Parameter adjustment device and computer-readable recording medium

Abstract

A parameter adjustment device according to the present disclosure is provided with: a simulation unit that simulates the operation of a control target, the value of a prescribed control parameter of which is set, on the basis of a control target model that models the operation of each part of the control target; a first search unit that performs simulation processing while changing the value of the control parameter within a prescribed first range, and searches for a candidate set of values of the control parameter near the boundary of a prescribed constraint condition on the basis of a first index value calculated from the simulation processing results; a candidate implementation unit that sets, for the control target, the candidate set of values of the control parameter, verifies the operation of the control target, and calculates a second index value on the basis of the verification results; and a search range determination unit that determines a second range narrower than the first range on the basis of the first index value and the second index value.

Description

Parameter adjustment device and computer-readable recording medium

[0001] The present disclosure relates to a parameter adjustment device and a computer-readable recording medium.

[0002] One method for adjusting control parameters set when controlling industrial machinery is to utilize simulation. In this method, a predetermined first range for the control parameter is first determined in advance, and simulations are repeatedly performed while changing the value of the control parameter within the first range. Then, optimal value candidates for the control parameter are extracted based on the results of the simulation. Then, a predetermined second range is determined near each of the extracted candidate values ​​for the control parameter. Then, the machine is repeatedly operated while adjusting the control parameter within the second range. Based on the results of this operation, the optimal value of the control parameter is determined (see, for example, Patent Document 1).

[0003] JP 2017-102619 A

[0004] Setting a narrow search range for control parameters may result in an inability to obtain appropriate control parameters. However, setting a wide search range for control parameters will require a significant amount of time for simulation calculations and machine-based operations. This places a heavy burden on workers, particularly when adjusting control parameters by actually operating the machine. Therefore, there is a need for technology that can appropriately narrow down the search range for control parameters.

[0005] One aspect of the present disclosure is a parameter adjusting device including: a controlled object model that models the operation of each part of an industrial machine that is a controlled object; a simulation unit that simulates the operation of the controlled object, to which predetermined control parameter values ​​are set, based on the controlled object model; a first search unit that repeatedly causes the simulation unit to perform a simulation process while changing the value of the control parameter within a predetermined first range, and searches for candidate sets of control parameter values ​​that will yield the first index value near a boundary of a predetermined constraint, based on a first index value calculated from the results of the simulation process; a candidate implementation unit that sets the candidate sets of control parameter values ​​searched for by the first search unit as the controlled object, performs operation verification, and calculates a second index value based on the results of the operation verification; and a search range determination unit that determines a second range of values ​​that is narrower than the first range, based on the first index value and the second index value.

[0006] Fig. 1 is a schematic hardware configuration diagram of a parameter adjustment device according to a first embodiment of the present disclosure; Fig. 2 is a flowchart showing a schematic block diagram of functions provided in the parameter adjustment device according to the first embodiment of the present disclosure, illustrating a schematic flow of processing related to preparation for a search; Fig. 3 is a flowchart showing a schematic flow of a first search process; Fig. 4 is a flowchart showing a schematic flow of a second range determination process; and Fig. 5 is a flowchart showing a schematic flow of a second search process.

[0007] Hereinafter, each embodiment of the present disclosure will be described with reference to the drawings. [First Embodiment] Fig. 1 is a schematic hardware configuration diagram showing the main parts of a parameter adjustment device according to one embodiment of the present disclosure. The parameter adjustment device 1 according to this embodiment can be implemented, for example, as a control device that controls industrial machinery based on a control program. The parameter adjustment device 1 can also be implemented, for example, as a personal computer installed alongside a control device that controls industrial machinery, or as another computer such as a computer, fog computer, or cloud server connected to the control device via a wired or wireless network. In this embodiment, an example is shown in which the parameter adjustment device 1 is implemented as a computer connected to a control device that controls industrial machinery 4 via a network 5.

[0008] The CPU 11 provided in the parameter adjustment device 1 according to this embodiment is a processor that provides overall control of the parameter adjustment device 1. The CPU 11 reads a system program stored in the ROM 12 via the bus 22 and controls the entire parameter adjustment device 1 in accordance with the system program. The RAM 13 temporarily stores temporary calculation data, display data, and various data input from outside.

[0009] The nonvolatile memory 14 is composed of, for example, a battery-backed memory or an SSD (Solid State Drive), not shown, and retains its stored state even when the parameter adjustment device 1 is powered off. The nonvolatile memory 14 stores programs and data read from the external device 72 via the interface 15, programs and data input via the input device 71, and programs and data acquired from the industrial machine 4 or other devices via the network 5. The stored data may include, for example, data related to physical quantities such as motor current, voltage, torque, position, speed, and acceleration of the drive unit detected by a sensor 8 attached to the industrial machine 4. The programs and data stored in the nonvolatile memory 14 may be loaded into the RAM 13 during execution / use. Furthermore, various system programs, such as known analysis programs, are pre-written in the ROM 12.

[0010] The interface 15 is an interface for connecting the CPU 11 of the parameter adjustment device 1 to an external device 72 such as a USB device. For example, system programs, programs related to the operation of the industrial machine 4, setting data, etc. are read from the external device 72. In addition, programs and setting data created and edited within the parameter adjustment device 1 can be stored in external storage means via the external device 72.

[0011] The interface 20 is an interface for connecting the CPU 11 of the parameter adjustment device 1 to a wired or wireless network 5. The network 5 may communicate using technologies such as serial communication such as RS-485, Ethernet (registered trademark), optical communication, wireless LAN, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc. The network 5 is connected to a control device that controls the industrial machine 4, a fog computer 6, a cloud server 7, etc., and exchanges data with the parameter adjustment device 1.

[0012] The display device 70 displays various data read into the memory, data obtained as a result of executing programs, etc., output via the interface 17. An input device 71, which is comprised of a keyboard, pointing device, etc., transmits instructions, data, etc., based on operations by an operator to the CPU 11 via the interface 18.

[0013] 2 is a schematic block diagram showing the functions of the parameter adjustment node 1 according to the first embodiment of the present disclosure. Each function of the parameter adjustment node 1 according to this embodiment is realized by the CPU 11 of the parameter adjustment node 1 shown in FIG. 1 executing a system program and controlling the operation of each part of the parameter adjustment node 1.

[0014] The parameter adjustment device 1 of this embodiment includes a simulation unit 100, a first search unit 110, a candidate implementation unit 120, a search range determination unit 130, a second search unit 140, and an output unit 150. The RAM 13 to the nonvolatile memory 14 of the parameter adjustment device 1 also include a model storage unit 210, which is an area in which a control object model that models the operation of the industrial machine 4 to be controlled is stored in advance. The RAM 13 to the nonvolatile memory 14 of the parameter adjustment device 1 also include an evaluation program 220 to be used for simulation processing and operation control of the actual control object.

[0015] The simulation unit 100 executes a predetermined simulation process for simulating the operation of the controlled object based on the controlled object model stored in the model storage unit 210. The controlled object model stored in the model storage unit 210 is a model of the operation of each part of the industrial machine 4, including drive parts such as servo motors, transmission parts such as ball screws, and movable parts such as tables. The controlled object model is expressed using mathematical expressions, transfer functions, block diagrams, etc. that reflect the characteristics of the industrial machine 4 that is the controlled object. By preparing the controlled object model, it is possible to simulate the operation of the industrial machine 4 that is the controlled object.

[0016] [Correction based on Rule 91 07.01.2025] The simulation process executed by the simulation unit 100 virtually executes the operation of the industrial machine 4, which is the control target, by calculation based on a control target model when predetermined command information is issued to the industrial machine 4. As a result of the simulation process, virtual feedback information of the industrial machine 4, which is the control target, is generated. Examples of the virtual feedback information include virtual physical quantities related to the operation of the industrial machine 4, such as the position, speed, acceleration, and torque of the motor for each operation cycle.

[0017] When performing simulation processing, the simulation unit 100 reflects the given control parameters in calculations related to the controlled object model. The control parameters differ depending on the type of industrial machine 4 that is the controlled object. For example, if the controlled industrial machine 4 is a machine tool operated by a servo motor, the control parameters include setting values ​​related to each motor, such as linear acceleration, linear jerk, corner speed difference, post-interpolation acceleration / deceleration time constant, speed gain, position loop gain, and feedforward coefficient. Furthermore, if the controlled industrial machine is an electric discharge machine, the control parameters include setting values ​​related to the power supply, such as electric discharge machining pulse voltage and electric discharge machining current, in addition to the above.

[0018] The first search unit 110 repeatedly performs simulation processing by the simulation unit 100 while varying the control parameters within a first range related to the control parameters. Then, based on the results of the simulation processing, the first search unit 110 searches for a most suitable set of control parameter values ​​within the first range in which a predetermined index value satisfies a predetermined constraint, as well as sets of control parameter values ​​nearby the most suitable set, as candidate sets of control parameter values. The predetermined index value is an index used for constraints and evaluations related to the operation results of the controlled object. Examples of index values ​​include machining accuracy (machining error), machined surface quality (position deviation, vibration), and machining time (cycle time).

[0019] The first range is a range of values ​​that each control parameter can take in the simulation process. The first range may be set by an operator, or may be determined according to a predetermined rule based on the specifications of the industrial machine 4 to be controlled, command information included in the evaluation program 220 used in the simulation, and the like. The first range is a predetermined range within which an index value used to evaluate the results of the simulation process can be calculated. Because the first range is a search range in the simulation process by the simulation unit 100, there is no problem in setting it to a relatively wide range. For example, the first range may be the minimum to maximum values ​​that can be set for each control parameter, or the range of values ​​that can be set based on the command information.

[0020] The first search unit 110 causes the simulation unit 100 to execute a simulation process based on predetermined command information included in the evaluation program 220 while varying the values ​​of each control parameter within a first range. Then, as a result, virtual feedback information is acquired, and a predetermined index value is evaluated based on the acquired virtual feedback information. The first search unit 110 executes this process using a known algorithm, such as machine learning search (e.g., reinforcement learning or Bayesian optimization), rule-based search, or a predetermined exhaustive search algorithm. When appropriate constraints and evaluation methods are set, the search by the first search unit 110 results in a set of values ​​of at least one or more control parameters near the boundaries of the constraints. The first search unit 110 may store the results of the simulation process (index values), the determination results of the constraints, and the evaluation results for the sets of values ​​of the control parameters evaluated. The first search unit 110 outputs to the candidate implementation unit 120 a set of the best control parameter values ​​within a range that satisfies one set of constraint conditions, or two sets of control parameter values ​​that are near the boundary of the constraint conditions, where a first set of control parameter values ​​satisfies the constraint conditions and a second set of control parameter values ​​that do not satisfy the constraint conditions.

[0021] The candidate implementation unit 120 sets, for the industrial machine 4 to be controlled, a set of control parameter values ​​near the boundary of the constraint conditions found by the first search unit 110, and commands the industrial machine 4 to perform operation verification based on the command information included in the evaluation program 220. Then, as a result of this operation, feedback information related to the operation is obtained from the industrial machine 4. An index value is then calculated to evaluate the obtained feedback information.

[0022] The search range determination unit 130 determines a second range by correcting the first range based on the operational results of the candidate implementation unit 120 verifying the operation of the industrial machine 4, which is the actual control target, for the set of control parameter values ​​near the boundary of the constraint condition searched by the first search unit 110.

[0023] When the first search unit 110 finds only one set of control parameter values ​​near the boundary of the constraint condition, the search range determination unit 130 may determine a predetermined range based on the set of control parameter values ​​as the second range. For example, the second range may be a range from the lower limit of the first range for each control parameter to the value of the control parameter near the boundary of the constraint condition. Alternatively, the second range may be a range from the upper limit of the first range for each control parameter to the value of the control parameter near the boundary of the constraint condition. When the candidate implementation unit 120 determines that the results of the operation verification of the industrial machine 4, which is the actual control target, do not satisfy the constraint condition, the second range for each control parameter may be determined based on the value of the control parameter near the boundary of the constraint condition, in a direction that satisfies the constraint condition. Alternatively, when the constraint condition is not satisfied, the second range may be determined based on the value of the control parameter near the boundary of the constraint condition, in a direction that improves the evaluation.

[0024] When the first search unit 110 finds two sets of control parameter values ​​near the boundary of the constraint condition, the search range determination unit 130 may determine the second range based on the results of simulating each of the sets of control parameter values ​​near the boundary of the constraint condition and the results of operational verification of the industrial machine 4, which is the actual controlled object. For example, when the results of operational verification of the industrial machine 4, which is the actual controlled object, show that the constraint condition is satisfied regardless of the set of control parameter values ​​used, the search range determination unit 130 calculates, as the modeling error, the difference in index values ​​between the results of simulating the second set of control parameter values ​​and the results of operational verification of the industrial machine 4, which is the actual controlled object. Next, the search range determination unit 130 selects a third set of control parameter values ​​that does not satisfy the constraint condition from the sets of control parameter values ​​found by the first search unit 110, taking the calculated modeling error into consideration. Then, the range between the selected third set of control parameter values ​​and the second set of control parameter values ​​may be determined as the second range.

[0025] Furthermore, if the results of verifying operation on the industrial machine 4 that is the actual controlled object show that the index value satisfies the constraint condition when operated with one set of control parameter values, but does not satisfy the constraint condition when operated with the other set of control parameter values, then the range between the first set of control parameter values ​​and the second set of control parameter values ​​can be defined as the second range.

[0026] Furthermore, if the result of verifying operation using the industrial machine 4 that is the actual control target shows that the constraint conditions are not satisfied regardless of the set of control parameter values ​​used, the search range determination unit 130 calculates, as a modeling error, the difference in index values ​​between the processing result of the simulation processing for the first set of control parameter values ​​and the result of verifying operation using the industrial machine 4 that is the actual control target. Next, taking the calculated modeling error into consideration, the first search unit 110 selects a third set of control parameter values ​​that satisfies the constraint conditions from the sets of control parameter values ​​searched for. Then, the range between the selected set of third control parameter values ​​and the set of first control parameter values ​​can be determined as a second range.

[0027] The second search unit 140 searches for a suitable set of control parameters while changing the control parameters within the second range determined by the search range determination unit 130. According to this embodiment, the second search unit 140 repeatedly operates the industrial machine 4, which is the actual controlled object, in accordance with the command information included in the evaluation program 220. Then, based on the results of the operation of the industrial machine 4, which is the actual controlled object, the second search unit 140 searches for the most suitable set of control parameter values ​​within the second range within which the index values ​​satisfy the constraint conditions. The second search unit 140 performs this process using known algorithms, such as machine learning search (e.g., reinforcement learning and Bayesian optimization), rule-based search, and a predetermined exhaustive search algorithm. The second search unit 140 operates the industrial machine 4, which is the actual controlled object, and calculates index values ​​from the operation results. Then, the second search unit 140 determines whether the calculated index values ​​satisfy the constraint conditions and performs evaluation using the index values. The operation of the industrial machine 4 for which a set of control parameter values ​​is set takes the same time as the actual cycle time. However, since the second range determined by the search range determination unit 130 is sufficiently narrow compared to the first range, the search does not take an enormous amount of time. The second search unit 140 searches for the best set of control parameter values ​​that satisfies at least one set of constraints. The second search unit 140 outputs the searched set of control parameter values ​​to the output unit 150.

[0028] The output unit 150 outputs the set of control parameter values ​​searched by the second search unit 140 to a predetermined device, equipment, etc. The output unit 150 may, for example, display the set of control parameter values ​​on the display device 70, or may record the set of control parameter values ​​in a predetermined area provided on the non-volatile memory 14. In addition, the set of control parameter values ​​may be transmitted to a control device of the industrial machine 4 for setting, or may be transmitted to a higher-level computer such as the fog computer 6 or the cloud server 7.

[0029] 3 to 6, a specific example of operation of the parameter adjustment device 1 according to this embodiment having the above configuration will be described. In this example, the industrial machine 4 is a machine tool that operates each axis with a servo motor. The linear acceleration [mm / sec] of each axis of the industrial machine 4 is2 ] and corner speed difference [mm / sec] are the control parameters to be searched. Furthermore, the maximum value of machining error and cycle time are selected as index values, and the constraint condition is that the maximum value of machining error is equal to or less than a predetermined target error, and the evaluation method is such that the shorter the cycle time, the better. Note that the index value may also be the machined surface quality, in which case the predetermined value of machined surface quality may be set as the constraint condition. Machining error and cycle time are in a trade-off relationship. Attempting to reduce machining error tends to increase cycle time, while attempting to shorten cycle time increases machining error. When the machining error is within the target error, a solution that minimizes the cycle time exists for any combination of control parameter values ​​near the boundary.

[0030] 3 is a flowchart showing a general flow of processing related to preparation for search. First, the simulation unit 100 reads from the model storage unit 210 a model of a machine tool in which each axis is operated by a servo motor in accordance with an operator's instructions (step SA01). Next, the first search unit 110 selects linear acceleration and corner speed difference as control parameters to be searched in accordance with the operator's instructions (step SA02). Furthermore, a constraint condition is set to limit the machining error to a target error of 10 μm or less, and an evaluation method is set so that the shorter the cycle time, the higher the evaluation (step SA03).

[0031] Next, the first search unit 110 sets a first range for each control parameter (step SA04). The first range is 10 to 10,000 [mm / sec] for linear acceleration. 2 ], and the corner speed difference is set to 10 to 5000 [mm / sec]. This setting may be made by an operator's instruction. Alternatively, the maximum and minimum values ​​of each control parameter may be inquired of the control device of the industrial machine 4 to be controlled, and the maximum and minimum values ​​may be set according to the response.

[0032] 4 is a flowchart showing the general flow of the first search process. After the above settings are completed and the operator issues a command to start the search, the first search unit 110 selects initial values ​​for each control parameter from within the first range and sets them for the simulation model retrieved by the simulation unit 100 (step SB01). Next, the first search unit 110 sets the evaluation program 220 to be executed in the simulation process for the simulation unit 100 (step SB02). The first search unit 110 then commands the selected simulation model to execute the simulation process using the set control parameter values.

[0033] The simulation unit 100 executes the simulation process in accordance with the command from the first search unit 110 (step SB03). When the simulation process is completed, the first search unit 110 acquires the processing results from the simulation unit 100 (step SB04). The acquired processing results are the machining path and cycle time used to calculate the index value. The first search unit 110 calculates the index value based on the acquired processing results (step SB05). The machining error, which is the index value, can be calculated as the maximum difference between the path commanded by the command information and the machining path in the virtual machining process in the simulation process. The cycle time acquired as the processing result is used as the index value. If the search of the first range is not completed at this point (No in step SB06), the first search unit 110 adjusts the values ​​of the control parameters and sets them in the simulation model (step SB07). Then, the process proceeds to step SB03. On the other hand, if the search of the first range has ended at this point (Yes in step SB06), control parameters near the boundary of the constraint are selected as candidates from the searched control parameters (step SB08). The processes in steps SB03 to SB07 are performed using a known search algorithm such as machine learning search, rule-based search, a predetermined exhaustive search algorithm, or Bayesian optimization, as described above.

[0034] Here, as a result of the first search process, a set A of values ​​of the first control parameters that satisfy the constraint conditions is selected as a candidate of the control parameters (linear acceleration=2300 [mm / sec] 2 ], corner speed difference = 1100 [mm / sec], at this time, machining error 9 [μm], cycle time 35.2 [sec]), and a set B of values ​​of the second control parameters that do not satisfy the constraints (linear acceleration = 2400 [mm / sec] 2 ], corner speed difference=1200 [mm / sec], machining error is 12 [μm], and cycle time is 33.0 [sec]) is selected by the first search unit 110.

[0035] 5 is a flowchart showing a schematic flow of the second range determination process. Once candidate sets of control parameter values ​​are selected, the candidate implementation unit 120 sets the selected candidate control parameter values ​​for the industrial machine 4, which is the actual controlled object, performs operation verification based on the command information included in the evaluation program 220, and acquires operation results when the industrial machine 4 is operated with each candidate set of control parameter values ​​(step SC01). The acquired operation results are the machining path and cycle time used to calculate the index value. The candidate implementation unit 120 calculates the index value based on the acquired operation results (step SC02). The machining error, which is the index value, can be calculated as the maximum value of the difference between the path commanded by the command information and the actual machining path. The cycle time acquired as the processing result is used as the index value.

[0036] Next, the search range determination unit 130 determines a second range by correcting the first range based on the index value calculated by the candidate implementation unit 120 (step SC03). For example, suppose that the first control parameter value set A is set for the industrial machine 4, which is the actual controlled object, and operational verification is performed, resulting in a machining error of 12 μm as an index value. Furthermore, suppose that the second control parameter value set B is set for the industrial machine 4, which is the actual controlled object, and operational verification is performed, resulting in a machining error of 14 μm as an index value. Neither the first control parameter value set nor the second control parameter value set satisfies the constraint. In such a case, the search range determination unit 130 calculates the difference in index value between the result of a simulation process performed for the first control parameter value set A and the result of operational verification performed on the industrial machine 4, which is the actual controlled object. Here, when the first control parameter value set A is set and the simulation process is performed, the machining error is 9 μm. Furthermore, when the control parameter set was set to the actual industrial machine 4 to be controlled and its operation was verified, the machining error as an index value was 12 μm, with the difference being 3 μm. This value is estimated to be a modeling error that occurred when the actual industrial machine 4 to be controlled was modeled. Therefore, candidate sets of control parameter values ​​are reselected taking into account the constraints and the modeling error. In this case, the target error set as the constraint is 10 μm, and the modeling error is 3 μm. Therefore, a set of control parameter values ​​that, in the results of the simulation process, results in a machining error of 10 μm - 3 μm = 7 μm or less is likely to satisfy the constraints even when set to the actual industrial machine 4 to be controlled and operated. Therefore, from among the sets of control parameter values ​​that, in the results of the simulation process, results in a machining error of 7 μm or less, the set with the shortest cycle time is selected as the third control parameter set C. Thereafter, when the third control parameter value set C is selected, the candidate implementation unit 120 performs operation verification for the third control parameter value set C on the industrial machine 4 that is the actual control target.Then, if the index values ​​calculated from the verification results satisfy the constraint conditions, the range between the third control parameter value set C and the first control parameter value set A can be determined as the second range. On the other hand, if the constraint conditions are not satisfied, a set of control parameter values ​​with an even smaller machining error can be selected as the third control parameter value set C, and the same process can be repeated. Following the above-mentioned process flow, for example, the third control parameter value set C can be set to linear acceleration = 2000 [mm / sec]. 2 ], corner speed difference=900 [mm / sec] (at this time, the machining error is 7 [μm], and the cycle time is 36.0 [sec]) is selected as the third control parameter value set C. At this time, the search range determination unit 130 sets the range of linear acceleration to 2000 to 2300 [mm / sec]. 2 ], and the range of the corner speed difference is determined to be 900 to 1100 [mm / sec], which is the second range.

[0037] Note that a case where the constraints are satisfied for both the first set of control parameter values ​​and the second set of control parameter values ​​will also be mentioned. For example, suppose that the first set of control parameter values ​​A is set for the industrial machine 4, which is the actual controlled object, and the operational verification results in a machining error of 7 μm as an index value. Furthermore, suppose that the second set of control parameter values ​​B is set for the industrial machine 4, which is the actual controlled object, and the operational verification results in a machining error of 9 μm as an index value. This satisfies the constraints for both the first set of control parameter values ​​and the second set of control parameter values. In such a case, the search range determination unit 130 calculates the difference in index values ​​between the result of a simulation process for the second set of control parameter values ​​B and the result of operational verification using the industrial machine 4, which is the actual controlled object. Here, when the second set of control parameter values ​​B was set and the simulation process was performed, the machining error was 12 μm. Furthermore, when the second set of control parameter values ​​B was set and the operational verification results were performed using the industrial machine 4, which is the actual controlled object, the machining error was 9 μm as an index value, and the difference therebetween is 3 μm. This value is estimated to be a modeling error that occurred when the actual industrial machine to be controlled was modeled. Therefore, a candidate set of control parameter values ​​is reselected taking into account the constraints and the modeling error. In this case, the target error set as the constraint is 10 μm, and the modeling error is 3 μm. Therefore, if the simulation results show that a control parameter value set produces a machining error of 10 μm + 3 μm = 13 μm or more, the constraint is likely not satisfied even when set and operated in the actual industrial machine 4 to be controlled. Therefore, among the control parameter value sets that produce a machining error of 13 μm or more in the simulation results, the one with the shortest cycle time is selected as the third control parameter value set C. After selecting the third control parameter value set C, the candidate implementation unit 120 then performs operation verification of the third control parameter value set C on the actual industrial machine 4 to be controlled.If the index values ​​calculated from the verification results do not satisfy the constraint conditions, the range between the third control parameter value set C and the second control parameter value set B can be determined as the second range. On the other hand, if the constraint conditions are satisfied, a set of control parameter values ​​with an even larger machining error can be reselected as the third control parameter value set C, and the same process can be repeated.

[0038] 6 is a flowchart showing a schematic flow of the second search process. Once the above process is completed and the second range is determined, the second search unit 140 selects an initial value for each control parameter from the second range and sets it for the industrial machine 4, which is the actual controlled object (step SD01). Next, the second search unit 140 sets the evaluation program 220 to be executed for the industrial machine 4, which is the actual controlled object (step SD02). Then, the second search unit 140 instructs the industrial machine 4, which is the actual controlled object, to operate with the set control parameter values.

[0039] The industrial machine 4, which is the actual controlled object, performs an operation based on the command information included in the evaluation program 220 (step SD03). When the operation based on the command information of the industrial machine 4, which is the actual controlled object, is completed, the second search unit 140 acquires an operation result from the industrial machine 4, which is the actual controlled object (step SD04). The acquired operation result is the machining path and cycle time used to calculate the index value. The second search unit 140 calculates an index value based on the acquired operation result (step SD05). The machining error, which is the index value, can be calculated as the maximum difference between the path commanded by the evaluation program 220 and the actual machining path. The cycle time acquired as the processing result is used as the index value. If the search of the second range is not completed at this point (No in step SD06), the second search unit 140 adjusts the value of the control parameter and sets it for the industrial machine 4, which is the actual controlled object (step SD07). Then, the process proceeds to step SD03. On the other hand, if the search for the second range has ended at this point (Yes in step SD06), the control parameter that satisfies the constraints (the machining error as an index value is equal to or less than the target error) and has the highest evaluation (the cycle time as an index value is the fastest) is selected from the searched control parameters (step SD08). The processes in steps SD03 to SD07 are performed using known search algorithms such as machine learning search, rule-based search, a predetermined exhaustive search algorithm, and Bayesian optimization, as described above.

[0040] [Correction based on Rule 91 07.01.2025] Here, as a result of the second search process, a set of control parameter values ​​that satisfy the constraint conditions (linear acceleration = 2250 [mm / sec] 2 ], corner speed difference=1000 [mm / sec], machining error is 10 [μm], and cycle time is 35.9 [sec]) is selected by the second search unit 140. The output unit 150 displays the set of control parameter values ​​selected by the second search unit 140 on the display device 70, and also transmits and sets the set of control parameter values ​​to the industrial machine 4 that is the actual target of control (step SD09).

[0041] The parameter adjustment device 1 according to this embodiment, which has the above configuration, narrows the range of control parameters to near the boundaries of constraint conditions through simulation processing that can be processed at high speed. Because the narrowed range of control parameters is sufficiently narrow, it becomes possible to adjust the control parameters using, for example, an industrial machine 4, which is an actual controlled object. Although control parameters can be adjusted accurately in the industrial machine 4, it is difficult to adjust the control parameters over a wide range because it takes time to obtain operation results. However, because the adjustment range can be narrowed down in advance to a range that is somewhat likely through simulation processing that can be processed at high speed, it is expected that the adjustment time using the industrial machine 4 can be shortened.

[0042] Although the embodiments of the present disclosure have been described in detail above, the present disclosure is not limited to the individual embodiments described above. Various additions, substitutions, modifications, partial deletions, etc. are possible in these embodiments without departing from the gist of the invention or the idea and intent of the present disclosure derived from the content described in the claims and their equivalents. For example, in the above-described embodiments, the order of each operation and the order of each process are shown as examples and are not limited to these. The same applies when numerical values ​​or mathematical expressions are used in the description of the above-described embodiments.

[0043] The following are supplementary notes related to embodiments of the present disclosure. (Supplementary Note 1) A parameter adjusting device (1) according to one aspect of the present disclosure includes: a controlled object model that models the operation of each part included in an industrial machine (4) that is a controlled object; a simulation unit (100) that simulates the operation of the controlled object, to which predetermined control parameter values ​​are set, based on the controlled object model; a first search unit (110) that repeatedly causes the simulation unit (100) to perform a simulation process while changing the value of the control parameter within a predetermined first range, and searches for candidate sets of control parameter values ​​that will yield the first index value near a boundary of a predetermined constraint, based on a first index value calculated from the result of the simulation process; a candidate implementation unit (120) that sets the candidate sets of control parameter values ​​searched for by the first search unit (110) as the controlled object, performs operation verification, and calculates a second index value based on the result of the operation verification; and a search range determination unit (130) that determines a second range of values ​​that is narrower than the first range, based on the first index value and the second index value. (Supplementary Note 2) A parameter adjusting device (1) according to another aspect of the present disclosure further includes a second search unit (140) that searches for the control parameter by repeatedly operating the controlled object while changing the value of the control parameter within the second range. (Supplementary Note 3) In the parameter adjusting device (1) according to another aspect of the present disclosure, the first index value and the second index value are any of machining accuracy, machined surface quality, and machining time, and at least one of the first search unit (110) and the second search unit (140) sets the constraint condition related to the machining accuracy or the machined surface quality and searches for the control parameter by a machine learning technique so as to minimize the machining time under the constraint condition.(Supplementary Note 4) A computer-readable recording medium having a program recorded thereon according to one aspect of the present disclosure has recorded thereon a program that causes a computer to operate as: a simulation unit (100) that simulates the operation of a control object, to which predetermined control parameter values ​​are set, based on a control object model that models the operation of each part of at least an industrial machine (4) that is a control object; a first search unit (110) that repeatedly causes the simulation unit (100) to perform a simulation process while changing the value of the control parameter within a predetermined first range, and that searches for candidate sets of control parameter values ​​that will yield the first index value near a boundary of a predetermined constraint, based on a first index value calculated from the result of the simulation process; a candidate implementation unit (120) that sets the candidate sets of control parameter values ​​searched by the first search unit (110) as the control object, verifies the operation, and calculates a second index value based on the result of the operation verification; and a search range determination unit (130) that determines a second range of values ​​that is narrower than the first range, based on the first index value and the second index value.

[0044] REFERENCE SIGNS LIST 1 Parameter adjustment device 4 Industrial machine 5 Network 6 Fog computer 7 Cloud server 8 Sensor 11 CPU 12 ROM 13 RAM 14 Non-volatile memory 15, 17, 18, 20 Interface 22 Bus 70 Display device 71 Input device 72 External device 100 Simulation unit 110 First search unit 120 Candidate implementation unit 130 Search range determination unit 140 Second search unit 150 Output unit 210 Model storage unit 220 Evaluation program

Claims

1. A parameter adjusting device comprising: a controlled object model which models the operation of each part of an industrial machine that is to be controlled; a simulation unit which simulates the operation of the controlled object, for which predetermined control parameter values ​​are set, based on the controlled object model; a first search unit which repeatedly causes the simulation unit to perform a simulation process while changing the value of the control parameter within a predetermined first range, and searches for candidate sets of control parameter values ​​which will result in the first index value near a boundary of a predetermined constraint condition, based on a first index value calculated from the result of the simulation process; a candidate implementation unit which sets the candidate sets of control parameter values ​​searched for by the first search unit as the controlled object, performs operation verification, and calculates a second index value based on the result of the operation verification; and a search range determination unit which determines a second range of values ​​which is narrower than the first range, based on the first index value and the second index value.

2. The parameter adjustment device according to claim 1, further comprising a second search unit that searches for the control parameter by repeatedly operating the controlled object while changing the value of the control parameter within the second range.

3. The parameter adjusting device according to claim 2, wherein the first index value and the second index value are either machining accuracy, machined surface quality, or machining time, and at least one of the first search unit and the second search unit sets the constraint condition related to the machining accuracy or the machined surface quality, and searches for the control parameter using a machine learning technique so as to maximize the machining time under the constraint condition.

4. A computer-readable recording medium having recorded thereon a program for causing a computer to operate as: a simulation unit that simulates the operation of a controlled object, in which predetermined control parameter values ​​are set, based on a controlled object model that models the operation of at least each part of the industrial machine that is the controlled object; a first search unit that repeatedly causes the simulation unit to perform a simulation process while changing the value of the control parameter within a predetermined first range, and searches for candidate sets of control parameter values ​​that will result in the first index value near the boundary of a predetermined constraint condition, based on a first index value calculated from the result of the simulation process; a candidate implementation unit that sets the candidate sets of control parameter values ​​searched for by the first search unit as the controlled object, verifies the operation, and calculates a second index value based on the result of the operation verification; and a search range determination unit that determines a second range, which is a range of values ​​narrower than the first range, based on the first index value and the second index value.

Images