接ビードを模擬形成するシミュレーションモデルから抽出される解析物理量を含めて統合する、
請求項2に記載の積層造形物の製造方法。 - 前記統合処理は、前記観測物理量と前記状態物理量とを重み付け係数で線形結合する、
請求項1~4のいずれか1項に記載の積層造形物の製造方法。 - 前記重み付け係数の初期値を、前記観測モデル及び前記状態遷移モデルが有する誤差の分散指標に応じて決定する、
請求項5に記載の積層造形物の製造方法。 - 前記重み付け係数を、前記軌道計画で規定する溶着狙い位置ごとに調整する、
請求項5に記載の積層造形物の製造方法。 - 前記観測モデルの重み付け係数を、前記溶接ビードを形成する溶接トーチの運棒動作又は運棒姿勢に応じて調整する、
請求項5に記載の積層造形物の製造方法。 - 前記状態遷移モデルとして、溶接ビード形状が定常的に推移するモデルと、前記溶接ビード形状が曲線的に変化するモデルとを、同一のビード形成軌道上で組合せることを特徴とする、
請求項1~4のいずれか1項に記載の積層造形物の製造方法。 - 前記状態遷移モデルとして、溶接ビード形状が定常的に推移するモデルと、前記溶接ビード形状が曲線的に変化するモデルとを、同一のビード形成軌道上で組合せることを特徴とする、
請求項5に記載の積層造形物の製造方法。 - 前記観測モデルは、前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含み、
前記状態遷移モデルは、前記軌道計画に基づく前記積層造形物の形状指標を表す計画形状データを含み、
前記点群データを複数の単位区画に分割し、前記計画形状データをそれぞれの前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとを前記単位区画ごとに比較した差分値を求め
前記差分値に応じて前記溶接ビードの形成条件を更新する、
請求項1に記載の積層造形物の製造方法。 - 複数の前記単位区画のうち前記差分値が予め定めた閾値より大きい単位区画を抽出し、抽出した前記単位区画に対応する領域に対して前記溶接ビードの形成条件を更新する、
請求項11に記載の積層造形物の製造方法。 - 前記差分値は、前記点群データの前記単位区画内に含まれる前記形状指標を平均化した値と、前記計画形状データの前記形状指標を前記単位区画に対応する領域内で平均化した値との偏差量である、
請求項12に記載の積層造形物の製造方法。 - 前記差分値を、前記溶接ビードにより形成されるビード層毎に求め、
前記形状指標を計測したビード層における前記差分値が予め定めた限界値以下の場合には、当該ビード層の次に積層するビード層における前記溶接ビードの形成条件を更新し、
前記差分値が予め定めた限界値を超える場合には、前記形状指標を計測したビード層の以降に積層する複数のビード層における前記溶接ビードの形成条件をそれぞれ更新する、
請求項11に記載の積層造形物の製造方法。 - 溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造装置であって、
少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定するモデル設定部と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する物理量抽出部と、
前記観測物理量と前記状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める統合処理部と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する制御条件更新部と、
を備える積層造形物の製造装置。 - 請求項15に記載の積層造形物の製造装置の制御を支援する制御支援装置であって、
前記観測モデルは、前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含み、
前記状態遷移モデルは、前記軌道計画に基づく前記積層造形物の形状指標を表す計画形状データを含み、
前記観測モデルから前記点群データを取得し、前記状態遷移モデルから前記計画形状データを取得するデータ取得部と、
取得した前記点群データを複数の単位区画に分割し、取得した前記計画形状データを前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとの前記単位区画ごとの差分値を算出する演算部と、
複数の前記単位区画のうち前記差分値が予め定めた閾値より大きい単位区画を抽出する抽出部と、
抽出した前記単位区画及び当該単位区画に対応する前記差分値の情報を出力する情報出力部と、
を備える制御支援装置。 - 前記情報出力部からの出力情報を表示する表示部を備える、
請求項16に記載の制御支援装置。 - 前記演算部は、前記点群データの座標系を、前記積層造形物の製造装置の駆動座標系に座標変換する、
請求項16に記載の制御支援装置。 - 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造手順を、コンピュータに実行させるプログラムであって、
前記コンピュータに、
前記造形装置が有する少なくとも一以上の計測器から収集された計測情報から得られる観測モデルと、前記軌道計画から抽出される前記積層造形物の形状指標の状態遷移を表す状態遷移モデルとを設定する機能と、
前記観測モデルから前記形状指標の観測物理量を抽出し、前記状態遷移モデルから前記形状指標の状態物理量を抽出する機能と、
前記観測物理量と状態物理量とを統合処理して前記形状指標の状態遷移推定値を求める機能と、
前記状態遷移推定値に応じて前記軌道計画が定める前記溶接ビードの形成条件を更新する機能と、
を実現させるプログラム。 - 造形装置により溶加材を溶融及び凝固させて形成する溶接ビードを、予め定めた軌道計画に基づいて繰り返し積層する積層造形物の製造手順を、コンピュータに実行させるプログラムであって、
前記コンピュータに、
前記積層造形物の形状指標を計測するセンサから収集され、前記積層造形物の計測された前記形状指標を表す点群データを含む観測モデルと、前記軌道計画に基づく前記積層造形物の前記形状指標を表す計画形状データを含み、前記形状指標の状態遷移を表す状態遷移モデルとを設定する機能と、
前記観測モデルから前記点群データを取得し、前記状態遷移モデルから前記計画形状データを取得する機能と、
取得した前記点群データを複数の単位区画に分割し、前記計画形状データを前記単位区画に対応する領域に分割して、前記点群データと前記計画形状データとの前記単位区画ごとの差分値を算出する機能と、
算出した前記単位区画及び当該単位区画に対応する前記差分値の情報を出力する機能と、
を実現させるプログラム。
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US18/729,923 US20250178127A1 (en) | 2022-01-19 | 2022-12-20 | Laminate molded object manufacturing method and manufacturing device, control support device, and program |
| EP22922165.0A EP4450198A4 (en) | 2022-01-19 | 2022-12-20 | Laminate molded object manufacturing method and manufacturing device, control support device, and program |
| CN202280089239.3A CN118613343A (zh) | 2022-01-19 | 2022-12-20 | 层叠造型物的制造方法及制造装置、控制支援装置以及程序 |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2022-006772 | 2022-01-19 | ||
| JP2022006772 | 2022-01-19 | ||
| JP2022-152600 | 2022-09-26 | ||
| JP2022152600 | 2022-09-26 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023140015A1 true WO2023140015A1 (ja) | 2023-07-27 |
Family
ID=87348123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2022/046977 Ceased WO2023140015A1 (ja) | 2022-01-19 | 2022-12-20 | 積層造形物の製造方法及び製造装置、制御支援装置、並びにプログラム |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20250178127A1 (ja) |
| EP (1) | EP4450198A4 (ja) |
| JP (1) | JP2023105801A (ja) |
| WO (1) | WO2023140015A1 (ja) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117798463A (zh) * | 2023-12-29 | 2024-04-02 | 浙大城市学院 | 金属结构件多轴3d打印装置及电弧增材方法 |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015160217A (ja) * | 2014-02-26 | 2015-09-07 | 国立大学法人東京農工大学 | 三次元造形装置、三次元造形物の造形方法、および三次元製造装置の制御プログラム |
| JP6576593B1 (ja) | 2018-11-09 | 2019-09-18 | 三菱電機株式会社 | 積層造形装置 |
| JP2020192752A (ja) * | 2019-05-29 | 2020-12-03 | 株式会社神戸製鋼所 | 造形物の製造方法、造形物の製造手順生成方法、造形物の製造システム、造形物の製造手順生成装置、及びプログラム |
| JP6797324B1 (ja) * | 2020-05-20 | 2020-12-09 | 株式会社神戸製鋼所 | 積層造形方法 |
| JP2021138030A (ja) * | 2020-03-04 | 2021-09-16 | 株式会社神戸製鋼所 | 積層造形物の材質予測方法、材質データベースの構築方法、積層造形物の積層計画方法、及び積層造形物の製造方法 |
| JP6989059B1 (ja) * | 2020-10-15 | 2022-01-05 | 三菱電機株式会社 | 積層造形方法、積層造形装置及び積層造形システム |
| JP2022006772A (ja) | 2020-06-25 | 2022-01-13 | ダイムラー・アクチェンゲゼルシャフト | 電動車両 |
| JP2022152600A (ja) | 2021-03-29 | 2022-10-12 | ソマール株式会社 | 粘着シート |
-
2022
- 2022-12-20 WO PCT/JP2022/046977 patent/WO2023140015A1/ja not_active Ceased
- 2022-12-20 US US18/729,923 patent/US20250178127A1/en active Pending
- 2022-12-20 EP EP22922165.0A patent/EP4450198A4/en active Pending
-
2023
- 2023-01-10 JP JP2023001859A patent/JP2023105801A/ja active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015160217A (ja) * | 2014-02-26 | 2015-09-07 | 国立大学法人東京農工大学 | 三次元造形装置、三次元造形物の造形方法、および三次元製造装置の制御プログラム |
| JP6576593B1 (ja) | 2018-11-09 | 2019-09-18 | 三菱電機株式会社 | 積層造形装置 |
| JP2020192752A (ja) * | 2019-05-29 | 2020-12-03 | 株式会社神戸製鋼所 | 造形物の製造方法、造形物の製造手順生成方法、造形物の製造システム、造形物の製造手順生成装置、及びプログラム |
| JP2021138030A (ja) * | 2020-03-04 | 2021-09-16 | 株式会社神戸製鋼所 | 積層造形物の材質予測方法、材質データベースの構築方法、積層造形物の積層計画方法、及び積層造形物の製造方法 |
| JP6797324B1 (ja) * | 2020-05-20 | 2020-12-09 | 株式会社神戸製鋼所 | 積層造形方法 |
| JP2022006772A (ja) | 2020-06-25 | 2022-01-13 | ダイムラー・アクチェンゲゼルシャフト | 電動車両 |
| JP6989059B1 (ja) * | 2020-10-15 | 2022-01-05 | 三菱電機株式会社 | 積層造形方法、積層造形装置及び積層造形システム |
| JP2022152600A (ja) | 2021-03-29 | 2022-10-12 | ソマール株式会社 | 粘着シート |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP4450198A4 |
Also Published As
| Publication number | Publication date |
|---|---|
| EP4450198A1 (en) | 2024-10-23 |
| EP4450198A4 (en) | 2025-04-09 |
| US20250178127A1 (en) | 2025-06-05 |
| JP2023105801A (ja) | 2023-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3863791B1 (en) | System and method for weld path generation | |
| CN110312588B (zh) | 层叠控制装置、层叠控制方法以及存储介质 | |
| US12393171B2 (en) | Closed-loop robotic deposition of material | |
| EP4137259B1 (en) | Laminate molding method | |
| CN116615301B (zh) | 附加制造装置、附加制造系统、附加制造方法、记录介质及学习装置 | |
| JP6921361B1 (ja) | 付加製造装置および付加製造方法 | |
| US20180307207A1 (en) | Closed-loop robotic deposition of material | |
| JP6989059B1 (ja) | 積層造形方法、積層造形装置及び積層造形システム | |
| JP7391709B2 (ja) | 造形物の製造方法、造形物の製造装置、及びプログラム | |
| CN118492415B (zh) | 用于同步复合增材制造的多轴联动控制方法及系统 | |
| CN117565402A (zh) | 3d打印控制执行方法、系统及3d打印设备 | |
| JP6753989B1 (ja) | 積層造形物の積層計画方法、積層造形物の製造方法及び製造装置 | |
| US10579046B2 (en) | Closed-loop robotic deposition of material | |
| WO2023140015A1 (ja) | 積層造形物の製造方法及び製造装置、制御支援装置、並びにプログラム | |
| JP2024012925A (ja) | ロボットの制御方法、およびロボットの制御装置 | |
| JP7505999B2 (ja) | 積層造形物の変形予測方法 | |
| JP7166489B1 (ja) | 付加製造システム、付加製造装置、情報処理装置および付加製造方法 | |
| CN118661071A (zh) | 缺陷检测方法、层叠造型物的制造方法、缺陷检测装置以及层叠造型装置 | |
| JP7418547B2 (ja) | 数値制御装置および付加製造装置の制御方法 | |
| JP7353245B2 (ja) | 付加製造装置および付加製造方法 | |
| EP4549071A1 (en) | Control information correction method, control information correction device, and program | |
| CN118613343A (zh) | 层叠造型物的制造方法及制造装置、控制支援装置以及程序 | |
| JP7601747B2 (ja) | 溶接パス自律最適化装置、溶接パス自律最適化システム、および、溶接パス自律最適化方法 | |
| CN117836082A (zh) | 造型装置的控制方法、造型装置以及程序 | |
| WO2021029297A1 (ja) | 積層造形物の積層計画方法、積層造形物の製造方法及び製造装置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22922165 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 202280089239.3 Country of ref document: CN |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2022922165 Country of ref document: EP |
|
| ENP | Entry into the national phase |
Ref document number: 2022922165 Country of ref document: EP Effective date: 20240718 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWP | Wipo information: published in national office |
Ref document number: 18729923 Country of ref document: US |