Processing condition determination support device, processing condition determination support system, processing condition determination support method and processing condition determination support program

Abstract

A machining condition determination support device (1) that supports the determination of a machining condition comprises: a simulation execution unit (11) that performs a machining simulation of a workpiece for each of several machining condition schemes based on a material shape model (21) representing a shape of the workpiece before machining, a tool model (22) representing a shape of a tool, and tool motion command data (23) during the machining operation; and an adaptation direction determination unit (13) that receives a selection from an operator for a displayed simulation result and, based on the received selection, determines and presents an adaptation direction for a machining condition of a machining simulation to be tried next.

Description

Technical area

[0001] The present disclosure relates to a machining condition determination supporting apparatus, a machining condition determination supporting system, a machining condition determination supporting method, and a machining condition determination supporting program for supporting the determination of a machining condition used when a machine tool performs machining. Technological background

[0002] The quality of a machining result with a machine tool depends largely on the settings of the machining conditions. Traditionally, to determine a machining condition that achieves a desired machining result (hereinafter referred to as a suitable machining condition), trial machining is performed several times on site, and a final machining condition is determined from the trial machining results using the intuition and experience of a professional. In this method, since trial machining requires time, the number of machining condition patterns that can be tried is limited to a small number.

[0003] In recent years, machining simulation technology has been developed, and machining simulation has been performed on computers without performing actual machining. As a result, many machining condition schemes can be evaluated in a short time, and suitable machining conditions can be easily determined.

[0004] For example, a display device described in Patent Literature 1 includes an input means such as a slide bar configured to continuously change at least one of the machining conditions, and a display means that displays machining simulation results associated with the changed machining conditions. Subsequently, the display device prompts the worker to recognize the machining simulation results associated with the continuous changes in the machining conditions and optimize the machining results. Citation listPatent literature

[0005] Patent Literature 1: Japanese Patent Application Laid-Open No. 2004 - 280 267 Short descriptionTechnical problem

[0006] However, in the technology of Patent Literature 1, since the machining conditions include a plurality of parameters, finding a suitable machining condition requires considering a vast number of machining condition schemes based on combinations of multiple parameters. In the technology of Patent Literature 1 described above, it is only possible to interactively search for a suitable machining condition through systematic trial and error in what is called a round-robin manner. The number of combinations increases explosively with the number of parameters, resulting in a problem that a suitable machining condition cannot be efficiently obtained in a short time.

[0007] The present disclosure has been made in view of the foregoing, and an object thereof is to obtain a machining condition determination support apparatus capable of efficiently obtaining a suitable machining condition in a short time. Solution to the problem

[0008] In order to solve the problems described above and achieve the object, a machining condition determination supporting device according to the present disclosure supports the determination of a machining condition in machining a workpiece using a machine tool, and includes: a simulation executing unit that executes machining simulation of the workpiece for each of a plurality of machining condition schemes based on a material shape model representing a shape of the workpiece before machining, a tool model representing the shape of a tool used for machining, and tool movement command data indicating the movement of the tool during the machining operation.The machining condition determination support device according to the present disclosure further includes a simulation display unit that displays a simulation result, which is a result of the machining simulation, on a screen display. The machining condition determination support device according to the present disclosure further includes a selection receiving unit that receives a selection from an operator for the displayed simulation result, and an adjustment direction determining unit that determines and presents an adjustment direction of a machining condition of a machining simulation to be tried next based on the selection received by the selection receiving unit. Advantageous effects of the invention

[0009] The machining condition determination support device according to the present disclosure can achieve the effect of efficiently obtaining a suitable machining condition in a short time. Short description of drawings Fig. 1 is a block diagram illustrating the configuration of a machining condition determination supporting apparatus according to the first embodiment. Fig. 2 is a block diagram illustrating the configuration of the simulation execution unit included in the machining condition determination support apparatus according to the first embodiment. Fig. 3 is a diagram for explaining the machining simulation results displayed by the machining condition determination support apparatus according to the first embodiment. Fig. 4 is a diagram for explaining a machining condition adjustment direction calculated by the machining condition determination supporting apparatus according to the first embodiment. Fig. 5 is a diagram for explaining a range in which a machining condition can be selected and which is suitable for the Fig. 4 shown reference axis is set. Fig. 6 is a flowchart illustrating a procedure for machining condition determination support processing by the machining condition determination support apparatus according to the first embodiment. Fig. 7 is a block diagram illustrating the configuration of a machining condition determination supporting apparatus according to the second embodiment. Fig. 8 is a diagram for explaining a neural network used by the learning unit according to the second embodiment. Fig. 9 is a flowchart illustrating a procedure for machining condition adjustment direction presentation processing by the machining condition determination support apparatus according to the second embodiment. Fig. 10 is a flowchart illustrating a procedure of a learning process executed by the learning unit according to the second embodiment. Fig. 11 is a flowchart illustrating a procedure of an inference process executed by the inference unit according to the second embodiment. Fig. 12 is a diagram illustrating an exemplary hardware configuration implementing the machining condition determination support apparatus according to the first and second embodiments. Description of embodiments

[0010] Hereinafter, a machining condition determination supporting apparatus, a machining condition determination supporting system, a machining condition determination supporting method, and a machining condition determination supporting program according to embodiments of the present disclosure will be described in detail with reference to the drawings. First embodiment.

[0011] Fig. 1 is a block diagram illustrating the structure of a machining condition determination support device according to the first embodiment. The machining condition determination support device 1 is a computer that supports a process of determining a machining condition used when a machine tool machines a workpiece. A user (operator) of the machining condition determination support device 1 executes the process of determining a machining condition using the machining condition determination support device 1.

[0012] The machining condition determination support device 1 includes a simulation execution unit 11, a simulation display unit 12, an adjustment direction determination unit 13, a machining condition update unit 14, a selection reception unit 15, and a storage unit 20. The machining condition determination support device 1 is connected to a display screen 50, a computer mouse 51, and a keyboard 52. The machining condition determination support device 1 receives a user instruction corresponding to the operator's operation on the computer mouse 51 and keyboard 52. The machining condition determination support device 1 displays calculated information (such as simulation results, which will be described below) on the display screen 50. The storage unit 20 stores a material shape model 21, a tool model 22, tool movement command data 23, and a

[0013] Machining Condition Set 24. The material shape model 21 is a model that expresses the shape of the workpiece before machining, i.e., the shape of the material. This means that the material shape model 21 is a model of the original shape of the workpiece. The tool model 22 is a model that represents the shape of the tool used in machining.

[0014] Tool movement command data 23 is data that specifies a movement command for the tool during the machining operation of the machine tool. Tool movement command data 23 is extracted, for example, from a control program for controlling the machine tool. An example of tool movement command data 23 is G-code.

[0015] The machining condition set 24 is data that specifies a machining condition for the machine tool to perform a machining operation. The machining condition set 24 includes multiple machining conditions. Each machining condition includes multiple parameters. For example, each machining condition of the machining condition set 24 includes different parameters for controlling numerical control behavior, such as the tool feed speed, referred to as the F value, and the spindle rotation speed, referred to as the S value.

[0016] The simulation execution unit 11 reads the material shape model 21, the tool model 22, the tool movement command data 23, and the machining condition set 24 from the storage unit 20. The simulation execution unit 11 simulates a machining operation performed by the machine tool. One example of machining performed by the machine tool is cutting.

[0017] The simulation execution unit 11 sequentially moves the tool specified by the tool model 22 in accordance with the tool movement command specified by the tool movement command data 23 with respect to the original shape of the workpiece specified by the material shape model 21, thereby simulating machining by the machine tool.

[0018] The simulation execution unit 11 executes a machining simulation for the workpiece for each of a plurality of machining condition schemes. The simulation execution unit 11 simulates cutting by removing an area where the material shape model 21, which is the shape model of the workpiece, and the moving tool model 22 intersect. At this time, the simulation execution unit 11 calculates a simulation result for each machining condition by applying a specific machining condition in the machining condition set 24. The simulation result calculated by the simulation execution unit 11 is the shape of the workpiece machined by the tool. The simulation execution unit 11 transmits the simulation result to the simulation display unit 12. The simulation display unit 12 shows the

[0019] Machining simulation results for the respective machining conditions executed by the simulation execution unit 11 are displayed side by side on the screen display 50. The simulation display unit 12 sends the machining simulation results displayed on the screen display 50 to the adjustment direction determination unit 13.

[0020] The selection receiving unit 15 receives a selection from the operator regarding the simulation result to be displayed on the screen display 50. The selection receiving unit 15 receives, via the computer mouse 51 or the keyboard 52, the machining simulation result selected by the operator from the machining simulation results displayed on the screen display 50. The machining simulation result selected by the operator is a machining simulation result corresponding to the machining result (shape of the workpiece after machining, machining time, and the like) determined by the operator to be unsuitable.

[0021] The operator selects the machining simulation result determined as inappropriate from the machining simulation results displayed on the screen display 50 using the computer mouse 51 or keyboard 52. Subsequently, the selection receiving unit 15 identifies the machining simulation result determined as inappropriate from the machining simulation results displayed on the screen display 50.

[0022] The selection receiving unit 15 receives the machining simulation result selected by the operator as unsuitable. The machining simulation result selected by the operator as unsuitable can subsequently be referred to as an unsuitable simulation result. Additionally, the machining condition corresponding to the unsuitable simulation result can be referred to as an unsuitable machining condition.

[0023] The adjustment direction determining unit 13 determines and presents the adjustment direction of the machining condition for the machining simulation to be executed next based on the selection received from the selection receiving unit 15. Specifically, the adjustment direction determining unit 13 determines an adjustment direction (hereinafter referred to as a machining condition adjustment direction) to switch to a machining condition that is expected to be more desirable than the current state based on the dependency between the machining condition corresponding to the unsuitable simulation result and the machining condition corresponding to the remaining machining simulation result not selected by the operator.Hereinafter, the remaining machining simulation results not selected by the operator are referred to as remaining simulation results. The machining condition corresponding to a remaining simulation result can be referred to as a remaining machining condition. The adjustment direction determination unit 13 presents the machining condition adjustment direction to the operator by displaying the machining condition adjustment direction on the screen display 50. The adjustment direction determination unit 13 also sends the machining condition adjustment direction to the machining condition update unit 14. Details of the machining condition adjustment direction will be described later.

[0024] The machining condition update unit 14 receives, from the computer mouse 51 or the keyboard 52, the machining condition specified by the operator along the machining condition adjustment direction displayed on the screen display 50. In this case, the operator refers to the machining condition adjustment direction displayed on the screen display 50, determines a machining condition predicted to be capable of producing a suitable machining result, and inputs the machining condition to the machining condition determination support device 1 using the computer mouse 51 or the keyboard 52. Subsequently, the machining condition update unit 14 acquires the machining condition specified by the operator.

[0025] After receiving the machining condition determined by the operator with reference to the machining condition adjustment direction, the machining condition updating unit 14 updates the machining condition corresponding to the machining simulation result determined as unsuitable in the machining condition set 24 into a new machining condition determined by the operator.

[0026] The machining condition determination support device 1 repeats the machining simulation, the process of obtaining an operation by the operator, and the process of updating the machining condition until a machining simulation result corresponding to the advantageous machining desired by the operator is obtained.

[0027] As described above, the operation by the operator includes an operation of selecting an inappropriate machining simulation result by the operator and an operation of determining a new machining condition along the presented machining condition adjustment direction.

[0028] Next, the configuration of the simulation execution unit 11 is described. Fig. 2 is a block diagram illustrating the configuration of the simulation execution unit included in the machining condition determination support apparatus according to the first embodiment.

[0029] The simulation execution unit 11 includes a motion command analysis unit 41, a numerical control unit 42, a mechanical response calculation unit 43, and a cutting simulation unit 44. The motion command analysis unit 41 reads the tool motion command data 23 from the storage unit 20. The motion command analysis unit 41 analyzes the tool motion command data 23, which is described in a standard format such as G-code, and calculates a motion command 61 for the tool. The motion command analysis unit 41 sends the calculated motion command 61 to the numerical control unit 42.

[0030] The numerical control unit 42 reads a machining condition (here, machining condition 24a) from the storage unit 20 in the machining condition set 24. The numerical control unit 42 sequentially reads machining conditions one by one from the machining condition set 24. The numerical control unit 42 calculates a servo command 62 for operating the machine tool using the read machining condition 24a and the motion command 61. The servo command 62 for operating the machine tool is a command to a servo motor that drives the tool. The numerical control unit 42 sends the calculated servo command 62 to the mechanical response calculation unit 43.

[0031] The numerical control unit 42 receives servo feedback (FB) data 63 from the mechanical response calculation unit 43 as feedback data for the servo command 62. The numerical control unit 42 executes feedback control by calculating the servo command 62 based on the servo FB data 63.

[0032] The mechanical response calculation unit 43 simulates a dynamic response of the machine tool to the calculated servo command 62 and calculates a tool motion trajectory 64, which is a tool motion path, as a result of the simulation. The mechanical response calculation unit 43 sends the calculated tool motion trajectory 64 to the cutting simulation unit 44.

[0033] The cutting simulation unit 44 reads the tool model 22 and the material shape model 21 from the storage unit 20. The cutting simulation unit 44 simulates cutting of the workpiece based on the tool movement trajectory 64, the tool model 22, and the material shape model 21. That is, the cutting simulation unit 44 simulates machining with the material shape model 21 as the original shape. The result of this simulation is a post-machining shape 45 of the workpiece. After calculating the new post-machining shape 45 by simulating machining, the cutting simulation unit 44 updates the old post-machining shape with the calculated post-machining shape 45. The cutting simulation unit 44 sends the post-machining shape 45, which is a result of the simulation, to the simulation display unit 12.

[0034] During machining with the machine tool, depending on machining conditions, the actual tool movement trajectory 64 may be internally rotated or externally rotated, and may follow a path different from the original tool movement trajectory expected from the tool movement command data 23. Therefore, different simulation results are generated depending on machining conditions.

[0035] It is noted that the data of the machining simulation result calculated by the simulation executing unit 11 may be any data as long as the operator can visually recognize the quality of the machining result caused by the difference in machining conditions, and is not limited to the example of the first embodiment.

[0036] Fig. 3 is a diagram for explaining machining simulation results displayed by the machining condition determination support device according to the first embodiment. Fig. 3 illustrates a state (screen display 71) in which machining simulation results corresponding to a plurality of machining conditions are displayed side by side, and a state (screen display 72) in which a machining simulation result determined to be unsuitable from the machining point of view has been selected.

[0037] Machining simulation results equivalent to the number of machining condition schemes in the machining condition set 24 are displayed on the screen display 71, which is an example of the screen display 50. Fig. 3 illustrates a case where four machining simulation results SA to SD are displayed on the screen display 71 as machining simulation results corresponding to four machining conditions.

[0038] The machining simulation result determined as inappropriate by the operator from among the machining simulation results SA to SD displayed on the screen display 71 is selected. For example, the operator selects the machining simulation result determined as inappropriate using the computer mouse 51.

[0039] On the screen display 72, which is an example of the screen display 50, the machining simulation result selected by the operator is highlighted. Fig. 3 illustrates a case where the machining simulation result SB selected by the operator is highlighted on the screen display 72. The machining simulation result SB thus selected is transmitted to the adjustment direction determination unit 13 via the selection receiving unit 15. Then, the machining condition corresponding to the selected machining simulation result SB is updated.

[0040] In this way, the simulation execution unit 11 displays the machining simulation results side by side. As a result, the operator can glance at the simulation results, and even an inexperienced operator can easily select a suitable or unsuitable machining simulation result.

[0041] Note that an END button is displayed on the screen display 71. When the operator presses the END button 75 on the screen display 71, the process of determining a machining condition ends. The machining condition set 24 stored in the storage unit 20 at the time the END button 75 is pressed is the determined machining condition. The machining condition set 24 stored in the storage unit 20 at the time the END button 75 is pressed is the updated last combination of machining conditions.

[0042] Fig. 4 is a diagram for explaining a machining condition adjustment direction calculated by the machining condition determination support device according to the first embodiment. Fig. 4, an example of the machining condition adjustment direction calculated by the adjustment direction determination unit 13 will be described.

[0043] For example, the simulation execution unit 11 executes the machining simulation on the machining conditions (here, machining conditions RA to RD (not shown)) including the parameters P1 to P3. The machining conditions RA to RD used by the simulation execution unit 11 to execute the machining simulation are plotted as points in a vector space based on the parameters P1 to P3 that constitute the machining conditions.

[0044] In Fig. 4, points of machining conditions RA to RD are represented as points QA to QD. This means that machining condition RA corresponds to point QA, and machining condition RB corresponds to point QB. Machining condition RC corresponds to point QC, and machining condition RD corresponds to point QD.

[0045] The simulation display unit 12 displays the unsuitable machining condition and the remaining machining conditions corresponding to the machining simulation results side by side in the vector space based on the machining parameters.

[0046] For example, if the operator selects the machining simulation result corresponding to machining condition RB as the unsuitable machining simulation result, the machining simulation results corresponding to machining conditions RA, RC, and RD are the remaining machining simulation results. In this case, the machining condition RB corresponding to the unsuitable machining simulation result is the unsuitable machining condition, and the machining conditions RA, RC, and RD corresponding to the remaining machining simulation results are the remaining machining conditions.

[0047] The machining condition determination support device 1 may use one or two of the machining conditions RA, RC, and RD as the remaining machining simulation results instead of all of the machining conditions RA, RC, and RD.

[0048] When the unsuitable machining condition is the machining condition RB, the adjustment direction determining unit 13 calculates vectors from the point QB to the points QA, QC, and QD corresponding to the remaining machining conditions (remaining machining conditions). That is, the adjustment direction determining unit 13 calculates a vector from the point QB to the point QA, a vector from the point QB to the point QC, and a vector from the point QB to the point QD. Further, the adjustment direction determining unit 13 calculates a combined vector of the vectors from the point QB to the remaining points QA, QC, and QD as a reference axis 80 of the adjustment direction and presents the calculated vector to the operator. In the first embodiment, the direction of the combined vector, the adjustment direction, and the direction of the reference axis 80 are the same.

[0049] The operator specifies a new machining condition along the presented reference axis 80 as a new machining condition for the next machining simulation. For example, the operator specifies a point on the reference axis 80 as a new machining condition. Note that the machining condition at this time is a preliminary determination, and the machining condition at the time the END button 75 is pressed is the final machining condition.

[0050] The new machining condition that can be set by the operator is not necessarily limited to the machining condition on the reference axis 80. The adjustment direction determining unit 13 can set a specific range around the reference axis 80 as a range in which a new machining condition can be selected.

[0051] Fig. 5 is a diagram for explaining a range in which a machining condition can be selected and which is suitable for the Fig. 4 is set. For example, the adjustment direction determining unit 13 may set as a vertex the point QB (point corresponding to the machining condition RB selected by the operator as unsuitable), which is the starting point of the reference axis 80, and a tapered portion 81, that is, a portion having a conical shape and extending from the point QB in the direction of the reference axis 80, as a portion (machining condition selectable portion) in which a new machining condition can be selected. In this case, the length of the reference axis 80 corresponds to the height of the tapered portion 81.

[0052] Additionally, the adjustment direction determination unit 13 may take a point sequence on the reference axis 80 or in the conical portion 81 at a specific interval, and set points included in this point sequence as points from which a new machining condition can be selected. Thus, once a simulation result corresponding to an unsuitable machining condition is selected from the simulation results for the plurality of machining condition sets 24, the machining condition determination support device 1 executes the setting of a new machining condition along the machining condition adjustment direction calculated by the above method.

[0053] The machining condition determination support device 1 repeatedly executes a machining simulation for a machining condition, calculates a machining condition adjustment direction based on an unsuitable machining condition selected by an operator, and updates it to a new machining condition set by the operator along the machining condition adjustment direction. As a result, the machining condition determination support device 1 can adjust the machining conditions toward a desired machining result.

[0054] Fig. 6 is a flowchart illustrating a procedure for machining condition determination support processing by the machining condition determination support device according to the first embodiment. The simulation execution unit 11 executes machining simulation with the machine tool using the material shape model 21, the tool model 22, the tool movement command data 23, and the machining condition set 24 (step S10).

[0055] The simulation display unit 12 displays the simulation result of the machining simulation for each machining condition performed by the simulation execution unit 11 on the screen display 50 (step S20).

[0056] The machining condition determination support device 1 determines whether the END button 75 has been pressed by the operator (step S25). In response to a determination that the END button 75 has not been pressed by the operator (step S25, NO), the operator selects the most unsuitable machining simulation result from the simulation results displayed on the screen display 50. The adjustment direction determination unit 13 receives the unsuitable machining simulation result selected by the operator via the selection receiving unit 15 (step S30).

[0057] The adjustment direction determining unit 13 calculates and presents the machining condition adjustment direction based on the dependency between the unsuitable machining condition corresponding to the unsuitable machining simulation result and the remaining machining condition corresponding to the remaining machining simulation result (step S40).

[0058] Subsequently, the operator can change the machining condition along the machining condition adjustment direction displayed on the screen display 50. The machining condition update unit 14 receives the machining condition determined by the operator along the machining condition adjustment direction displayed on the screen display 50 (step S50).

[0059] Upon receiving the machining condition designated by the operator, the machining condition updating unit 14 updates the machining condition corresponding to the machining simulation result determined as inappropriate in the machining condition set 24 as the new machining condition designated by the operator (step S60).

[0060] Thereafter, the machining condition determination support device 1 returns to the processing of step S10 and executes the processing of steps S10 and S20. Then, the machining condition determination support device 1 determines whether the END button 75 has been pressed by the operator (step S25).

[0061] The machining condition determination support device 1 repeats the processing of steps S30 to S60 and the processing of steps S10, S20, and S25 until the END button 75 is pressed by the operator. That is, the processing of steps S30 to S60 and the processing of steps S10, S20, and S25 are repeated until a desired simulation result is obtained.

[0062] In response to a determination that the END button 75 has been pressed by the operator (step S25, YES), the machining condition determination support device 1 ends the machining condition determination support processing. At this time, the final machining condition is the machining condition determined by the machining condition determination support device 1.

[0063] Note that the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the machining condition update unit 14, the selection reception unit 15, and the storage unit 20 may be arranged in different locations. In this case, at least some of the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the machining condition update unit 14, the selection reception unit 15, and the storage unit 20 are connected to a network and transmit and receive data via the network.That is, the machining condition determination support processing described in the first embodiment can be implemented by a machining condition determination support system in which at least one of the simulation executing unit 11, the simulation displaying unit 12, the adjustment direction determining unit 13, the machining condition updating unit 14, the selection receiving unit 15, and the storage unit 20 are connected to a network.

[0064] The machining condition determination support device 1, which calculates the adjustment direction of the machining condition with respect to the unsuitable machining condition, can obtain a suitable machining condition with a small number of trials compared to a case where the machining simulation is performed under round-robin machining conditions. Therefore, the machining condition determination support device 1 can efficiently determine a suitable machining condition in a short time.

[0065] As described above, the machining condition determination support device 1 according to the first embodiment receives the unsuitable simulation result selected by the operator from a machining quality perspective with respect to the displayed simulation results. Then, the machining condition determination support device 1 determines and displays the adjustment direction of the machining condition of the machining simulation to be tried next based on the relationship between the unsuitable machining condition corresponding to the unsuitable simulation result and the remaining machining condition corresponding to the remaining simulation result.In addition, the machining condition determination support device 1 receives the next machining condition set by the operator along the displayed adjustment direction and replaces the unsuitable machining condition with the received next machining condition. As a result, the machining condition determination support device 1 can adjust the machining conditions toward a desired machining result. Therefore, the machining condition determination support device 1 can efficiently obtain a suitable machining condition in a short time. Second embodiment.

[0066] Next, the second embodiment will be described with reference to Fig. 7 to 11. In the second embodiment, the machining condition determination support device learns the relationship between an unsuitable machining condition, a remaining machining condition, and a machining condition adjustment direction corresponding to a machining condition set by the operator, and generates a learned model as a learning result. Then, the machining condition determination support device applies an unsuitable machining condition and a remaining machining condition to the learned model to infer the machining condition adjustment direction corresponding to the machining condition expected to be set by the operator.

[0067] Fig. 7 is a block diagram illustrating the configuration of a machining condition determination support device according to the second embodiment. Fig. 7, which have the same functionalities as those of the machining condition determination support device 1 according to the Fig. 1 are denoted by the same reference numerals, and their duplicate descriptions are omitted. The following mainly describes differences from the first embodiment.

[0068] The machining condition determination support device 2 learns the machining condition adjustment direction and uses the learning result to infer the machining condition adjustment direction and present the machining condition adjustment direction to the operator. The machining condition determination support device 2 includes a learning unit 31, an inference unit 32, and a learning data storage unit 30 in addition to the components included in the machining condition determination support device 1. The learning unit 31 is connected to the adjustment direction determination unit 13 and the machining condition update unit 14, but the connecting lines are not shown.

[0069] The learning unit 31 generates a learned model 33, which is learned data obtained by learning the relationship between an unsuitable machining condition and a remaining machining condition, and a machining condition adjustment direction corresponding to the machining condition set by the operator. The machining condition adjustment direction corresponding to the machining condition set by the operator corresponds to the direction of the vector from the point indicating the unsuitable machining condition to the point indicating the machining condition set by the operator in the vector space based on the machining parameters.

[0070] The learned model 33 is a model for inferring the machining condition adjustment direction from the unsuitable machining condition and the remaining machining condition. The learning data storage unit 30 stores the learned model 33 generated by the learning unit 31.

[0071] The inference unit 32 applies the unsuitable machining condition and the remaining machining condition to the learned model 33 to infer the machining condition adjustment direction corresponding to the unsuitable machining condition and the remaining machining condition. In addition, the inference unit 32 calculates a range in which a machining condition can be selected from the inferred machining condition adjustment direction. Details of the learning unit 31 and the inference unit 32 will be described later.

[0072] The adjustment direction determination unit 13 calculates the machining condition adjustment direction based on the machining condition corresponding to the unsuitable simulation result selected by the operator and the machining condition corresponding to the remaining simulation result. However, the machining condition adjustment direction calculated by the adjustment direction determination unit 13 may not be the shortest route leading to the final machining condition.

[0073] Therefore, the machining condition determination support device 2 according to the second embodiment learns and infers the machining condition adjustment direction using the learning unit 31 and the inference unit 32, which are learning inference mechanisms, to determine the machining condition adjustment direction. The inferred machining condition adjustment direction is a machining condition adjustment direction obtained by correcting the machining condition adjustment direction determined by the adjustment direction determination unit 13. That is, the machining condition adjustment direction inferred using the learning unit 31 and the inference unit 32 is the machining condition adjustment direction corresponding to the machining condition predicted to be determined by the operator.By presenting the inferred machining condition adjustment direction to the operator, the machining condition determination support device 2 can obtain a suitable machining condition in fewer procedures than in the first embodiment.

[0074] The learning unit 31 may use a known algorithm such as supervised learning, unsupervised learning, or reinforcement learning as the learning algorithm. As an example, a case will be described in which a neural network is applied to the learning algorithm used in the learning unit 31.

[0075] For example, the learning unit 31 learns the machining condition adjustment direction corresponding to the unsuitable machining condition and the remaining machining condition through what is called supervised learning according to a neural network model. Here, supervised learning refers to a method that provides pairs of inputs and outputs (labels) (learning data) to the learning unit 31 to learn features contained in these learning data and infer results from the inputs.

[0076] A neural network comprises an input layer composed of several neurons, an intermediate layer (hidden layer) composed of several neurons, and an output layer composed of several neurons. The number of intermediate layers can be one, two, or more.

[0077] Fig. Fig. 8 is a diagram for explaining a neural network used by the learning unit according to the second embodiment. For example, in the case of the Fig. In the three-layer neural network shown in Figure 8, a plurality of inputs are provided to the input layers (X1 to X3), and their values ​​are multiplied by weights W1 (w11 to w16) for input to the intermediate layers (Y1 to Y2). Then, the results are further multiplied by weights W2 (w21 to w26) for output from the output layers (Z1 to Z3). The output results vary depending on the values ​​of weights W1 and W2.

[0078] The neural network used by the learning unit 31 learns the machining condition adjustment direction corresponding to the inappropriate machining condition and the remaining machining condition by what is called supervised learning, according to the learning data generated based on a combination of the inappropriate machining condition and the remaining machining condition and the machining condition adjustment direction corresponding to the machining condition set by the operator.

[0079] That is, the neural network learns by adjusting the weights W1 and the weights W2 so that the result output from the output layer by inputting the inappropriate processing condition and the remaining processing condition, which are the first input, approaches the processing condition adjustment direction, which is the second input (correct answer).

[0080] The neural network learns the correspondence dependency between the unsuitable machining condition and the remaining machining condition and the machining condition adjustment direction, thereby generating the learned model 33 that can output the appropriate machining condition adjustment direction when the unsuitable machining condition and the remaining machining condition are input. In this way, the learning unit 31 learns the learned model 33 that can output the correct machining condition adjustment direction in response to the input of the unsuitable machining condition and the remaining machining condition.

[0081] The learning unit 31 generates the learned model 33 by performing the above learning and outputs it. The learning data storage unit 30 stores the learned model 33 output by the learning unit 31.

[0082] During the learning phase, the learning unit 31 receives machining conditions corresponding to the unsuitable simulation result selected by the operator and the remaining simulation results from the adjustment direction determination unit 13. In addition, the learning unit 31 receives the machining condition adjustment direction corresponding to the machining condition set by the operator along the machining condition adjustment direction from the machining condition updating unit 14.

[0083] In the second embodiment, the machining condition adjustment direction is presented to the operator by a combination of the reference axis 80 of the machining condition adjustment direction and the conical portion 81 set around the reference axis 80, as shown in Fig. 5 is shown.

[0084] The learning unit 31 generates the learned model 33, which estimates the dependency between the unsuitable machining condition, the remaining machining condition, and the machining condition adjustment direction corresponding to the machining condition set by the operator, from each trial (machining simulation) conducted so far. In this way, the learning unit 31 improves the learned model 33 so that the deviation between the reference axis of the machining condition adjustment direction presented to the operator and the reference axis of the machining condition adjustment direction (the next machining condition to be tried) expected to be selected by the operator is minimized.

[0085] In the inference phase, the inference unit 32 receives an unsuitable machining condition and a remaining machining condition from the adjustment direction determination unit 13. The inference unit 32 applies the unsuitable machining condition and the remaining machining condition to the learned model 33 to infer the reference axis of the adjustment direction of the next machining condition to be tried. Then, the inference unit 32 calculates a range within which a machining condition is selectable from the inferred reference axis of the machining condition adjustment direction and presents the range to the operator via the adjustment direction determination unit 13 and the screen display 50. As a result, the machining condition determination support device 2 can obtain a suitable machining condition with a smaller number of trials than the machining condition determination support device 1.

[0086] Fig. 9 is a flowchart illustrating a procedure for machining condition adjustment direction presentation processing by the machining condition determination support device according to the second embodiment. The learning unit 31 of the machining condition determination support device 2 receives the unsuitable machining condition corresponding to the unsuitable simulation result selected by the operator from the adjustment direction determination unit 13 (step S110). In addition, the learning unit 31 receives the remaining machining condition corresponding to the remaining simulation result from the adjustment direction determination unit 13 (step S120).

[0087] In addition, the learning unit 31 receives from the machining condition updating unit 14 the machining condition adjustment direction corresponding to the machining condition set by the operator along the machining condition adjustment direction (step S130). Note that the remaining simulation result is determined by selecting the unsuitable simulation result and then the operator sets the machining condition, but the processing in steps S110 to S130 can be performed in any order. That is, the learning unit 31 can receive the unsuitable machining condition, the remaining machining condition, and the machining condition adjustment direction corresponding to the machining condition set by the operator in any order.

[0088] The learning unit 31 generates the learned model 33 for inferring the machining condition adjustment direction from the unsuitable machining condition and the remaining machining condition (step S140). The learned model 33 is stored in the learning data storage unit 30.

[0089] Subsequently, when presenting the machining condition adjustment direction, the machining condition determination support device 2 infers the machining condition adjustment direction using the learned model 33. Specifically, the inference unit 32 reads the learned model 33 from the learning data storage unit 30 (step S150).

[0090] In addition, the inference unit 32 receives the unsuitable machining condition corresponding to the unsuitable simulation result from the adjustment direction determination unit 13 (step S160). The inference unit 32 receives the remaining machining condition corresponding to the remaining simulation result from the adjustment direction determination unit 13 (step S170). Note that the processing in steps S150 to S170 can be performed in any order.

[0091] The inference unit 32 applies the unsuitable machining condition and the remaining machining condition to the learned model 33 to infer the machining condition adjustment direction corresponding to the machining condition predicted to be set by the operator (step S180). The inference unit 32 calculates a range in which a machining condition is selectable from the machining condition adjustment direction and sends the range in which a machining condition is selectable to the adjustment direction determination unit 13.

[0092] The inference unit 32 presents the range in which a machining condition can be selected to the operator by displaying the range in which a machining condition can be selected on the screen display 50 via the adjustment direction determination unit 13 (step S190). The range in which a machining condition can be selected includes information regarding the machining condition adjustment direction (reference axis).

[0093] Now, a procedure for the process in which the learning unit 31 learns the learned model 33 will be described with reference to Fig. 10 described. Fig. 10 is a flowchart illustrating a procedure for a learning process executed by the learning unit according to the second embodiment.

[0094] The learning unit 31 acquires learning data used for learning (step S210). Specifically, the learning unit 31 acquires the unsuitable machining condition and the remaining machining condition, as well as the machining condition adjustment direction corresponding to the machining condition set by the operator.

[0095] In step S210, the learning unit 31 simultaneously acquires the unsuitable machining condition, the remaining machining condition, and the machining condition adjustment direction corresponding to the machining condition set by the operator, but these can also be acquired at any time. That is, the learning unit 31 only needs to be able to receive the unsuitable machining condition and the remaining machining condition in conjunction with the machining condition adjustment direction corresponding to the machining condition set by the operator, and can acquire the unsuitable machining condition, the remaining machining condition, and the machining condition adjustment direction corresponding to the machining condition set by the operator at different times.

[0096] The learning unit 31 performs the learning process using the unsuitable machining condition and the remaining machining condition, and the machining condition adjustment direction corresponding to the machining condition set by the operator (step S220). Specifically, the learning unit 31 learns the machining condition adjustment direction corresponding to the unsuitable machining condition and the remaining machining condition through what is called supervised learning, according to the learning data generated based on the related combination of the unsuitable machining condition and the remaining machining condition and the machining condition adjustment direction corresponding to the machining condition set by the operator, and generates the learned model 33.

[0097] After generating the learned model 33, the learning unit 31 outputs the learned model 33 to the learning data storage unit 30 (step S230). The learning data storage unit 30 stores the learned model 33 generated by the learning unit 31.

[0098] Next, with reference to Fig. 11 describes a procedure for the process in which the inference unit 32 infers the machining condition adjustment direction using the learned model 33. Fig. 11 is a flowchart illustrating a procedure for an inference process executed by the inference unit according to the second embodiment.

[0099] The inference unit 32 acquires inference data to be used in inferring the machining condition adjustment direction (step S310). Specifically, the inference unit 32 acquires an unsuitable machining condition and a remaining machining condition. The inference unit 32 reads the learned model 33 from the learning data storage unit 30.

[0100] The inference unit 32 inputs an unsuitable machining condition and a remaining machining condition into the learned model 33 (step S320) to obtain a suitable machining condition adjustment direction.

[0101] The inference unit 32 outputs the data inferred using the learned model 33 (step S330). Specifically, the inference unit 32 outputs a suitable machining condition adjustment direction obtained by the learned model 33, thereby presenting the suitable machining condition adjustment direction to the operator via the adjustment direction determination unit 13 and the screen display 50.

[0102] Now, a hardware configuration of the machining condition determination support devices 1 and 2 will be described. Fig. 12 is a diagram illustrating an exemplary hardware configuration implementing the machining condition determination support apparatus according to the first and second embodiments.

[0103] Since the machining condition determination support devices 1 and 2 have similar hardware configurations, the hardware configuration of the machining condition determination support device 1 will be described here. The machining condition determination support device 1 can be implemented by an input device 300, a processor 100, a data storage 200, and an output device 400. The processor 100 is exemplified by a central processing unit (CPU, also referred to as a central computing device, processing device, computing device, microprocessor, microcomputer, or digital signal processor (DSP)) or a large-scale integration (LSI) system. The data storage 200 is exemplified by a random access memory (RAM) or a read-only memory (ROM).

[0104] The machining condition determination support device 1 is implemented by the processor 100, which reads and executes a computer-readable machining condition determination support program stored in the data storage 200 to perform the operation of the machining condition determination support device 1. It can also be said that the machining condition determination support program, which is a program for performing the operation of the machining condition determination support device 1, causes a computer to execute the procedure or method related to the machining condition determination support device 1.

[0105] The machining condition determination supporting program executed by the machining condition determination supporting device 1 has a module configuration including the simulation executing unit 11, the simulation displaying unit 12, the adjustment direction determining unit 13, the selection receiving unit 15, and the machining condition updating unit 14. In the machining condition determination support program, the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the selection reception unit 15, and the machining condition update unit 14 are loaded into the main storage device, and the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the selection reception unit 15, and the machining condition update unit 14 are generated on the main storage device.

[0106] The input device 300 receives an operation by the operator and sends the operation to the processor 100. The data storage 200 corresponds to the storage unit 20. The data storage 200 stores the material shape model 21, the tool model 22, the tool movement command data 23, and the machining condition set 24. In addition, the data storage 200 stores the machining simulation result, the adjustment direction of the machining condition, and the like, which were calculated by the processor 100.

[0107] The output device 400 outputs the machining condition used by the processor 100 for machining simulation, the machining simulation result calculated by the processor 100, and the like on the display screen 50. The output device 400 outputs the adjustment direction of the machining condition calculated by the processor 100 on the display screen 50. The output device 400 can also output the machining condition set by the operator on the display screen 50.

[0108] The material shape model 21, the tool model 22, the tool movement command data 23, and the machining condition set 24 in the data memory 200 are read by the processor 100. The data memory 200 is also used as temporary storage when the processor 100 executes various processes.

[0109] The machining condition determination support program may be stored on a computer-readable storage medium in an installable or executable file and provided as a computer program product. Alternatively, the machining condition determination support program may be provided to the machining condition determination support device 1 via a network, such as the Internet. Note that part of the functionality of the machining condition determination support device 1 may be implemented by dedicated hardware, such as a dedicated circuit, and the other part may be implemented by software or firmware.

[0110] One or more of the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the selection reception unit 15, and the machining condition update unit 14 may be implemented by the input device 300, the processor 100, the data storage 200, and the output device 400 described above.

[0111] Note that the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the selection reception unit 15, the machining condition update unit 14, the storage unit 20, the learning unit 31, the inference unit 32, and the learning data storage unit 30 may be arranged at different locations from each other. In this case, at least some of the simulation execution unit 11, the simulation display unit 12, the adjustment direction determination unit 13, the selection reception unit 15, the machining condition update unit 14, the storage unit 20, the learning unit 31, the inference unit 32, and the learning data storage unit 30 are connected to a network and transmit and receive data via the network.That is, the machining condition determination support processing described in the second embodiment can be implemented by a machining condition determination support system in which at least some of the simulation executing unit 11, the simulation displaying unit 12, the adjustment direction determining unit 13, the selection receiving unit 15, the machining condition updating unit 14, the storage unit 20, the learning unit 31, the inference unit 32, and the learning data storing unit 30 are connected to a network.

[0112] As described above, in the second embodiment, the machining condition determination support device 2 learns the relationship between the unsuitable machining condition and the remaining machining condition, and the machining condition adjustment direction corresponding to the machining condition set by the operator, and generates the learned model. Then, the machining condition determination support device 2 applies an unsuitable machining condition and a remaining machining condition to the learned model to infer the machining condition adjustment direction corresponding to the machining condition predicted to be set by the operator.As a result, the machining condition determination support device 2 can present an even more suitable machining condition adjustment direction than the machining condition determination support device 1, and can efficiently obtain a suitable machining condition in a shorter time than the machining condition determination support device 1.

[0113] The configurations described in the above embodiments are examples. The embodiments can be combined with other known technologies and with each other, and some of the configurations may be omitted or modified within a range that does not deviate from the gist of the invention. List of reference symbols 1.2 Machining condition determination support device; 11 Simulation execution unit; 12 simulation display unit; 13 Adjustment direction determination unit; 14 Processing condition update unit; 15 Selection receiving unit; 20 storage units; 21 Material form model; 22 tool model; 23 tool movement command data; 24 Machining condition set; 24a Processing condition; 30 learning data storage unit; 31 learning units; 32 inference unit; 33 learned model; 41 Motion command analysis unit; 42 numerical control unit; 43 Mechanical response calculation unit; 44 Cutting simulation unit; 45 Form post-processing; 50 screen display; 51 computer mouse; 52 keyboard; 61 movement command; 62 servo command; 63 servo FB data; 64 tool movement trajectory; 71, 72 screen display; 75 END button; 80 reference axis; 81 conical area; 100 processor; 200 data storage; 300 input device; 400 output device; P1 to P3 parameters; QA to QD point; RA to RD processing condition; SA to SD machining simulation result. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] JP 2004 - 280 267

[0005]

Claims

[1] A machining condition determination support device that supports the determination of a machining condition when machining a workpiece using a machine tool, the machining condition determination support device comprising: a simulation executing unit for performing machining simulation of the workpiece for each of a plurality of machining condition schemes based on a material shape model representing a shape of the workpiece before machining, a tool model representing a shape of a tool to be used for machining, and tool movement command data indicating the movement of the tool during the machining operation; a simulation display unit for displaying on a screen display a simulation result which is a result of the machining simulation; a selection receiving unit for receiving a selection from an operator for the displayed simulation result; and an adjustment direction determining unit for determining and presenting an adjustment direction of a machining condition of a machining simulation to be tried next, based on the selection received from the selection receiving unit. [2] The machining condition determination support device according to claim 1, wherein the selection receiving unit receives an unsuitable simulation result, which is a simulation result selected by an operator as being the most unsuitable from a point of view of machining quality with respect to the displayed simulation result, and the adjustment direction determining unit determines and presents an adjustment direction of a machining condition of a machining simulation to be tried next based on a dependency between an unsuitable machining condition, which is a machining condition corresponding to the unsuitable simulation result, and a remaining machining condition, which is a machining condition corresponding to a remaining simulation result, which is an unselected remaining simulation result. [3] The machining condition determination support device according to claim 2, further comprising: a machining condition updating unit for obtaining a next machining condition set by an operator along the presented adjustment direction, and for replacing the inappropriate machining condition with the received next machining condition. [4] A machining condition determination support device according to claim 2, wherein the machining condition includes several machining parameters, and the adjustment direction determining unit determines an adjustment direction of a machining condition for a machining simulation to be tried next based on a vector from the unsuitable machining condition to the remaining machining condition in a vector space based on the machining parameters. [5] The machining condition determination support device according to claim 4, wherein the simulation display unit displays the adjustment direction of the machining condition determined by the adjustment direction determination unit in the vector space based on the machining parameters. [6] The machining condition determination support device according to any one of claims 1 to 5, wherein the simulation display unit presents the simulation results by displaying the simulation results side by side. [7] The machining condition determination support device according to claim 2, further comprising: a learning unit for generating a learned model for inferring an adjustment direction of the next machining condition from the unsuitable machining condition, the remaining machining condition, and the machining condition set by the operator; and an inference unit to use the learned model to infer an adaptation direction of the next processing condition corresponding to the unsuitable processing condition and the remaining processing condition. [8] A machining condition determination support system that supports the determination of a machining condition when machining a workpiece using a machine tool, the machining condition determination support system comprising: a simulation executing unit for performing machining simulation of the workpiece for each of a plurality of machining condition schemes based on a material shape model representing a shape of the workpiece before machining, a tool model representing a shape of a tool to be used for machining, and tool movement command data indicating the movement of the tool during the machining operation; a simulation display unit for displaying on a screen display a simulation result which is a result of the machining simulation; a selection receiving unit for receiving a selection from an operator for the displayed simulation result; and an adjustment direction determining unit for determining and presenting an adjustment direction of a machining condition of a machining simulation to be tried next on the basis of the selection received from the selection receiving unit, wherein at least some of the simulation execution unit, the simulation display unit, the selection receiving unit and the adjustment direction determining unit are connected to a network. [9] A machining condition determination support method for supporting a determination of a machining condition in machining a workpiece using a machine tool, the machining condition determination support method comprising: a simulation execution step of executing, by means of a machining condition determination support device, a machining simulation of the workpiece for each of a plurality of machining condition schemes based on a material shape model representing a shape of the workpiece before machining, a tool model representing a shape of a tool to be used for machining, and tool movement command data indicating a movement of the tool during the machining operation; a simulation display step of displaying, on a screen display, a simulation result which is a result of the machining simulation by the machining condition determination supporting device; a selection receiving step of receiving, by the machining condition determination support device, a selection by an operator for the displayed simulation result; and an adjustment direction determination step of determining and presenting an adjustment direction of a machining condition for a machining simulation to be tried next based on the selection, by the machining condition determination support device. [10] A machining condition determination support program for supporting the determination of a machining condition when machining a workpiece using a machine tool, the machining condition determination support program causing a computer to execute: a simulation executing step of executing a machining simulation of the workpiece for each of a plurality of machining condition schemes based on a material shape model representing a shape of the workpiece before machining, a tool model representing a shape of a tool to be used in machining, and tool movement command data indicating the movement of the tool during the machining operation; a simulation display step of displaying a simulation result, which is a result of the machining simulation, on a screen display; a selection receiving step of receiving a selection by an operator for the displayed simulation result; and an adjustment direction determining step of determining and presenting an adjustment direction for a machining condition of a machining simulation to be tried next based on the selection.

Images