JP7852406B2 - Processing support system - Google Patents

Description

This invention relates to a processing support system.

Conventionally, it is known that to determine processing conditions in industrial machinery, the knowledge of skilled workers is compiled into a database and utilized. For example, Patent Document 1 describes a database-based knowledge model. A knowledge model is composed of multiple factors defined by various technical terms and the relationships between these factors. It is claimed that processing conditions can be optimized using this knowledge model.

Using the machining support system based on the knowledge model described above, inputting workpiece information, target cycle time, target machining accuracy, etc., as input factors will output cycle time, machining conditions, and machining accuracy predictions as output factors.

Japanese Patent Publication No. 2020-177547

According to the above technology, when a set of input factors, including workpiece information, target cycle time, and target machining accuracy, is input, only one set of output factors corresponding to this set of input factors is output. Therefore, if multiple machining conditions are required, or if customer requirements change, multiple sets of input factors must be input, making the input process cumbersome.

This invention was made in view of the above problems, and aims to provide a processing support system that can output multiple sets of output factor information by simply inputting one set of input factor information.

One aspect of the present invention is,
A machining support system that utilizes a knowledge model that defines the relationships between factors in the machining of a workpiece by a machining device,
A storage device for storing the aforementioned knowledge model,
A processing unit that performs calculations using the aforementioned knowledge model,
Equipped with,
The aforementioned arithmetic processing unit is
A basic input factor information acquisition unit that acquires a set of basic input factor information,
An input factor information generation unit generates multiple sets of variation input factor information that are different from the set of basic input factor information based on the set of basic input factor information,
An output factor information generation unit generates, based on the knowledge model, a set of basic output factor information corresponding to the set of basic input factor information, and a set of variation output factor information corresponding to the set of variation input factor information.
A selection unit that selects at least one set of teaching output factor information from the aforementioned multiple sets of variation output factor information,
A teaching processing unit that teaches the aforementioned set of basic output factor information and teaches at least the selected set of teaching output factor information,
Equipped with ,
The aforementioned set of basic input factor information includes the target machining accuracy,
The selection unit is part of a machining support system that selects at least one set of teaching output factor information before and after the target machining accuracy from among the multiple sets of variation output factor information, based on the target machining accuracy .
Furthermore, other embodiments of the present invention include:
A machining support system that utilizes a knowledge model that defines the relationships between factors in the machining of a workpiece by a machining device,
A storage device for storing the aforementioned knowledge model,
A processing unit that performs calculations using the aforementioned knowledge model,
Equipped with,
The aforementioned arithmetic processing unit is
A basic input factor information acquisition unit that acquires a set of basic input factor information,
An input factor information generation unit generates multiple sets of variation input factor information that are different from the set of basic input factor information based on the set of basic input factor information,
An output factor information generation unit generates, based on the knowledge model, a set of basic output factor information corresponding to the set of basic input factor information, and a set of variation output factor information corresponding to the set of variation input factor information.
A selection unit that selects at least one set of teaching output factor information from the aforementioned multiple sets of variation output factor information,
A teaching processing unit that teaches the aforementioned set of basic output factor information and teaches at least the selected set of teaching output factor information,
Equipped with,
The selection unit is part of a machining support system that selects at least one set of teaching output factor information before and after a predicted value change point from among the multiple sets of variation output factor information, based on the predicted value change point where the predicted value of the machining accuracy of the workpiece changes.

According to one and other embodiments of the present invention, a set of output factor information corresponding to a set of input factor information, and at least one set of teaching output factor information are displayed. As a result, multiple sets of output factor information can be obtained by inputting only one set of input factor information, thus reducing the effort required to input input factor information.

As described above, according to the above embodiment, a processing support system can be provided that outputs multiple sets of output factor information by inputting only one set of input factor information.

This diagram shows the configuration of the processing support system according to Embodiment 1. This figure shows a arithmetic processing unit, storage device, input device, and display device according to Embodiment 1. This figure shows a knowledge network diagram that represents a knowledge model. This is a plan view showing the processing machine. This is a plan view showing a workpiece. This is a flowchart showing the main routine of the machining support system. This is a flowchart showing the output factor information generation process. This is a flowchart showing the selection process. This diagram shows the relationship between the basic input factor, the variation input factor, the basic output factor, the variation output factor, and the instructional output factor. This graph shows the changes in cycle time and runout depending on the rough finishing division method. This is a flowchart showing the selection process in the machining support system according to Embodiment 2. This graph shows the information of the teaching output factors selected before and after the change in roundness. This is a flowchart showing the main routine of the machining support system according to Embodiment 5. This is a flowchart of the machine learning process.

(Embodiment 1)
1. Configuration of the Machining Support System The configuration of the machining support system of Embodiment 1 will be described with reference to Figures 1 and 2. The machining support system is applied to the machining field, which processes workpieces. The machining field includes, for example, cutting, grinding, electrical discharge machining, and press working. The machining support system according to this embodiment is a system that utilizes a knowledge model that describes the knowledge of technical information possessed by skilled personnel in the machining field.

As shown in Figure 1, the machining support system 1 comprises a storage device 2 and an arithmetic processing unit 3. An input device 4 and a display device 5 are connected to the arithmetic processing unit 3. The machining support system 1 forms a network with multiple machining devices 6. In other words, the multiple machining devices 6 and the machining support system 1 are configured to communicate with each other.

The storage device 2 stores the knowledge model M, machine configuration information of the processing device 6, etc. The arithmetic processing unit 3 performs calculations using the knowledge model M. The processing device 6 can be, for example, a grinding machine, a turning machine, a machining center, a milling machine, a gear cutting machine, a boring machine, etc. In this embodiment, a grinding machine will be used as an example.

The machining support system 1 uses a knowledge model M to assist in the operation of the machining apparatus 6, determine operating conditions, assess the machine state, and determine the state of the workpiece W.

Generally, in the field of machining, operators determine machining conditions such as cutting speed and depth of cut per unit time by considering various information such as the material of the workpiece W, the material of the tool, the quality of the workpiece W, and the machining cycle time. In this case, a knowledge model M is a model of the operator's thought process when acquiring various input information and determining the machining conditions.

In other words, the knowledge model M defines factors such as the material of the workpiece W, the material of the tool, the quality of the workpiece W, the machining cycle time, the cutting speed, and the depth of cut, as well as each industrial technology element that appears in the operator's thought process, and the relationships between these factors are defined.

For example, when the machining support system 1 determines the machining conditions for the machining device 6, it can obtain input factors such as the material and quality of the workpiece W, and then, by using the knowledge model M, output machining conditions such as cutting speed and depth of cut as output factors.

As shown in Figure 2, the arithmetic processing unit 3 comprises a basic input factor information acquisition unit 3a, an input factor information generation unit 3b, an output factor information generation unit 3c, a selection unit 3d, and a teaching processing unit 3e.

The basic input factor information acquisition unit 3a acquires basic input factor information from the input device 4. Basic input factor information is a set of multiple input factor information. Basic input factor information includes workpiece information (shape, material, etc. of workpiece W), target cycle time, target machining accuracy (target roundness, target chatter, target runout, target phase angle, target burn, etc.).

The input factor information generation unit 3b generates multiple sets of variation input factor information from a set of basic input factor information. Variation input factor information is a set of multiple input factor information combined into one. Variation input factor information is different from basic input factor information. In other words, at least one input factor information included in the variation input factor information is different from the input factor information that constitutes the basic input factor information.

The output factor information generation unit 3c generates a set of basic output factor information and multiple sets of variation output factor information based on the knowledge model M stored in the storage device 2. The basic output factor information is a set of multiple output factor information grouped together and corresponds to the set of basic input factor information described above. Similarly, the variation output factor information is a set of multiple output factor information grouped together and corresponds to the variation input factor information described above. The basic output factor information and variation output factor information include cycle time, machining conditions (such as the rough finishing division method), machine configuration information of the machining device 6, and machining accuracy prediction results (roundness, chatter, runout, phase angle, burn, etc.).

The selection unit 3d selects at least one set of instructional output factor information from the multiple sets of variation output factor information generated by the output factor information generation unit 3c. The selection unit 3d may select one set of instructional output factor information, or it may select two or more sets.

The teaching processing unit 3e provides instruction to the operator by displaying a set of basic output factor information and at least one selected set of teaching output factor information on the display device 5. The teaching processing unit 3e may also provide instruction to the operator via a device other than the display device 5, such as a printing device, providing the set of basic output factor information and at least one selected set of teaching output factor information.

2. Knowledge Network Diagram N
Conceptually, the knowledge model M is represented in network form. An example of a knowledge network diagram N, which represents the knowledge model M in network form, will be explained with reference to Figure 3. In other words, the knowledge network diagram N is a graphical representation of the knowledge model M. In this example, we will use a knowledge network diagram N relating to the knowledge model M in the field of machining.

As shown in Figure 3, the knowledge network diagram N comprises multiple node shapes 21 and link shapes 22 connecting the node shapes 21. In this embodiment, the node shapes 21 are represented by letters enclosed in rectangles. In this embodiment, the link shapes 22 are represented by straight lines. However, the link shapes 22 may also be represented by arrow lines to indicate the direction of the definitions between factors. The node shapes 21 represent factors defined in industrial technology terminology in the knowledge model M.

Multiple factors can have either a technical hierarchical relationship (also called a superior-subordinate relationship) or a technical heterogeneous dependency relationship. In other words, the relationships between factors can be classified into these two types.

For example, machine specifications encompass rigidity, machine precision, responsiveness, and operating speed. In other words, as factors with a technical inclusion relationship, machine specifications are considered a higher-level conceptual factor, while rigidity, machine precision, responsiveness, and operating speed are considered lower-level conceptual factors. However, machine specifications may also include factors other than those mentioned above.

Examples of factors with technically heterogeneous dependencies include stiffness and cycle time. In the following, the relationship between two factors with a technical inclusion relationship will be simply referred to as an inclusion relationship, and the relationship between two factors with technically heterogeneous dependencies will be simply referred to as a heterogeneous dependency relationship.

Link figures 22 represent the relationships between node figures 21. Link figures 22 also represent the relationships between factors in the knowledge model M.

3. Processing equipment 6
The processing apparatus 6 in this embodiment will be described with reference to Figures 4 and 5. In this embodiment, a grinding machine will be used as an example of the processing apparatus 6. However, the processing apparatus 6 is not limited to a grinding machine, and any processing machine such as a lathe, machining center, milling machine, gear machine, boring machine, etc., can be appropriately selected.

The processing device 6 rotates the workpiece W around the center line C, rotates the grinding wheel 16 (which acts as a tool), and moves the grinding wheel 16 relatively close to the workpiece W in a direction intersecting the workpiece W's axis, thereby grinding the outer or inner surface of the workpiece W. The processing device 6 can be a table traverse type grinding machine, a grinding wheel base traverse type grinding machine, etc. Furthermore, the processing device 6 can be a cylindrical grinding machine, a cam grinding machine, etc.

In this embodiment, as shown in Figure 5, the workpiece W is, for example, a camshaft formed in an axial shape, and the outer circumferential surface of the workpiece W is the part to be machined. However, the shape of the workpiece W is not limited to a camshaft; it can be any shape, such as a cylinder or a cylindrical shape with an inner circumferential surface. If the workpiece W is cylindrical, the inner circumferential surface of the workpiece W can be the part to be machined.

In this embodiment, the workpiece W is substantially rod-shaped and supported at both ends by workpiece support members.

The configuration of the processing apparatus 6 will be explained with reference to Figure 4. In this embodiment, the processing apparatus 6 is an example of a cam grinding machine with a grinding wheel base traverse. However, a table traverse type can also be used for the processing apparatus 6. The processing apparatus 6 mainly comprises a bed 11, a headstock 12, a tailstock 13, a traverse base 14, a grinding wheel base 15, a grinding wheel 16, a sizing device 17, a grinding wheel adjustment device 18, and a coolant device 19.

The bed 11 is fixed to the mounting surface. The headstock 12 is located on the upper surface of the bed 11, on the front side in the X-axis direction (lower side in Figure 4) and one end side in the Z-axis direction (left side in Figure 4). The headstock 12 supports the workpiece W so that it can rotate around the Z-axis with respect to the center line C. The workpiece W is rotated by the drive of a motor 12a provided on the headstock 12. The tailstock 13 is located on the upper surface of the bed 11, opposite the headstock 12 in the Z-axis direction, that is, on the front side in the X-axis direction (lower side in Figure 4) and the other end side in the Z-axis direction (right side in Figure 4). In other words, the headstock 12 and tailstock 13 rotatably support the workpiece W at both ends.

As shown in Figure 5, the workpiece W comprises a cylindrical shaft portion 30 with its major axis in the Z-axis direction, and a plurality of cams 31-35 (five in this embodiment) arranged on the shaft portion 30. The plurality of cams are, from left to right in Figure 5, the first cam 31, the second cam 32, the third cam 33, the fourth cam 34, and the fifth cam 35. The first to fifth cams 31-35 are identical in shape and size. Although not shown in detail, the first to fifth cams 31-35 have a roughly egg-like shape when viewed from the Z-axis direction.

As shown in Figure 4, the traverse base 14 is mounted on the upper surface of the bed 11 so as to be movable in the Z-axis direction. The traverse base 14 moves by the drive of a motor 14a mounted on the bed 11. The grinding wheel holder 15 is mounted on the upper surface of the traverse base 14 so as to be movable in the X-axis direction. The grinding wheel holder 15 moves by the drive of a motor 15a mounted on the traverse base 14. The grinding wheel 16 is rotatably supported on the grinding wheel holder 15. The grinding wheel 16 rotates by the drive of a motor 16a mounted on the grinding wheel holder 15. The grinding wheel 16 is constructed by fixing multiple abrasive grains with a bonding agent.

The sizing device 17 measures the dimensions (diameter) of the workpiece W. The sizing device 17 functions as a detector 20 for obtaining the actual depth of cut in the workpiece W.

The grinding wheel correction device 18 corrects the shape of the grinding wheel 16. The grinding wheel correction device 18 is a device that performs truing of the grinding wheel 16. In addition to truing, or as an alternative to truing, the grinding wheel correction device 18 may also perform dressing of the grinding wheel 16. Furthermore, the grinding wheel correction device 18 also has a function to measure the dimensions (diameter) of the grinding wheel 16.

Here, truing is a reshaping process, which involves reshaping the grinding wheel 16 to match the shape of the workpiece W when the wheel wears down due to grinding, and eliminating runout of the grinding wheel 16 caused by uneven wear. Dressing is a sharpening process, which involves adjusting the amount of abrasive grain protrusion and creating cutting edges for the abrasive grains. Dressing corrects issues such as clogged or damaged grains and is usually performed after truing.

The coolant system 19 supplies coolant from the coolant nozzle to the grinding point of the workpiece W by the grinding wheel 16. The coolant system 19 cools the recovered coolant to a predetermined temperature and supplies it again to the grinding point. The coolant system 19 allows for adjustment of the coolant flow rate and supply timing. In Figure 4, reference numeral 19 indicates the position of the coolant nozzle. Although not shown, a temperature sensor (detector 20) for acquiring the temperature of the recovered coolant may also be provided.

4. Operation of Machining Support System 1 Next, the operation of Machining Support System 1 will be explained with reference to Figures 6 to 8.

4-1. Input Processing (S1)
As shown in Figure 6, when the machining support system 1 is started, input processing (S1) is executed. In input processing (S1), a set of basic input factor information is input to the basic input factor information acquisition unit 3a from the operator via an input device 4 such as a keyboard. However, a set of basic input factor information may also be input to the basic input factor information acquisition unit 3a from an external storage device such as a USB memory. As a result, the basic input factor information acquisition unit 3a acquires a set of basic input factor information.

4-2. Input Factor Information Generation Process (S2)
Next, the arithmetic processing unit 3 executes the input factor information generation process (S2). More specifically, the input factor information generation unit 3b generates multiple sets of variation input factor information based on each basic input factor information included in the acquired set of basic input factor information.

For example, let us describe a case where a set of basic input factor information includes a rough finishing division method as one of the basic input factor information. As described above, the workpiece W in this embodiment is a camshaft. There are ²⁵ possible rough finishing division methods that can be performed on the camshaft, corresponding to whether or not rough finishing division is performed on each of the first to fifth cams 31 to 35. The input factor information generation unit 3b generates multiple sets of variation input factor information corresponding to the ²⁵ possible methods.

The input factor information generation unit 3b may generate variation input factor information for all factor information included in a set of basic input factor information, or it may generate variation input factor information for some of the factor information.

4-3. Output Factor Information Generation Process (S3)
Next, the output factor information generation unit 3c executes the output factor information generation process (S3). Figure 7 shows a flowchart of the output factor information generation process.

The output factor information generation unit 3c acquires the knowledge model M from the storage device 2 (S31). Subsequently, the output factor information generation unit 3c acquires a set of basic input factor information and multiple variation input factor information (S32).

Next, the output factor information generation unit 3c generates one set of basic output factor information from one set of basic input factor information based on the knowledge model M (S33). Furthermore, the output factor information generation unit 3c generates multiple sets of variation output factor information from multiple sets of variation input factor information based on the knowledge model M (S33).

The output factor information generation process (S3) is completed. Upon completion of the output factor information generation process (S3), the selection process (S4) is executed.

4-4. Selection Process (S4)
Next, as shown in Figure 6, the selection unit 3d performs the selection process (S4). Figure 8 shows a flowchart of the selection process. When the selection process (S4) is performed, the selection unit 3d obtains the target machining accuracy from a set of basic input factor information (S41).

Next, the selection unit 3d calculates the change point at which the machining accuracy of the workpiece W changes (S42). For example, if runout is included as an example of the target machining accuracy of the workpiece W, the selection unit 3d calculates the predicted runout value for, for example, the ²⁵ types of rough finishing division described above. Then, it selects the rough finishing division before and after the target runout (S43).

Furthermore, if the cycle time is included as an example of the target machining accuracy of the workpiece W, the selection unit 3d calculates a predicted value of the cycle time for, for example, the ^2-5 rough finishing division patterns described above (S42). Then, it selects the rough finishing division patterns before and after the target cycle time (S43).

Furthermore, in the selection criteria for this configuration, the case where rough finishing division is performed on all of the first to fifth cams 31 to 35 is selected as a reference example.

The selection process (S4) is now complete. Upon completion of the selection process (S4), the teaching process (S5) is executed.

4-5. Instructional Processing (S5)
Next, as shown in Figure 6, the teaching processing unit 3e performs teaching processing (S5). Specifically, the teaching processing unit 3e displays a set of basic output factor information and at least one set of teaching output factor information on the display device 5. As a result, the set of basic output factor information and at least one set of teaching output factor information are taught to the operator.

Once the teaching process (S5) is complete, the operation of the machining support system 1 ends.

5. Relationship between Input and Output Factors During Processing The relationship between input and output factors during various processes will be explained with reference to Figure 9. In the input process (S1), the basic input factor information acquisition unit 3a acquires a set of basic input factor information. In Figure 9, the set of basic input factor information is represented by "In 1" enclosed in a rectangle.

Next, in the input factor information generation process (S2), the input factor information generation unit 3b generates multiple sets of variation input factor information from one set of basic input factor information. In Figure 9, the eight sets of variation input factor information are represented by the rectangles "In 2" to "In 9". The number of sets of variation input factor information is not limited to eight; it can be any number.

In the output factor information generation process (S3), the output factor information generation unit 3c generates one set of basic output factor information based on one set of basic input factor information. In Figure 9, the set of basic output factor information is represented by "Out 1," enclosed in a double-lined rectangle. The output factor information generation unit 3c generates multiple sets of corresponding variation output factor information based on multiple sets of variation input factor information. In Figure 9, eight sets of variation output factor information are represented by "Out 2" to "Out 9," enclosed in double-lined rectangles. The number of variation output factor information is the same as the number of variation input factor information.

In the selection process (S4), the selection unit 3d selects at least one set of instructional output factor information from multiple sets of variation output factor information. In Figure 9, the four sets of instructional output factor information are represented by double-lined rectangles labeled "Out 3," "Out 4," "Out 7," and "Out 8." The number of instructional factor information sets is not limited to four; any number of sets can be selected.

In the teaching process (S5), the teaching processing unit 3e displays one set of basic output factor information and at least one set of teaching output factor information on the display device 5. In Figure 9, the display device 5 shows the basic output factor information represented by "Out 1" enclosed in a double-lined rectangle, and the four sets of teaching output factor information selected in the selection process (S4) ("Out 3," "Out 4," "Out 7," and "Out 8").

The information shown in Figure 9 is a general explanation of the relationship between input and output factors during various processes; therefore, it does not limit the explanations of the input and output factor information examples described below.

6. Examples of Input Factor Information and Output Factor Information Next, examples of input factor information and output factor information relating to this form will be explained. However, the content and types of input factor information and output factor information are not limited to the following explanation.

In this configuration, one set of basic input factor information includes the target cycle time and target deviation. Other basic input factor information is omitted.

In this configuration, in the selection process (S4), the following four configurations are selected according to predetermined selection criteria.
- When rough finishing division is performed on all of the first to fifth cams 31 to 35 - When rough finishing division is performed only on the second, third and fourth cams 32, 33, and 34 - When rough finishing division is performed only on the third cam 33 - When rough finishing division is not performed on all of the first to fifth cams 31 to 35

When rough finishing division is performed on all of the first to fifth cams 31 to 35, the machining accuracy (e.g., runout) of the workpiece W improves, but the cycle time increases. Furthermore, when machining the workpiece W with a grinding machine, the area near the center of the workpiece W in the Z-axis direction is prone to deflection due to the load applied during grinding. Therefore, by improving the machining accuracy of the third cam 33, which is fixed near the center in the Z-axis direction among the first to fifth cams 31 to 35, the overall machining accuracy of the workpiece W can be improved. For this reason, if rough finishing division is performed at only one location, it is preferable to perform it on the third cam 33. Also, if rough finishing division is performed at multiple locations, it is preferable to perform it on the third cam 33, as well as on the second cam 32 and fourth cam 34 symmetrically with respect to the third cam 33.

Figure 10 shows how cycle time and runout change depending on the type of rough finishing division, with the horizontal axis representing cycle time and the vertical axis representing runout.

If rough finishing division is not performed on all of the first to fifth cams 31-35, the predicted runout of the workpiece W is approximately the same as the target runout, which is the target runout of the workpiece W after machining. The upper and lower limits of the predicted runout of the workpiece W are indicated by error bars. In other words, the predicted runout of the workpiece W includes the target runout between the upper and lower limits. On the other hand, the predicted cycle time is shorter than the target cycle time. In the prior art, because the upper limit of the predicted runout is greater than the target runout, the output factor information for cases where rough finishing division is not performed on all of the first to fifth cams 31-35 was not calculated at all. In this embodiment, the case where rough finishing division is not performed on all of the first to fifth cams 31-35 corresponds to teaching output factor information.

When rough finishing division is performed only on the third cam 33, the predicted runout of the workpiece W is smaller than the target runout in both its upper and lower limits. On the other hand, the predicted cycle time is shorter than the target cycle time. In the prior art, because the predicted runout was smaller than the target runout and the predicted cycle time was shorter than the target cycle time, the case where rough finishing division was performed only on the third cam 33 was calculated and communicated to the operator. In this embodiment, the case where rough finishing division is performed only on the third cam 33 corresponds to the basic output factor information.

When rough finishing division is performed on the second to fourth cams 34, the predicted runout of the workpiece W is smaller than the target runout at both the upper and lower limits. On the other hand, the predicted cycle time is longer than the target cycle time. In the conventional technology, because the predicted cycle time is longer than the target cycle time, the output factor information when rough finishing division is performed on the second to fourth cams 34 is not calculated at all. In this embodiment, when rough finishing division is performed on the second to fourth cams 34, it corresponds to teaching output factor information.

When rough finishing division is performed on all of the first to fifth cams 31-35, the predicted runout of the workpiece W is smaller than the target runout at both the upper and lower limits. On the other hand, the predicted cycle time is longer than the target cycle time. In the conventional technology, because the predicted cycle time is longer than the target cycle time, the output factor information when rough finishing division is performed on all of the first to fifth cams 31-35 was not calculated at all. In this embodiment, when rough finishing division is performed on all of the first to fifth cams 31-35, it corresponds to teaching output factor information.

7. Effects of this Embodiment In this embodiment, a set of output factor information corresponding to a set of input factor information that has been input, and at least one set of teaching output factor information are displayed. As a result, even when multiple processing conditions are required or when customer requests change, multiple sets of output factor information can be obtained by inputting only one set of input factor information, thus saving labor in the input of input factor information.

In this embodiment, one set of basic input factor information includes the target machining accuracy, and the selection unit 3d selects at least one set of teaching output factor information from among multiple sets of variation output factor information based on the target machining accuracy.

By inputting a set of input factor information, including the target machining accuracy, it is possible to obtain teaching output factor information regarding the machining accuracy of the workpiece W. This eliminates the need to input multiple sets of input factor information and engage in trial and error to explore the machining accuracy of the workpiece W, thus streamlining the input process.

Furthermore, in this embodiment, the selection unit 3d selects at least one set of teaching output factor information before and after the target machining accuracy.

By inputting a single set of input factor information, it is possible to obtain teaching output factor information before and after the target machining accuracy. This eliminates the need to input multiple sets of input factor information and engage in trial and error to explore the output factors before and after the machining accuracy of the workpiece W, thus streamlining the input process.

(Embodiment 2)
Next, Embodiment 2 will be described with reference to Figures 11 to 12. In this embodiment, one set of basic input factor information includes the target roundness, which is the target value of the roundness of the workpiece W.

Furthermore, in this embodiment, the configuration of the selection process (S4) differs from that of Embodiment 1. Figure 11 shows a flowchart of the selection process (S4) according to this embodiment.

When the selection process (S4) is executed, the selection unit 3d acquires the target machining accuracy, which will serve as the selection criterion, from a set of basic input factor information (S44). In this embodiment, the arithmetic processing unit 3 acquires the target roundness.

Next, in S45, the selection unit 3d calculates predicted values for the physical properties of the workpiece W. In this embodiment, the arithmetic processing unit 3 calculates predicted values for roundness for every 1-second increase in cycle time, from a lower limit (e.g., 20 seconds) to an upper limit (e.g., 10,000 seconds). However, the increase in cycle time may be any value of 2 seconds or more.

Next, the selection unit 3d calculates the predicted value change point where the predicted physical properties of the workpiece W change within the range from the lower limit to the upper limit of the cycle time (S45). In this embodiment, the predicted value change point where the predicted value of roundness changes abruptly is calculated.

Next, the selection unit 3d selects the cycle times corresponding to the points before and after the change in the predicted roundness value (S46). That is, it selects the cycle time before the rapid change in the predicted roundness value and the cycle time after the rapid change in the predicted roundness value.

Figure 12 shows a graph illustrating the change in roundness, with the vertical axis representing roundness and the horizontal axis representing cycle time. In this embodiment, the roundness R1 and cycle time T1 before the rapid decrease in roundness, and the roundness R2 and cycle time T2 after the rapid decrease in roundness are selected.

Furthermore, among the reference numerals used in Embodiment 2 and later, those identical to those used in previously described embodiments represent the same components, etc., as those in the previously described embodiments, unless otherwise specified.

In this embodiment, the selection unit 3d selects at least one set of teaching output factor information from among multiple sets of variation output factor information, based on the predicted value change point where the predicted machining accuracy of the workpiece W changes.

By inputting a single set of input factor information, it is possible to obtain teaching output factor information for the point where the predicted machining accuracy of the workpiece W changes. This eliminates the need to input multiple sets of input factor information and engage in trial and error to find the point where the machining accuracy of the workpiece W changes, thus streamlining the input process.

The selection unit 3d selects at least one set of teaching output factor information before and after the predicted value change point.

By inputting a single set of input factor information, it is possible to obtain teaching output factor information before and after the predicted value change point. This eliminates the need to input multiple sets of input factor information and engage in trial and error to search for output factor information before and after the change point where the machining accuracy of the workpiece W changes, thus streamlining the input process.

One set of basic input factor information includes the target cycle time, while multiple sets of variation input factor information include target cycle times different from those of the one set of basic input factor information.

By inputting a single set of input factor information, multiple target cycle times can be obtained. Based on these multiple target cycle times, multiple cycle times can then be obtained as output factor information. As a result, there is no need to input multiple sets of input factor information and engage in trial and error to explore cycle times, thus streamlining the input process.

(Embodiment 3)
Next, Embodiment 3 will be described. In this embodiment, the set of basic input factor information includes, for example, the factor information listed in Table 1.

As shown in Table 1, a set of basic input factor information for this embodiment includes the target cycle time (100 seconds), target roundness (2 μm), and target runout (20 μm).

The arithmetic processing unit 3, through teaching processing (S5), displays the basic output factor information shown in Table 2 on the display device 5.

As shown in Table 2, the basic output factor information includes cycle time prediction, roundness prediction, runout prediction, and interval prediction. The interval related to the interval prediction may be either the truing interval or the dressing interval.

Each predicted value included in the basic output factor information is calculated to match each target value included in the basic input factor information. Specifically, the predicted cycle time (99 seconds) is shorter than the target cycle time (100 seconds), the predicted roundness (1.0–1.5 μm) is smaller than the target roundness (2 μm), and the predicted runout (3–10 μm) is smaller than the target runout (20 μm). Furthermore, the interval predicted value in the basic output factor information for this configuration is set to 50.

Furthermore, the processing unit 3 selects the teaching output factor information listed in Tables 3 to 7 from multiple sets of variation output factor information based on the selection criteria stored in the storage device 2 (S4), and displays the teaching output factor information on the display device 5 through teaching processing (S5). The teaching output factor information and selection criteria listed in each table are explained below.

The teaching output factor information listed in Table 3 was selected based on the selection criterion of selecting variation factor information that includes interval prediction values that are a predetermined amount larger than the interval prediction values included in the basic output factor information related to Table 2. In this configuration, a set of variation output factor information containing 100 interval prediction values was selected, which is 50 more than the interval prediction values (50) related to the basic output factor information.

Comparing the basic output factor information listed in Table 2 with the teaching output factor information listed in Table 3, the predicted cycle time value in Table 3 (109 seconds) is longer than the predicted cycle time value in Table 2 (99 seconds).

The teaching output factor information listed in Table 4 was selected based on the selection criteria of selecting variation output factor information that includes deflection prediction values with improved accuracy by a predetermined amount compared to the deflection prediction values included in the basic output factor information related to Table 2. In this configuration, a set of variation output factor information corresponding to the same 50 interval prediction values as the 50 interval prediction values related to the basic output factor information listed in Table 2 was selected.

Comparing the basic output factor information listed in Table 2 with the teaching output factor information listed in Table 4, the predicted cycle time in Table 4 (109 seconds) is longer than the predicted cycle time in Table 2 (99 seconds). Furthermore, the predicted deflection value in Table 4 (0–3 μm) is smaller than the predicted deflection value in Table 2 (3–10 μm).

The teaching output factor information listed in Table 5 was selected based on two criteria: first, selecting variation output factor information that includes deflection prediction values with improved accuracy by a predetermined amount compared to the deflection prediction values included in the basic output factor information in Table 2; and second, selecting a set of variation output factor information corresponding to 100 interval prediction values, which is 50 more than the 50 interval prediction values listed in Table 2.

Comparing the basic output factor information in Table 2 with the teaching output factor information in Table 5, the predicted cycle time in Table 5 (119 seconds) is longer than the predicted cycle time in Table 2 (99 seconds). Furthermore, the predicted deflection range in Table 5 (0–3 μm) is smaller than the predicted deflection range in Table 2 (3–10 μm). Also, the interval in Table 4 (100 cycles) is greater than the interval in Table 2 (50 cycles).

The teaching output factor information listed in Table 6 was selected based on the selection criteria of selecting variation output factor information that includes a cycle time prediction value that is a predetermined amount shorter than the cycle time prediction value included in the basic output factor information related to Table 2. In Table 6, a set of variation output factor information corresponding to a cycle time prediction value of 89 seconds, which is 10 seconds shorter than the cycle time prediction value (99 seconds) listed in Table 2, was selected.

Comparing the basic output factor information listed in Table 2 with the teaching output factor information listed in Table 6, the predicted cycle time value in Table 6 (89 seconds) is shorter than the predicted cycle time value in Table 2 (99 seconds).

The teaching output factor information listed in Table 7 was selected based on the selection criteria of selecting variation output factor information that includes cycle time prediction values that are a predetermined amount shorter than the cycle time prediction values included in the basic output factor information related to Table 2. In Table 7, a set of variation output factor information corresponding to a cycle time prediction value of 79 seconds, which is 20 seconds shorter than the cycle time prediction value (99 seconds) listed in Table 2, was selected.

Comparing the basic output factor information in Table 2 with the teaching output factor information in Table 7, the predicted cycle time in Table 7 (79 seconds) is shorter than the predicted cycle time in Table 2 (99 seconds). Furthermore, the predicted roundness in Table 7 (1.5–2.0 μm) is greater than the predicted roundness in Table 2 (1.0–1.5 μm). Also, the predicted interval in Table 7 (30 intervals) is less than the predicted interval in Table 2 (50 intervals).

In this configuration, a set of output factor information corresponding to a set of input factor information, and at least one set of teaching output factor information are displayed. This allows for the acquisition of multiple sets of output factor information by simply inputting one set of input factor information, even when multiple processing conditions are required or when customer requirements change, thus streamlining the input process.

(Embodiment 4)
Next, Embodiment 4 will be described. In this embodiment, the operation of the input factor information generation process (S2) differs from that of Embodiment 1.

In the input factor information generation process (S2) according to this embodiment, the input factor information generation unit 3b acquires a set of basic input factor information. Next, the input factor information generation unit 3b acquires from the storage device 2 machine configuration information of another processing device 6, which is different from the machine configuration information of the processing device 6 included in the set of basic input factor information.

Next, the input factor information generation unit 3b generates multiple sets of variation input factor information by replacing the machine configuration information of the processing device 6 included in one set of basic input factor information with the machine configuration information of other processing devices 6.

In this embodiment, one set of basic input factor information includes the machine configuration information of the processing device 6, and multiple sets of variation input factor information include the machine configuration information of other processing machines that differ from the one set of basic input factor information.

By inputting a single set of input factor information, it is possible to obtain teaching output factor information for use with a different processing device 6 than the one initially input. This eliminates the need to input multiple sets of input factor information and engage in trial and error to find output factor information when the processing device 6 is changed, thus streamlining the input process.

(Embodiment 5)
Next, Embodiment 5 will be described with reference to Figures 13 and 14. Figure 13 shows a flowchart of the processing support system 1 according to this embodiment.

As shown in Figure 13, when the processing support system 1 according to this embodiment is started, input processing (S1) is executed. During input processing, a set of basic input factor information is input to the arithmetic processing unit 3 from the operator via an input device 4 such as a keyboard.

Next, the machine learning process (S6) is executed. As shown in Figure 14, when the machine learning process (S6) is executed, the arithmetic processing unit 3 acquires a set of basic input factor information. Next, the arithmetic processing unit 3 acquires from the storage device 2 the selection criteria used when the arithmetic processing unit 3 selects teaching output factor information from the variation output factor information in the selection process (S4).

Next, the processing unit 3 performs machine learning using a set of basic input factor information as training data. This updates the selection criteria for at least one set of teaching output factor information in the knowledge model M. The processing unit 3 stores the updated selection criteria in the storage device 2. This completes the machine learning process (S6).

As shown in Figure 13, once the machine learning process (S6) is completed, the input factor information generation process (S2) is executed. The subsequent processes are the same as in Embodiment 1, so their explanation is omitted.

In this configuration, the selection criteria for at least one set of instructional output factors in the knowledge model M are updated using machine learning, based on a set of basic input factor information received as training data.

The set of basic input factor information input into the machining support system 1 is tailored to the user's requirements. Therefore, by learning the set of basic input factor information requested by the user, the selection criteria for at least one set of teaching output factors in the knowledge model M are updated, allowing for the acquisition of teaching output factors that are more accurately adapted to the user's requirements.

The present invention is not limited to the embodiments described above, and can be applied to various embodiments without departing from its spirit.

When selecting teaching output factor information based on the change point where the predicted machining accuracy of the workpiece W changes, the type of machining accuracy of the workpiece W can be any of the following: roundness, runout, surface roughness, etc.

The system may also be configured to teach a predicted CP value, selected based on the CP value of the workpiece W, as teaching output factor information.

1. Machining support system, 2. Storage device, 3. Arithmetic processing unit, 3a. Basic input factor information acquisition unit, 3b. Input factor information generation unit, 3c. Output factor information generation unit, 3d. Selection unit, 3e. Teaching processing unit, 6. Machining device, M. Knowledge model, W. Workpiece

Claims (5)

A machining support system that utilizes a knowledge model that defines the relationships between factors in the machining of a workpiece by a machining device,
A storage device for storing the aforementioned knowledge model,
A processing unit that performs calculations using the aforementioned knowledge model,
Equipped with,
The aforementioned arithmetic processing unit is
A basic input factor information acquisition unit that acquires a set of basic input factor information,
An input factor information generation unit generates multiple sets of variation input factor information that are different from the set of basic input factor information based on the set of basic input factor information,
An output factor information generation unit generates, based on the knowledge model, a set of basic output factor information corresponding to the set of basic input factor information, and a set of variation output factor information corresponding to the set of variation input factor information.
A selection unit that selects at least one set of teaching output factor information from the aforementioned multiple sets of variation output factor information,
A teaching processing unit that teaches the aforementioned set of basic output factor information and teaches at least the selected set of teaching output factor information,
Equipped with ,
The aforementioned set of basic input factor information includes the target machining accuracy,
The selection unit is a machining support system that selects at least one set of teaching output factor information before and after the target machining accuracy from among the multiple sets of variation output factor information, based on the target machining accuracy . A machining support system that utilizes a knowledge model that defines the relationships between factors in the machining of a workpiece by a machining device,
A storage device for storing the aforementioned knowledge model,
A processing unit that performs calculations using the aforementioned knowledge model,
Equipped with,
The aforementioned arithmetic processing unit is
A basic input factor information acquisition unit that acquires a set of basic input factor information,
An input factor information generation unit generates multiple sets of variation input factor information that are different from the set of basic input factor information based on the set of basic input factor information,
An output factor information generation unit generates, based on the knowledge model, a set of basic output factor information corresponding to the set of basic input factor information, and a set of variation output factor information corresponding to the set of variation input factor information.
A selection unit that selects at least one set of teaching output factor information from the aforementioned multiple sets of variation output factor information,
A teaching processing unit that teaches the aforementioned set of basic output factor information and teaches at least the selected set of teaching output factor information,
Equipped with ,
The selection unit is a machining support system that selects at least one set of teaching output factor information before and after a predicted value change point from among the multiple sets of variation output factor information, based on the predicted value change point where the predicted value of the machining accuracy of the workpiece changes. The aforementioned set of basic input factor information includes the target cycle time,
The machining support system according to claim 1 or 2 , wherein the multiple sets of variation input factor information include a target cycle time different from that of the single set of basic input factor information. The aforementioned set of basic input factor information includes the mechanical configuration information of the processing apparatus,
The machining support system according to claim 1 or 2 , wherein the multiple sets of variation input factor information include machine configuration information of other machining equipment that is different from the one set of basic input factor information. The processing support system according to claim 1 or 2 , wherein the selection criteria for at least one set of teaching output factor information in a knowledge model are updated by machine learning using the input set of basic input factor information as training data.