WO2022224450A9 - 機械学習装置、加減速調整装置及びコンピュータ読み取り可能な記憶媒体 - Google Patents
機械学習装置、加減速調整装置及びコンピュータ読み取り可能な記憶媒体 Download PDFInfo
- Publication number
- WO2022224450A9 WO2022224450A9 PCT/JP2021/016479 JP2021016479W WO2022224450A9 WO 2022224450 A9 WO2022224450 A9 WO 2022224450A9 JP 2021016479 W JP2021016479 W JP 2021016479W WO 2022224450 A9 WO2022224450 A9 WO 2022224450A9
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- machining
- parameters
- unit
- machine tool
- data
- Prior art date
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 28
- 230000001133 acceleration Effects 0.000 title claims description 57
- 238000003860 storage Methods 0.000 title claims description 20
- 238000003754 machining Methods 0.000 claims abstract description 141
- 230000006870 function Effects 0.000 claims abstract description 57
- 238000004364 calculation method Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims description 39
- 238000000034 method Methods 0.000 claims description 13
- 230000009471 action Effects 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000010586 diagram Methods 0.000 description 5
- 238000009826 distribution Methods 0.000 description 5
- 230000036461 convulsion Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- 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
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
-
- 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
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
-
- 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
- G05B2219/32—Operator till task planning
- G05B2219/32186—Teaching inspection data, pictures and criteria and apply them for inspection
-
- 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
- G05B2219/32—Operator till task planning
- G05B2219/32187—Correlation between controlling parameters for influence on quality parameters
-
- 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
- G05B2219/32—Operator till task planning
- G05B2219/32193—Ann, neural base quality management
Definitions
- the present invention relates to a machine learning device, an acceleration/deceleration adjustment device, and a computer-readable storage medium.
- the machining speed for machining the workpiece is commanded as the axis movement speed within the machining program.
- the movement speed of the axis commanded in the machining program is the maximum speed of relative movement (tool movement) between the tool and the workpiece.
- the machine tool fluctuates the movement speed of the axes according to the parameters related to the control of each axis at the start of machining, corners, curved parts, etc., within a range that does not exceed the commanded maximum speed.
- target tolerances and machined surface quality are set in advance.
- a target machining time is also determined in advance.
- the operator of the machine tool adjusts parameters such as the acceleration/deceleration time constant and the movement speed commanded in the machining program while checking the machining error and machined surface quality of the product after machining.
- Patent Document 1 As a conventional technology for adjusting the parameters related to the control of each axis in the machining of products, there is a patent for obtaining an optimum speed distribution that balances machining errors, machined surface quality, and machining time using machine learning technology. An application has been filed (for example, Patent Document 1, etc.).
- an appropriate velocity distribution is set for a given machining purpose in a given machine tool. Therefore, there is also the problem that it is necessary to reconfigure the appropriate velocity distribution each time the machine tool for machining is changed or the purpose of machining is changed. Therefore, there is a demand for a technique that can adjust parameters in machining based on criteria other than velocity distribution.
- the acceleration/deceleration adjustment device solves the above problems by directly specifying permissible machining errors such as shape errors and positional deviations, and machined surface quality to enable quantitative control.
- machine learning that optimizes the combination of set values of parameters that control the amount of movement of each axis for each control cycle, including the N-order time differential element (N is a natural number) of the speed of each axis.
- one aspect of the present disclosure is a machine learning device that estimates a parameter related to the control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis of a machine tool that processes a workpiece.
- a state observation unit that observes, as data indicating an operating state of the machine tool, information relating to at least one of machining accuracy and machined surface quality in the machining, and machining time required for the machining
- a determination condition acquisition unit that acquires a target value related to data observed by a state observation unit as determination data; and the parameter based on the data observed by the state observation unit and the determination data acquired by the determination condition acquisition unit.
- a reward calculation unit for calculating a reward for processing based on the value
- a value function update unit for updating a value function for calculating the value of the processing state based on the parameter based on the reward
- a value function updated based on the value function a decision making unit that estimates a combination of the parameter setting values more suitable for the processing, and outputs the estimated combination of the parameter setting values.
- Another aspect of the present disclosure is an acceleration/deceleration adjustment device that adjusts parameters related to control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis of a machine tool that processes a workpiece.
- a state observation unit that observes, as data indicating an operating state of the machine tool, information relating to at least one of machining accuracy and machined surface quality in the machining, and machining time required for the machining;
- a determination condition acquisition unit that acquires a target value related to data observed by the state observation unit as determination data; a value function storage unit that stores a value function for calculating the value of the processed state based on the parameter; and the value function.
- a decision-making unit for estimating a combination of the parameter setting values more suitable for the processing based on the above, and outputting the estimated combination of the parameter setting values; and an action output unit that adjusts the parameters of the machine tool based on the combination.
- Another aspect of the present disclosure is a machine learning device that estimates a parameter related to control of the amount of movement for each control cycle, including an N-order time differential element (N is a natural number) of each axis of a machine tool that processes a workpiece.
- a computer-readable storage medium storing a program for operating a computer, wherein information relating to at least one of machining accuracy and machined surface quality in the machining and the machining time required for the machining are stored in the machine tool.
- a state observation unit that observes data indicating an operating state
- a determination condition acquisition unit that acquires a target value related to the data observed by the state observation unit as determination data, the data observed by the state observation unit, and the determination condition.
- a reward calculation unit that calculates a reward for processing based on the parameters based on the determination data acquired by the acquisition unit, and updates a value function for calculating a value of the processing state based on the parameters based on the reward.
- a value function updating unit, and a decision making unit that estimates a combination of setting values of the parameters that are more suitable for the processing based on the updated value function and outputs the estimated combination of setting values of the parameters, a computer comprising: is a computer-readable storage medium storing a program for operating the
- a computer-readable storage medium storing a program for operating a computer as a computer, wherein information related to at least one of machining accuracy and machined surface quality in the machining and the machining time required for the machining are stored in the machine tool a state observation unit that observes as data indicating the operation state of the state observation unit, a determination condition acquisition unit that acquires a target value related to the data observed by the state observation unit as determination data, and a processing state value based on the parameters.
- a decision-making unit that estimates a combination of setting values of the parameters that is more suitable for the processing based on the value function, and outputs the estimated combination of setting values of the parameters.
- a computer-readable storage medium storing a program that causes a computer to operate as an action output unit that adjusts the parameters of the machine tool based on the combination of the set values of the parameters output by the decision making unit.
- machining accuracy/machined surface quality target value shape error, positional deviation, etc.
- FIG. 2 is a hardware configuration diagram of an acceleration/deceleration adjusting device
- FIG. 3 is a block diagram showing functions of an acceleration/deceleration adjusting device
- FIG. It is a figure explaining calculation of processing accuracy. It is a figure explaining calculation of machined surface quality.
- 4 is a flowchart showing a schematic operation example of an acceleration/deceleration adjusting device; It is a figure explaining adjustment of a parameter.
- FIG. 5 is a block diagram showing functions of an acceleration/deceleration adjusting device according to another embodiment;
- FIG. 1 is a schematic hardware configuration diagram showing essential parts of an acceleration/deceleration adjusting device according to an embodiment of the present invention.
- the acceleration/deceleration adjusting device 1 of the present invention can be implemented, for example, as a control device that controls a machine tool based on a machining program.
- the acceleration/deceleration adjusting device 1 of the present invention may be a personal computer attached to a control device for controlling a machine tool based on a machining program, a personal computer connected to the control device via a wired/wireless network, a cell computer, It can be implemented on computers such as the fog computer 6 and the cloud server 7 .
- This embodiment shows an example in which the acceleration/deceleration adjusting device 1 is mounted on a personal computer connected to a control device that controls a machine tool via a network.
- the CPU 11 included in the acceleration/deceleration adjusting device 1 is a processor that controls the acceleration/deceleration adjusting device 1 as a whole.
- the CPU 11 reads a system program stored in the ROM 12 via the bus 22 and controls the entire acceleration/deceleration adjusting device 1 according to the system program.
- the RAM 13 temporarily stores calculation data, display data, various data input from the outside, and the like.
- the non-volatile memory 14 is composed of, for example, a memory backed up by a battery (not shown) or an SSD (Solid State Drive), and retains the stored state even when the power of the acceleration/deceleration adjusting device 1 is turned off.
- the nonvolatile memory 14 stores data read from the external device 72 via the interface 15, data input via the input device 71, and data obtained from the machine tool 3 (including data detected by the sensor 4). ) are stored.
- the data stored in the nonvolatile memory 14 may be developed in the RAM 13 at the time of execution/use.
- Various system programs such as a well-known analysis program are pre-written in the ROM 12 .
- a sensor 4 is attached to the machine tool 3 to detect physical quantities such as current, voltage, and vibration of each part during operation of the machine tool 3 .
- Examples of the machine tool 3 include a machining center and a lathe.
- the machine tool 3 transmits data such as the position, speed, acceleration, jerk, vibration, and machining time of each axis during machining via the network 5.
- the interface 15 is an interface for connecting the CPU 11 of the acceleration/deceleration adjusting device 1 and an external device 72 such as a USB device. From the external device 72 side, for example, a pre-stored machining program, data relating to the operation of each machine tool 3, and the like can be read. Also, the machining program and setting data edited in the acceleration/deceleration adjusting device 1 can be stored in the external storage means via the external device 72 .
- the interface 20 is an interface for connecting the CPU of the acceleration/deceleration adjusting device 1 and the wired or wireless network 5 .
- a machine tool 3 , a fog computer 6 , a cloud server 7 and the like are connected to the network 5 to exchange data with the acceleration/deceleration adjusting device 1 .
- An input device 71 composed of a keyboard, a pointing device, and the like passes commands, data, etc. based on operations by the operator to the CPU 11 via the interface 18 .
- the interface 21 is an interface for connecting the CPU 11 and the machine learning device 100 .
- the machine learning device 100 includes a processor 101 that controls the entire machine learning device 100, a ROM 102 that stores system programs and the like, a RAM 103 for temporary storage in each process related to machine learning, and a storage of models and the like. It has a non-volatile memory 104 that is used.
- the machine learning device 100 can observe each piece of information (for example, data detected during machining by the machine tool 3) that can be acquired by the acceleration/deceleration adjustment device 1 via the interface 21.
- the acceleration/deceleration adjustment device 1 also acquires processing results output from the machine learning device 100 via the interface 21, stores and displays the acquired results, and communicates the network 5 and the like to other devices. Send via.
- FIG. 2 is a schematic block diagram showing the functions of the acceleration/deceleration adjusting device 1 according to the first embodiment of the present invention.
- Each function provided in the acceleration/deceleration adjusting device 1 according to the present embodiment is performed by the CPU 11 provided in the acceleration/deceleration adjusting device 1 shown in FIG. It is realized by controlling the operation of each part of the device 1 and the machine learning device 100 .
- the acceleration/deceleration adjusting device 1 of this embodiment includes a state observing section 110 , a determination condition obtaining section 120 and an action output section 150 .
- the machine learning device 100 of the acceleration/deceleration adjusting device 1 also includes a learning section 130 and a decision making section 140 .
- a value function storage unit 138 that stores the value function as the result of machine learning by the learning unit 130 is prepared in advance.
- the state observation unit 110 observes information related to at least one of machining accuracy and machined surface quality, and machining time as data indicating the operating state of the machine tool. Observing here means acquiring data from the environment and calculating predetermined data based on the acquired data. First, the state observation unit 110 acquires various data detected during operation of the machine tool 3 as data indicating the operating state of machining by the machine tool 3 . The state observation unit 110 acquires, for example, the position, velocity, acceleration, jerk, vibration, and machining time of each axis during machining by the machine tool 3 as data indicating the operation status of machining by the machine tool 3 .
- the state observation unit 110 includes parameters (linear acceleration, linear acceleration, post-interpolation acceleration/deceleration time constant, corner speed difference, position loop gain, feedforward coefficient, etc.) and a machining program used for machining control are acquired as data indicating the operating state of machining by the machine tool 3 .
- the data acquired by the state observation unit 110 may be instantaneous values acquired at a predetermined timing.
- the data acquired by the state observation unit 110 may be time-series data acquired over a predetermined period of time.
- the state observation unit 110 measures the machining accuracy of the machining based on the position data, velocity data, acceleration data, jerk data, vibration data, etc. of each axis included in the data indicating the operating state of machining by the machine tool 3. and data related to machined surface quality. Examples of data related to the calculated machining accuracy and machined surface quality include shape errors, positional deviations, vibration errors, and the like.
- FIG. 3 shows the machining program path instructed by the machining program executed in the machine tool 3 and the movement path calculated based on the position of the motor.
- the horizontal axis indicates the X coordinate position
- the vertical axis indicates the Y coordinate position.
- the solid line arrow indicates the machining program path commanded by the machining program
- the dotted line arrow indicates the movement path calculated based on the position of the motor.
- each axis is moved along the machining program path commanded by the machining program based on the parameters related to the control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis.
- N is a natural number
- a movement command to be output to the motor of the axis is calculated.
- a movement command with priority given to machining speed, efficiency, etc. is calculated within the range of the set parameters. Therefore, when the motor is moved based on the calculated movement path, the movement path calculated based on the position of the motor (the relative movement path between the tool driven by the motor and the workpiece) follows the machining program path. do not become.
- the state observation unit 110 calculates a shape error indicating machining accuracy and a position deviation indicating machined surface quality based on the difference between this machining program path and the movement path calculated based on the position of the motor.
- shape error for example, the maximum value Emax of the shape error may be used.
- positional deviation an average value Emean, a variance value Edist, or the like may be used.
- FIG. 4 shows the machining program path instructed by the machining program executed in the machine tool 3 and the movement path (with/without vibration) calculated based on the position of the motor. Looking more closely at the path calculated based on the motor position, the motor position data oscillates. Since this vibration appears as scratches and streaks during machining, it affects the quality of the machined surface.
- the state observation unit 110 calculates a vibration error indicating the machined surface quality based on the difference between the movement command output to the motor and the position acquired from the motor.
- the vibration error for example, the maximum value Amax of the amplitude may be used, or the average value Amean of the amplitude may be used.
- the vibration error may be calculated using vibration data from the vibration measuring device in addition to the position data of the motor. Vibration data from the vibration measuring device can provide vibration closer to the machining point (contact position between the tool and the workpiece).
- the state observation unit 110 may acquire data directly from the machine tool 3 via the network 5.
- the state observation unit 110 may acquire data acquired and stored by the external device 72, the fog computer 6, the cloud server 7, or the like.
- the data acquired or calculated by the state observation unit 110 is input to the learning unit 130 and the decision making unit 140 .
- the determination condition acquisition unit 120 acquires determination data related to the purpose of processing in processing by the machine tool 3 .
- Judgment data related to the purpose of machining for example, machining accuracy such as a predetermined allowable machining accuracy (allowable shape error), allowable machining surface quality (allowable position deviation), allowable machining surface quality (allowable vibration error), etc. Acceptable values related to machined surface quality can be mentioned.
- the determination data related to the purpose of processing includes, for example, target processing time.
- the determination condition acquisition unit 120 may acquire tolerance values related to machining accuracy and machined surface quality set in the machine tool 3 via the network 5 .
- the determination condition acquisition unit 120 may acquire data stored by the external device 72, the fog computer 6, the cloud server 7, or the like.
- the determination condition acquisition unit 120 may prompt the operator to input tolerance values related to machining accuracy and machined surface quality, and target machining time from the input device 71 .
- the data acquired by the determination condition acquisition unit 120 is input to the learning unit 130 and the decision making unit 140 .
- the learning unit 130 executes processing related to machine learning based on the data indicating the operation state of machining by the machine tool 3 acquired by the state observation unit 110 and the determination data related to the purpose of machining acquired by the determination condition acquisition unit 120. .
- the learning unit 130 includes a reward calculator 132 and a value function updater 134 . Based on the reward calculated by the reward calculation unit 132, the learning unit 130 updates the value function by the value function update unit 134, so that every control cycle including the N-order time differential element (N is a natural number) of each axis and the correlation between the value of the combination and the value of the combination of parameters related to the control of the amount of movement of .
- the remuneration calculation unit 132 calculates a remuneration for the current operating state of the machine tool 3 based on data indicating the operating state of machining by the machine tool 3 and determination data related to the purpose of machining.
- the remuneration calculation unit 132 compares the values indicating the machining accuracy and the machined surface quality calculated by the state observation unit 110 with the determination data related to the purpose of machining, and calculates a predetermined remuneration calculation formula set in advance based on the comparison result. Calculate the reward by
- the determination data related to the purpose of machining includes allowable values related to machining accuracy and machined surface quality.
- the remuneration calculation unit 132 calculates a high remuneration when the calculated values indicating the machining accuracy and the machined surface quality are within the allowable values. Further, the remuneration calculation unit 132 calculates a low remuneration when the calculated value indicating the machining accuracy or the machined surface quality exceeds the allowable value. The remuneration calculation unit 132 may calculate a higher remuneration according to the degree of being within the allowable value. Further, the remuneration calculation unit 132 may calculate a lower remuneration according to the degree of exceeding the allowable value. The reward calculator 132 may calculate a negative reward.
- the reward calculation unit 132 compares the value indicating the machining time required for machining in the machine tool 3 with the target machining time included in the determination data related to the machining purpose, and based on the comparison result, a preset Additional remuneration is calculated according to a predetermined remuneration calculation formula.
- the remuneration calculation unit 132 calculates a high remuneration when the processing time required for processing is within the target processing time. Further, the remuneration calculation unit 132 calculates a low remuneration when the processing time required for processing exceeds the target processing time.
- the remuneration calculation unit 132 may calculate a higher remuneration according to the extent to which the processing time required for processing is within the processing time. Further, the remuneration calculation unit 132 may calculate a lower remuneration according to the degree of exceeding the processing time required for processing. The reward calculator 132 may calculate a negative reward. The remuneration calculation unit 132 adds the additional remuneration calculated in this way to the remuneration calculated based on the machining accuracy and the quality of the machined surface.
- the reward calculator 132 may store the machining time when the machining accuracy and the machined surface quality are within the allowable values. At this time, among the stored machining times, the shortest machining time is used as determination data relating to the machining purpose. Then, only when the machining accuracy and the machined surface quality are within the allowable range, the remuneration calculation unit 132 calculates the remuneration based on the machining time at that time and the above-described shortest machining time stored as a reference for remuneration calculation. calculate. By doing so, it becomes possible to search for the parameter that minimizes the machining time within the range of target machining accuracy and machined surface quality.
- the value function update unit 134 updates the value function stored in the value function storage unit 138 based on the reward calculated by the reward calculation unit 132.
- the value function used in the present invention is a combination of parameters related to the control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis set during machining by the machine tool 3. It is a state value function that calculates the value of being in that state.
- the state-value function V takes, for example, each state (a combination of parameters related to the control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis) and performs machining.
- the state value function may be defined as a function that returns the reward calculated by the reward calculation unit 132 as a value when It should be noted that the state value function is preferably set to output high values for all possible states at the stage when learning is started.
- the state value function uses, as input data, a combination of parameters related to the control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis set during machining by the machine tool 3, and the parameters It may be constructed as a multi-layer neural network, etc., in which the value of the state of combination is used as output data.
- the decision-making unit 140 determines parameters related to the control of the amount of movement for each control cycle, including the N-order time differential element (N is a natural number) of each axis, based on the value function created by the machine learning performed by the learning unit 130. output a combination of The decision-making unit 140 uses the value function to obtain a parameter combination having a higher value than the parameter combination set in the current machining. The decision making unit 140 compares, for example, the value calculated from the currently set parameter combination with the value calculated from the parameter combination when each parameter is changed by a predetermined amount.
- a combination of currently set parameters a combination of parameters that change linear acceleration by + ⁇ A, a combination of parameters that change linear acceleration by - ⁇ A, and a parameter that changes acceleration/deceleration time constant after interpolation by + ⁇ .
- Values are calculated using the value function for each of the combination, the parameter combination obtained by changing the post-interpolation acceleration/deceleration time constant by - ⁇ , and the parameter combination with the higher value is obtained. Then, the obtained parameter combination is output as a parameter combination more suitable for the current machining.
- the decision making unit 140 may output the current parameter combination as a higher value parameter combination. good.
- the decision making section 140 may randomly output a combination of parameters other than the one currently set. In the initial stage of learning by the learning unit 130, the decision making unit 140 preferably outputs a random combination of parameters with a certain probability regardless of the value calculated by the value function ( ⁇ -greedy method). By doing so, it is possible to efficiently search for a more appropriate combination of parameters.
- the action output unit 150 determines whether to continue parameter adjustment based on the combination of parameters output by the decision making unit 140 . Then, when it is determined to continue adjusting the parameters, the machine tool 3 is set to the combination of parameters output by the decision making section 140, and is instructed to perform the machining operation again. For example, when the combination of parameters output by the decision making unit 140 is different from the combination of parameters currently set in the machine tool 3, the action output unit 150 may determine to continue parameter adjustment. In addition, the action output unit 150 records the number of times the parameter of the machine tool 3 has been changed since the start of parameter adjustment, and if the number of times the parameter has been changed is within a predetermined number of times, the parameter It may be determined to continue the adjustment of .
- the action output unit 150 may directly set a combination of parameters for the machine tool 3 via the network 5.
- the action output unit 150 may also transmit a combination of parameters to the fog computer 6 or the cloud server 7 via the network 5 to indirectly prompt the machine tool 3 to set the parameters.
- the action output unit 150 may display the combination of parameters on the display device 70 and prompt the operator to set the parameters in the machine tool 3 .
- FIG. 5 is a flow chart illustrating a process for finding a combination of parameters using the acceleration/deceleration adjusting device 1 having the above configuration.
- the acceleration/deceleration adjusting device 1 When an operator or the like instructs adjustment of parameters related to control of the amount of movement for each control cycle, including the N-th order time differential element (N is a natural number) of each axis, the acceleration/deceleration adjusting device 1 is first preliminarily A combination of set parameters (for example, linear acceleration of 4000 mm/sec 2 , corner speed difference of 800 mm/min, post-interpolation acceleration/deceleration time constant of 32 msec, etc.) is set for the machine tool 3 . Then, the machine tool 3 is instructed to run idle according to a predetermined machining program (step SA01).
- a predetermined machining program step SA01
- the state observation unit 110 collects data indicating the operating state of the machine tool 3 (time-series data of the motor speed, time-series data of the motor speed, etc.). , machining time, etc.) are acquired (step SA02). Then, based on the acquired data indicating the operating state of the machine tool 3, data related to machining accuracy and machined surface quality (for example, machining accuracy (shape error): 80 ⁇ m, machined surface quality (positional deviation): 8 ⁇ m, machining surface quality (vibration error): 0.09 ⁇ m, etc.) is calculated (step SA02).
- machining accuracy shape error
- machined surface quality positional deviation
- machining surface quality machining surface quality (vibration error): 0.09 ⁇ m, etc.
- the determination condition acquisition unit 120 obtains determination data related to the purpose of machining in the machining by the machine tool 3 (machining accuracy (allowable shape error): 100 ⁇ m, machining surface quality (allowable position deviation): 10 ⁇ m, machining surface quality (allowable vibration error): 0.1 ⁇ m, allowable processing time 12.0 sec, etc.) is acquired (step SA03).
- Reward calculation unit 132 based on the data related to the machining accuracy and machined surface quality input from the state observation unit 110, and the determination data related to the processing purpose input from the determination condition acquisition unit 120, the combination of the current parameters (step SA04).
- the value function updating unit 134 updates the value function stored in the value function storage unit 138 (step SA05).
- the decision-making unit 140 Based on the updated value function, the decision-making unit 140 obtains a combination of parameters considered to be more appropriate for the current machining, and outputs the obtained parameter combination.
- the action output unit 150 having received this input determines whether or not to continue parameter adjustment. Then, a command is issued to perform idling again according to the machining program with the set parameters (step SA06).
- the acceleration/deceleration adjusting device 1 having the above configuration sets the target value (shape error, positional deviation, etc.) of the actual machining accuracy/machined surface quality, thereby setting parameters more suitable for the machining accuracy/machined surface quality. Quantitative control becomes possible. By quantitatively controlling the combination of parameter setting values, it is possible to maintain a combination of parameter setting values suitable for a given machining purpose, and switch between them as appropriate according to the machining purpose. Furthermore, the optimization of the combination of parameter setting values by the acceleration/deceleration adjusting device 1 according to the present embodiment can be performed while observing the environment (control device, machine tool) from the outside. Therefore, there is no need to install new software on the environment side, and it can be used in a wide range of environments.
- the present invention is not limited to the above-described examples of the embodiments, and can be implemented in various modes by adding appropriate modifications.
- the evaluation of machining accuracy and machined surface quality is performed based on a predetermined remuneration calculation formula set in advance. good too.
- the range of each parameter output by the decision making unit 140 may be set in advance. By configuring in this way, it is possible to limit the search range of parameters.
- the decision-making unit 140 determines the combination of parameters to be output based on the value output by the value function. You may make it output the combination of the parameter created by this.
- FIG. 6 is a table showing parameter adjustment directions for solving the problems to be solved ascertained from the machining results, and their priorities. In this way, by storing the relationship between the problem to be solved and the parameter adjustment method as a rule, the decision-making unit 140 can refer to the rule and determine the parameter to be adjusted. .
- a weight can be set for the remuneration related to the machining accuracy and the machined surface quality calculated by the remuneration calculation unit 132 and the additional remuneration related to the machining time.
- FIG. 7 shows a configuration example when the learning unit 130 is removed.
- the value function is not updated, but by giving the initial parameters to the acceleration/deceleration adjusting device 1, the decision making unit 140 searches for more appropriate parameters, Adjustments of machine 3 parameters can be made.
Landscapes
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Manufacturing & Machinery (AREA)
- General Engineering & Computer Science (AREA)
- Numerical Control (AREA)
Abstract
Description
そのため、速度分布以外の基準に基づいて加工におけるパラメータの調整を行える技術が望まれている。
そして、本開示の一態様は、ワークの加工を行う工作機械が備える各軸のN階時間微分要素(Nは自然数)を含む制御周期毎の移動量の制御に係るパラメータを推定する機械学習装置であって、前記加工における加工精度及び加工面品位の少なくともいずれかに係る情報と、前記加工に掛かった加工時間とを、前記工作機械の動作状態を示すデータとして観測する状態観測部と、前記状態観測部が観測したデータに係る目標値を判定データとして取得する判定条件取得部と、前記状態観測部が観測したデータと、前記判定条件取得部が取得した前記判定データとに基づいて前記パラメータに基づく加工に対する報酬を算出する報酬計算部と、前記報酬に基づいて前記パラメータに基づく加工状態の価値を算出するための価値関数を更新する価値関数更新部と、更新された前記価値関数に基づいて前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する意思決定部と、を備える機械学習装置である。
本開示の他の態様は、ワークの加工を行う工作機械が備える各軸のN階時間微分要素(Nは自然数)を含む制御周期毎の移動量の制御に係るパラメータを推定する機械学習装置としてコンピュータを動作させるプログラムを記憶したコンピュータ読み取り可能な記憶媒体であって、前記加工における加工精度及び加工面品位の少なくともいずれかに係る情報と、前記加工に掛かった加工時間とを、前記工作機械の動作状態を示すデータとして観測する状態観測部と、前記状態観測部が観測したデータに係る目標値を判定データとして取得する判定条件取得部と、前記状態観測部が観測したデータと、前記判定条件取得部が取得した前記判定データとに基づいて前記パラメータに基づく加工に対する報酬を算出する報酬計算部と、前記報酬に基づいて前記パラメータに基づく加工状態の価値を算出するための価値関数を更新する価値関数更新部と、更新された前記価値関数に基づいて前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する意思決定部として、コンピュータを動作させるプログラムを記憶したコンピュータ読み取り可能な記憶媒体である。
図1は本発明の一実施形態による加減速調整装置の要部を示す概略的なハードウェア構成図である。本発明の加減速調整装置1は、例えば加工プログラムに基づいて工作機械を制御する制御装置として実装することができる。また、本発明の加減速調整装置1は、加工プログラムに基づいて工作機械を制御する制御装置に併設されたパソコンや、有線/無線のネットワークを介して制御装置と接続されたパソコン、セルコンピュータ、フォグコンピュータ6、クラウドサーバ7などのコンピュータ上に実装することができる。本実施形態では、加減速調整装置1を、ネットワークを介して工作機械を制御する制御装置と接続されたパソコンの上に実装した例を示す。
本実施形態の加減速調整装置1は、状態観測部110、判定条件取得部120、行動出力部150を備える。また、加減速調整装置1の機械学習装置100は、学習部130、意思決定部140を備える。更に、機械学習装置100のRAM103乃至不揮発性メモリ104上には、学習部130による機械学習の結果としての価値関数を記憶する価値関数記憶部138が予め用意されている。
報酬計算部132は、更に工作機械3における加工に掛かった加工時間を示す値と、加工目的に係る判定データに含まれる目標とする加工時間とを比較し、その比較結果に基づいて予め設定された所定の報酬算出式により追加の報酬を算出する。報酬計算部132は、加工に掛かった加工時間が目標とする加工時間内に収まる場合に高い報酬を算出する。また、報酬計算部132は、加工に掛かった加工時間が目標とする加工時間を超える場合に低い報酬を算出する。報酬計算部132は、加工に掛かった加工時間に収まる度合いに応じてより高い報酬を算出してもよい。また、報酬計算部132は、加工に掛かった加工時間を超える度合いに応じてより低い報酬を算出してもよい。報酬計算部132は、マイナスの報酬を算出してもよい。報酬計算部132は、このようにして算出した追加の報酬を、加工精度や加工面品位に基づいて算出した報酬に加算する。
なお、加工時間を報酬として考慮する場合、報酬計算部132は、加工精度や加工面品位が許容値内に収まる場合における加工時間を記憶するようにしてもよい。この時、記憶した加工時間の内で、最も短い加工時間を加工目的に係る判定データとする。そして、報酬計算部132は、加工精度や加工面品位が許容値内に収まる場合にのみ、その時の加工時間と上記した記憶している最も短い加工時間を報酬算出の基準とした上で報酬を算出する。このようにすることで、目標とする加工精度や加工面品位の範囲内で、最も加工時間が短くなるパラメータを探索できるようになる。
例えば、上記した実施形態では、加工精度や加工面品位の評価を予め設定された所定の報酬算出式に基づいて行っているが、評価に係る評価用プログラムを外部から登録できるように構成してもよい。このように構成することで、加工の内容が変わって学習したいパラメータが追加された場合等においても、四角コーナやR角コーナなどの専用の評価用プログラムを提供することで効率よく加工精度・加工面品位を評価することができるようになる。
報酬計算部132が算出する加工精度や加工面品位に係る報酬と、加工時間に係る追加の報酬とに対して、重みを設定できるように構成してもよい。このように構成することで、加工品質を重視する場合には加工精度や加工面品位に係る報酬の重みを増加させ、一方で加工時間を重視する場合には加工時間に係る報酬の重みを増加させる、といったように、加工の目的に合わせた微調整を行うことが可能となる。
1 加減速調整装置
3 工作機械
4 センサ
5 ネットワーク
6 フォグコンピュータ
7 クラウドサーバ
11 CPU
12 ROM
13 RAM
14 不揮発性メモリ
15 インタフェース
17,18,20,21 インタフェース
22 バス
70 表示装置
71 入力装置
72 外部機器
100 機械学習装置
101 プロセッサ
102 ROM
103 RAM
104 不揮発性メモリ
110 状態観測部
120 判定条件取得部
130 学習部
132 報酬計算部
134 価値関数更新部
138 価値関数記憶部
140 意思決定部
150 行動出力部
Claims (9)
- [規則91に基づく訂正 02.06.2023]
ワークの加工を行う工作機械が備える各軸のN階時間微分要素(Nは自然数)を含む制御周期毎の移動量の制御に係るパラメータを推定する機械学習装置であって、
前記加工における加工精度及び加工面品位の少なくともいずれかに係る情報と、前記加工に掛かった加工時間とを、前記工作機械の動作状態を示すデータとして観測する状態観測部と、
前記状態観測部が観測したデータに係る目標値を判定データとして取得する判定条件取得部と、
前記状態観測部が観測したデータと、前記判定条件取得部が取得した前記判定データとに基づいて前記パラメータに基づく加工に対する報酬を算出する報酬計算部と、
前記報酬に基づいて前記パラメータに基づく加工状態の価値を算出するための価値関数を更新する価値関数更新部と、
更新された前記価値関数に基づいて前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する意思決定部と、
を備える機械学習装置。 - 前記加工精度及び加工面品位の少なくともいずれかを評価可能な評価用プログラムを登録可能であり、該評価用プログラムを用いて前記加工精度及び加工面品位の少なくともいずれかに係る報酬を算出する、
請求項1に記載の機械学習装置。 - 前記意思決定部は、更新された前記価値関数に基づいて前記加工における加工時間がより短い、前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する、
請求項1に記載の機械学習装置。 - ワークの加工を行う工作機械が備える各軸のN階時間微分要素(Nは自然数)を含む制御周期毎の移動量の制御に係るパラメータを調整する加減速調整装置であって、
前記加工における加工精度及び加工面品位の少なくともいずれかに係る情報と、前記加工に掛かった加工時間とを、前記工作機械の動作状態を示すデータとして観測する状態観測部と、
前記状態観測部が観測したデータに係る目標値を判定データとして取得する判定条件取得部と、
前記パラメータに基づく加工状態の価値を算出するための価値関数を記憶する価値関数記憶部と、
前記価値関数に基づいて前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する意思決定部と、
前記意思決定部が出力した前記パラメータの設定値の組み合わせに基づいて、前記工作機械の前記パラメータを調整する行動出力部と、
を備える加減速調整装置。 - 前記判定条件取得部は、複数の目標値を設定可能である、
請求項1または3に記載の加減速調整装置。 - 前記意思決定部が出力する前記パラメータの設定値の範囲を設定可能である、
請求項1または3に記載の加減速調整装置。 - 前記パラメータの調整回数を設定可能である、
請求項1または3に記載の加減速調整装置。 - [規則91に基づく訂正 02.06.2023]
ワークの加工を行う工作機械が備える各軸のN階時間微分要素(Nは自然数)を含む制御周期毎の移動量の制御に係るパラメータを推定する機械学習装置としてコンピュータを動作させるプログラムを記憶したコンピュータ読み取り可能な記憶媒体であって、
前記加工における加工精度及び加工面品位の少なくともいずれかに係る情報と、前記加工に掛かった加工時間とを、前記工作機械の動作状態を示すデータとして観測する状態観測部と、
前記状態観測部が観測したデータに係る目標値を判定データとして取得する判定条件取得部と、
前記状態観測部が観測したデータと、前記判定条件取得部が取得した前記判定データとに基づいて前記パラメータに基づく加工に対する報酬を算出する報酬計算部と、
前記報酬に基づいて前記パラメータに基づく加工状態の価値を算出するための価値関数を更新する価値関数更新部と、
更新された前記価値関数に基づいて前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する意思決定部と
してコンピュータを動作させるプログラムを記憶したコンピュータ読み取り可能な記憶媒体。 - ワークの加工を行う工作機械が備える各軸のN階時間微分要素(Nは自然数)を含む制御周期毎の移動量の制御に係るパラメータを調整する加減速調整装置としてコンピュータを動作させるプログラムを記憶したコンピュータ読み取り可能な記憶媒体であって、
前記加工における加工精度及び加工面品位の少なくともいずれかに係る情報と、前記加工に掛かった加工時間とを、前記工作機械の動作状態を示すデータとして観測する状態観測部と、
前記状態観測部が観測したデータに係る目標値を判定データとして取得する判定条件取得部と、
前記パラメータに基づく加工状態の価値を算出するための価値関数を記憶する価値関数記憶部と、
前記価値関数に基づいて前記加工により適している前記パラメータの設定値の組合せを推定し、推定した前記パラメータの設定値の組み合わせを出力する意思決定部と、
前記意思決定部が出力した前記パラメータの設定値の組み合わせに基づいて、前記工作機械の前記パラメータを調整する行動出力部と
してコンピュータを動作させるプログラムを記憶したコンピュータ読み取り可能な記憶媒体。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/016479 WO2022224450A1 (ja) | 2021-04-23 | 2021-04-23 | 機械学習装置、加減速調整装置及びコンピュータ読み取り可能な記憶媒体 |
JP2023516011A JPWO2022224450A1 (ja) | 2021-04-23 | 2021-04-23 | |
US18/551,192 US20240176309A1 (en) | 2021-04-23 | 2021-04-23 | Machine learning device, acceleration and deceleration adjustment device, and computer-readable storage medium |
CN202180097101.3A CN117203591A (zh) | 2021-04-23 | 2021-04-23 | 机器学习装置、加减速调整装置以及计算机可读取的存储介质 |
DE112021007088.6T DE112021007088T5 (de) | 2021-04-23 | 2021-04-23 | Maschinenlernvorrichtung, beschleunigungs- und verlangsamungseinstellvorrichtung und computerlesbares speichermedium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2021/016479 WO2022224450A1 (ja) | 2021-04-23 | 2021-04-23 | 機械学習装置、加減速調整装置及びコンピュータ読み取り可能な記憶媒体 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2022224450A1 WO2022224450A1 (ja) | 2022-10-27 |
WO2022224450A9 true WO2022224450A9 (ja) | 2023-09-07 |
Family
ID=83723463
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/016479 WO2022224450A1 (ja) | 2021-04-23 | 2021-04-23 | 機械学習装置、加減速調整装置及びコンピュータ読み取り可能な記憶媒体 |
Country Status (5)
Country | Link |
---|---|
US (1) | US20240176309A1 (ja) |
JP (1) | JPWO2022224450A1 (ja) |
CN (1) | CN117203591A (ja) |
DE (1) | DE112021007088T5 (ja) |
WO (1) | WO2022224450A1 (ja) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118534842A (zh) * | 2024-07-24 | 2024-08-23 | 通用技术集团机床工程研究院有限公司 | 速度规划方法、插补点确定方法、装置及相关设备 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017030067A (ja) * | 2015-07-30 | 2017-02-09 | ファナック株式会社 | 加工時間測定機能とオンマシン測定機能を有する制御装置付き加工装置 |
JP6219897B2 (ja) * | 2015-09-28 | 2017-10-25 | ファナック株式会社 | 最適な加減速を生成する工作機械 |
JP6625914B2 (ja) * | 2016-03-17 | 2019-12-25 | ファナック株式会社 | 機械学習装置、レーザ加工システムおよび機械学習方法 |
JP6499710B2 (ja) * | 2017-04-20 | 2019-04-10 | ファナック株式会社 | 加減速制御装置 |
-
2021
- 2021-04-23 WO PCT/JP2021/016479 patent/WO2022224450A1/ja active Application Filing
- 2021-04-23 JP JP2023516011A patent/JPWO2022224450A1/ja active Pending
- 2021-04-23 CN CN202180097101.3A patent/CN117203591A/zh active Pending
- 2021-04-23 DE DE112021007088.6T patent/DE112021007088T5/de active Pending
- 2021-04-23 US US18/551,192 patent/US20240176309A1/en active Pending
Also Published As
Publication number | Publication date |
---|---|
US20240176309A1 (en) | 2024-05-30 |
CN117203591A (zh) | 2023-12-08 |
DE112021007088T5 (de) | 2023-12-07 |
JPWO2022224450A1 (ja) | 2022-10-27 |
WO2022224450A1 (ja) | 2022-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9964931B2 (en) | Numerical controller with machining condition adjustment function which reduces chatter or tool wear/breakage occurrence | |
CN109960219B (zh) | 控制装置以及机器学习装置 | |
CN109581962B (zh) | 数值控制系统 | |
JP6219897B2 (ja) | 最適な加減速を生成する工作機械 | |
US10216169B2 (en) | Automatic parameter adjustment device for adjusting parameter corresponding to processing condition | |
US11640557B2 (en) | Machine learning device, numerical control system, and machine learning method | |
JP6499710B2 (ja) | 加減速制御装置 | |
JP6756676B2 (ja) | 製造システム | |
CN109613886B (zh) | 热位移校正系统 | |
JP6557285B2 (ja) | 制御装置及び機械学習装置 | |
US20230271276A1 (en) | Method, Control Unit and Laser Cutting System for Combined Path and Laser Process Planning for Highly Dynamic Real-Time Systems | |
KR102224970B1 (ko) | 제어 장치 및 기계 학습 장치 | |
US20210174227A1 (en) | Machine learning apparatus, control device, machining system, and machine learning method for learning correction amount of workpiece model | |
US20170090452A1 (en) | Machine tool for generating speed distribution | |
US11698616B2 (en) | Managing apparatus and managing system | |
WO2022224450A9 (ja) | 機械学習装置、加減速調整装置及びコンピュータ読み取り可能な記憶媒体 | |
CN114424129A (zh) | 控制系统 | |
JP6618656B1 (ja) | 保守支援システム、数値制御装置および保守支援システムの制御方法 | |
US20240033873A1 (en) | Device and method for machining a workpiece | |
JP7504687B2 (ja) | 切削加工支援システム | |
US20200183352A1 (en) | Machine tool | |
US20200089175A1 (en) | Characteristic judgment apparatus, characteristic judgment method, and charcteristic judgment program | |
WO2024122066A1 (ja) | パラメータ調整装置及びコンピュータ読み取り可能な記録媒体 | |
WO2023153446A1 (ja) | 提案装置、提案システム、提案方法、及び、プログラム | |
CN118905724A (zh) | 用于火箭箱底镜像铣削的数据驱动自适应壁厚控制方法和系统 |
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: 21937941 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023516011 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18551192 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 112021007088 Country of ref document: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202180097101.3 Country of ref document: CN |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21937941 Country of ref document: EP Kind code of ref document: A1 |