Processing system, processing method and program

Abstract

The processing system includes: an input unit (11) for receiving setting data; a processing device (10) installed in a factory for processing an object based on the setting data; a model information acquisition unit (13) for acquiring model information representing a first model (41), which is applied to the setting data to suppress errors in the processing results of the processing device (10) due to inherent characteristics of the processing device (10) before being installed in the factory; and a learning unit (16) for learning a second model (42) to suppress errors between the processing results of the processing device and the target value. The processing device processes the object (21) in the factory using the first model (41) based on the setting data input in the factory. The processing device (10) processes the object (22) based on the results of applying the first model (41) and the second model (42) to the setting data.

Description

Technical Field

[0001] This disclosure relates to processing systems, processing methods, and procedures. Background Technology

[0002] In the field of FA (Factory Automation), various devices are used to build systems that implement processing steps similar to those on a production line. The devices that constitute such a system are typically a number of machines of a specific type, each with the capability appropriate for the processing steps to be performed. However, even devices of the same type may not perform at exactly the same capacity when installed in the field; sometimes, the capabilities of the devices may differ.

[0003] Therefore, a technique is considered that involves learning a model to estimate the output when conditions are changed, thereby estimating the capability difference of the device and enabling the device to perform processing that takes the capability difference into account (for example, see Patent Document 1). When using this technique, factors that may affect the capability of the device can be specified as conditions. Furthermore, if the capability difference can be estimated using a model, a model for obtaining an output that suppresses capability differences can also be obtained.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2021-170163 Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] However, there are many factors that may affect the device's ability to perform in the field, and model learning may generate a large computational load.

[0009] This disclosure was made under the above circumstances, and its purpose is to reduce the computational load of learning models with poor ability to perform suppression devices.

[0010] Methods for solving problems

[0011] To achieve the above objectives, the processing system of this disclosure comprises: an input unit for receiving setting data; a processing device installed in a factory for processing an object according to the setting data; a model information acquisition unit for acquiring model information representing a first model, which is applied to the setting data to suppress a first error contained in the first processing result of the processing device before being installed in the factory due to the inherent characteristics of the processing device; an environment information acquisition unit for acquiring factory environment information representing the environment of the processing device installed in the factory; an error information acquisition unit for acquiring first error information representing a second error between the second processing result of the processing device processing the object in the factory using the first model based on the setting data input in the factory and a target value; and a learning unit for learning a second model for suppressing the second error based on the factory environment information and the first error information, wherein the processing device processes the object based on the results of applying the first model and the second model to the setting data.

[0012] Invention Effects

[0013] According to this disclosure, the model information acquisition unit acquires model information representing a first model, which is applied to setting data to suppress a first error contained in the first processing result of the processing device before it is installed in the factory due to the inherent characteristics of the processing device. The learning unit learns a second model to suppress a second error between the second processing result of the processing device for the object being processed in the factory and the target value. Thus, separate from the first model for suppressing capability differences arising before installation in the factory, the second model for suppressing capability differences arising from the environment of the processing device installed in the factory is learned. Therefore, factors that may affect the capability of the device are identified, and the second model for addressing capability differences caused by these factors is learned. Therefore, the computational load for learning the model for suppressing capability differences of the device can be reduced. Attached Figure Description

[0014] Figure 1 This is a diagram showing an outline of the processing system of Embodiment 1.

[0015] Figure 2 This is a diagram showing the hardware structure of the processing device in Embodiment 1.

[0016] Figure 3 This is a diagram showing the functional structure of the processing device in Embodiment 1.

[0017] Figure 4 This is a diagram illustrating an example of model information for the first model of Embodiment 1.

[0018] Figure 5 This is a diagram illustrating an example of the second model of Embodiment 1.

[0019] Figure 6 This is a flowchart illustrating the model application process of Implementation Method 1.

[0020] Figure 7 This is a diagram showing the functional structure of the processing device in a modified example.

[0021] Figure 8 This is a diagram showing an outline of the processing apparatus of Embodiment 2.

[0022] Figure 9 This is a diagram showing the functional structure of the processing device in Embodiment 2.

[0023] Figure 10 This is a flowchart illustrating the first model generation process of Implementation Method 2.

[0024] Figure 11 This is a flowchart illustrating the model application process of Implementation Method 2.

[0025] Figure 12 This is a diagram showing the functional structure of the processing device 10 in Embodiment 3.

[0026] Figure 13 This is a diagram showing the structure of the formula library in Embodiment 3.

[0027] Figure 14 This is a diagram showing the functional structure of the processing device in Embodiment 4.

[0028] Figure 15 This is a diagram used to illustrate the estimation model of implementation method 4.

[0029] Figure 16 This is a diagram illustrating an example of the estimation model for Implementation 4.

[0030] Figure 17 This is a flowchart illustrating the estimation model application process of Implementation 4.

[0031] Figure 18 This is a flowchart illustrating the estimation model generation process of Implementation 4.

[0032] Figure 19 This is a flowchart illustrating the model improvement process of Implementation 5.

[0033] Figure 20 This is a diagram used to illustrate the search for the combination of the second model and the estimation model in Implementation 5.

[0034] Figure 21 This is a diagram showing the structure of the processing system in a modified example.

[0035] Figure 22This is a diagram showing the structure of the processing device in a modified example.

[0036] Figure 23 This is a diagram showing the first model with variations. Detailed Implementation

[0037] Hereinafter, the processing system of the present disclosure will be described in detail with reference to the accompanying drawings.

[0038] Implementation method 1.

[0039] like Figure 1 As shown, the processing system 1000 of this embodiment is constructed by installing processing device 10b, one of the processing devices 10a and 10b of the same model produced at the production site 100a, at the factory 100. The processing system 1000 includes processing device 10b. Hereinafter, without distinguishing between processing devices 10a and 10b, it will sometimes be referred to as processing device 10.

[0040] Processing device 10 is a FA device or equipment, such as a machine tool, for processing objects 20 and 21. Object 20 is the object to be processed in production site 100a, and object 21 is the object to be processed in factory 100. Processing by processing device 10 includes, for example, cutting or grinding objects 20 and 21 as workpieces, assembling objects 20 and 21 as products, or assembling objects 20 and 21 as part of a product to other components. Processing device 10 processes objects 20 and 21 according to setting data input by the user. Setting data refers to parameters set in processing device 10 to enable processing device 10 to process objects 21 and 22, such as the worktable supporting the workpiece, the position and speed of the tool, and the rotational speed of the tool.

[0041] However, the processing device 10 has mechanical errors during the production stage at the production site 100a. That is, processing devices 10a and 10b each have inherent characteristics, which may cause errors in the processing results of the object 20 at the production site 100a compared to the target value. Figure 1 The diagram shows processing device 10a having characteristic A and processing device 10b having characteristic B. The processing result is, for example, the cutting result of object 20, and the error in the processing result is the dimensional error of the object 20 being cut. The target value can be a specification value directly represented by the set data, or it can be a value expected by the user as a value corresponding to the set data. The target value may or may not be included in the set data. However, even if the target value is not included in the set data, it usually corresponds to the set data and can also be separately input by the user into processing device 10.

[0042] Since the processing device 10 may produce errors in the processing results due to its inherent characteristics, a first model for suppressing such errors is learned at the production site 100a, and the learned first model is assembled into the processing device 10, thereby homogenizing the capacity of the processing device 10 removed from the production site 100a.

[0043] Then, when the processing device 10 is installed in the factory 100, errors may occur in the processing results of the object 21 due to the installation environment of the processing device 10 in the factory 100. The installation environment includes, for example, temperature, humidity, or the type of material inserted into the processing device 10. Therefore, the processing device 10 learns a second model to suppress errors caused by the installation environment and processes the object 21 using the learned second model, thereby performing processing suitable for the installation environment. By learning and using the second model, the processing devices 10a and 10b reduce deviations in the processing results of the object 21 even under different environments, thus achieving homogenization capabilities. Furthermore, in Figure 1 The learning associated with processing device 10b is shown in the example. The learning associated with processing device 10a is also performed in the same way as that with processing device 10b.

[0044] The processing device 10 consists of hardware elements for functioning as a computer. Specifically, such as... Figure 2 As shown, the processing device 10 includes a processor 101, a main storage unit 102, an auxiliary storage unit 103, an input unit 104, an output unit 105, and a communication unit 106. The main storage unit 102, the auxiliary storage unit 103, the input unit 104, the output unit 105, and the communication unit 106 are all connected to the processor 101 via an internal bus 107.

[0045] The processor 101 includes a CPU (Central Processing Unit) as a processing circuit. The processor 101 performs various functions and executes the processing described later by executing the program P1 stored in the auxiliary storage unit 103.

[0046] The main storage unit 102 contains RAM. Program P1 is loaded from the auxiliary storage unit 103 into the main storage unit 102. Furthermore, the main storage unit 102 is used as the operating area of ​​the processor 101.

[0047] The auxiliary storage unit 103 includes non-volatile memory such as EEPROM (Electrically Erasable Programmable Read-Only Memory) and HDD (Hard Disk Drive). In addition to program P1, the auxiliary storage unit 103 also stores various data used in the processing of the processor 101. Following the instructions of the processor 101, the auxiliary storage unit 103 provides the processor 101 with data used by the processor 101. Furthermore, the auxiliary storage unit 103 stores data provided from the processor 101.

[0048] The input unit 104 includes input devices such as hard switches, input keys, keyboards, and indicator devices. The input unit 104 acquires information input by the user from the processing device 10 and notifies the processor 101 of the acquired information.

[0049] The output unit 105 includes output devices such as LEDs (Light Emitting Diodes), LCDs (Liquid Crystal Displays), and speakers. The output unit 105 displays various information to the user according to the instructions of the processor 101.

[0050] The communication unit 106 includes a communication interface circuit for communicating with an external device. The communication unit 106 receives signals from the outside and outputs data represented by those signals to the processor 101. Furthermore, the communication unit 106 sends signals representing data output from the processor 101 to the external device.

[0051] Through the aforementioned hardware structure, the processing device 10 performs various functions within the factory 100. Specifically, as... Figure 3 As shown, the processing device 10, as its function, includes: an input unit 11 into which setting data is input; a processing unit 12 that performs processing based on the setting data; a model information acquisition unit 13 that acquires model information representing the first model 41 after learning at the production site 100a; an environmental information acquisition unit 14 that acquires environmental information representing the setting environment; an error information acquisition unit 15 that acquires error information representing the error when the object 21 is processed using the first model 41; and a learning unit 16 that learns a second model 42 for error suppression based on the environmental information and the error information. Figure 3 In the diagram, solid arrows represent the flow of information before the second model learns, while dashed arrows represent the flow of information after the second model learns.

[0052] The input unit 11 is mainly implemented through the input unit 104 or the communication unit 106. The input unit 11 receives setting data input by the user. Among the objects 21 and 22 processed based on the setting data, object 21 is the processing object before learning the second model 42, and object 22 is the processing object after learning the second model 42. The input unit 11 is an example of an input unit that receives setting data.

[0053] The processing unit 12 is primarily implemented using a processing module for processing the processor 101 and objects 21 and 22. The processing module includes, for example, a motor for moving the tool and the workbench. Before learning the second model 42, the processing unit 12 applies the first model 41 to the setting data input to the input unit 11 and processes object 21. When the learning unit 16 learns the second model 42 through the processing of object 21, the processing unit 12 sequentially applies the first model 41 and the second model 42 to the setting data and processes object 22. Furthermore, the processing unit 12 provides the output of the first model 41 to the learning unit 16 for learning the second model 42.

[0054] The model information acquisition unit 13 is implemented by at least one of the processor 101, the input unit 104, and the communication unit 106. The model information acquisition unit 13 can read model information registered in the auxiliary storage unit 103 of the processing device 10, which is being moved from the production site 100a, or it can read model information from a recording medium such as a memory card attached to the processing device 10 during the move-out process. Furthermore, the model information acquisition unit 13 can acquire model information directly input by the user, or it can receive model information via a communication line or network.

[0055] exist Figure 4 The image shows a simple example of the model information for model 41. Figure 4 The model information is represented as follows: when the value of the set data is above zero and less than 10, the model output obtained by applying the first model 41 to the set data is the sum of the set data value plus 1; when the value of the set data is above 10 and less than 20, the sum of the set data value plus 2 is used as the model output. Furthermore, the first model 41 is not limited to the following... Figure 4 The transformation table shown can also be represented as a function model using mathematical expressions. The model information acquisition unit 13 is an example of a model information acquisition unit that acquires model information representing a first model, which is applied to setting data to suppress a first error contained in the first processing result of the processing device due to inherent characteristics of the processing device before it is installed in the factory. Here, the first processing result and the first error distinguish the processing result and error in the production site from the second processing result and the second error in the factory 100 described later.

[0056] return Figure 3 The environmental information acquisition unit 14 is implemented by at least one of the processor 101, the input unit 104, and the communication unit 106. The environmental information acquisition unit 14 acquires environmental information indicating the environment in which the processing apparatus 10 is set up when the processing unit 12 processes the object 21 using the first model 41 instead of the second model 42. The environmental information acquisition unit 14 can read environmental information from the main storage unit 102, the auxiliary storage unit 103, or a recording medium; it can also acquire environmental information directly input by the user; or it can receive environmental information from sensors measuring environmental conditions via a communication line or network. The environment represented by the environmental information can be, as described above, temperature, humidity, and material type; it can also be the electrical environment provided to the processing apparatus 10 in the factory 100, the quality of air or gas, or the output of other devices connected to the processing apparatus 10. The environmental information acquisition unit 14 is an example of an environmental information acquisition unit that acquires factory environmental information indicating the environment of the processing apparatus set up in the factory.

[0057] The error information acquisition unit 15 is implemented by at least one of the processor 101, the input unit 104, and the communication unit 106. The error information acquisition unit 15 acquires the difference between the measured value and the target value (i.e., the error) from the measuring device 30, which measures the processing result of the object 21 processed by the processing unit 12. Alternatively, the error information acquisition unit 15 can acquire the error information by acquiring both the processing result and the target value separately. That is, the error information can also be information representing both the measured value and the target value of the processing result. The error information acquisition unit 15 can acquire the error information through communication with the measuring device 30, by reading the error information from a recording medium, or by acquiring error information directly input by the user. The error information acquisition unit 15 is an example of an error information acquisition unit that acquires first error information representing a second error between the second processing result and the target value of the processing device, which processes the object in the factory using a first model based on setting data input in the factory.

[0058] The learning unit 16 is primarily implemented via the processor 101. The learning unit 16 learns a second model 42 to suppress errors based on the output of the first model 41, the error resulting from processing the object 21 using that output, and the settings environment when processing the object 21 using that output. Figure 5 The image shows a simple example of the second model 42 that has been learned. Figure 5The second model 42 represents the following cases: when the output of the first model 41 is 1 or more and less than 11, if the ambient temperature is less than 15°C, the sum obtained by adding 0.4 to the output of the first model 41 is used as the output of the second model 42; if the ambient temperature is 15°C or more, the sum obtained by adding 0.3 to the output of the first model 41 is used as the output of the second model 42. Furthermore, it represents the following cases: when the output of the first model 41 is 12 or more and less than 22, if the ambient temperature is less than 20°C, the difference obtained by subtracting 0.2 from the output of the first model 41 is used as the output of the second model 42; if the ambient temperature is 20°C or more, the difference obtained by subtracting 0.3 from the output of the first model 41 is used as the output of the second model 42. Additionally, the second model 42 is not limited to the following cases. Figure 5 The transformation table shown can also be represented as a function model using mathematical expressions. Return Figure 3 When learning the second model 42, the learning unit 16 provides the second model 42 to the processing unit 12. The learning unit 16 is equivalent to an example of a learning unit that learns the second model to suppress the second error based on environmental information and error information. The second model 42 is equivalent to an example of a model for obtaining a value corrected from the output value of the first model 41 based on the output value of the first model 41 and environmental information.

[0059] Next, refer to Figure 6 The model application processing performed by the processing device 10 will be explained. Figure 6 The model application process shown is performed as an adjustment operation or initialization process from the time the processing device 10 is set in the factory 100 until it begins normal operation.

[0060] In the model application processing, the model information acquisition unit 13 acquires model information (step S1). Then, the input unit 11 accepts the input setting data (step S2), the processing unit 12 applies the first model 41 to the setting data and processes the object 21 (step S3), the error information acquisition unit 15 acquires error information (step S4), and the environmental information acquisition unit 14 acquires environmental information (step S5).

[0061] Next, the learning unit 16 performs steps S2 to S5 to determine whether the amount of accumulated data, including the output, error information, and environmental information of the interrelated first model 41, exceeds a threshold (step S6). If it is determined that the amount of data does not exceed the threshold (step S6: No), the processing device 10 repeatedly performs the processing after step S2. Thus, data is accumulated as a combination of the output, error information, and environmental information of the first model 41.

[0062] If the data volume is determined to exceed a threshold (step S6: Yes), the learning unit 16 uses the accumulated data to learn the second model 42 (step S7). For example, after generating a temporary model representing the relationship between the output, error, and setting environment of the first model 41 through regression analysis, the learning unit 16 obtains a transformation formula for the output of the first model 41 that reduces the error as the second model 42. However, the learning unit 16's learning of the second model 42 is not limited to this; it can be supervised learning, represented by neural networks, or reinforcement learning.

[0063] Next, the input unit 11 accepts the newly input setting data (step S8), and the processing unit 12 applies the first model 41 and the second model 42 to the new setting data to process the object 22 (step S9). As a result, the second model can be applied to perform processing with smaller errors.

[0064] As explained above, the model information acquisition unit 13 acquires model information representing the first model 41, which is applied to setting data to suppress errors in the processing results of the processing device 10 caused by the inherent characteristics of the processing device 10 before it is installed in the factory 100. Furthermore, the learning unit 16 learns a second model 42 to suppress errors in the processing results of the processing device 10 that processes the object 21 in the factory 100. Thus, separate from the first model 41 which suppresses capability differences arising before it is installed in the factory 100, the second model 42 is learned to suppress capability differences arising from the environment of the processing device 10 installed in the factory 100. Therefore, by distinguishing factors that may affect the device's capability, the second model 42 is learned in the factory 100 to address capability differences caused by these factors. Therefore, the computational load for learning the model to suppress capability differences of the device can be reduced.

[0065] That is, information representing the inherent characteristics of the processing device 10 is excluded from the information used by the learning unit 16 for learning the second model 42. Therefore, useless computational processing caused by the same type of learning being performed on both the production site 100a and the factory 100 can be avoided.

[0066] Furthermore, a second model 42 for obtaining the output of the processing device 10 based on the output of the first model 41 has been described, but it is not limited thereto. For example... Figure 7 As shown, the first model 41 and the second model 42 can also be applied in parallel to the set data. After the second model 42 has been learned, the sum of the outputs of the first model 41 and the second model 42 is input to the processing unit 17 that processes the objects 21 and 22. The method of applying the second model 42 is arbitrary, as long as it is used to correct the processing of the object 21 based on the first model 41. Furthermore, in Figure 7In the example, the learning unit 16 can also collect set data for learning the second model 42.

[0067] Implementation method 2.

[0068] Next, Embodiment 2 will be described focusing on its differences from Embodiment 1 described above. Furthermore, structures that are the same as or equivalent to those in Embodiment 1 will be referred to using the same reference numerals. For example... Figure 8 As shown, the difference between this embodiment and Embodiment 1 is that the processing device 10 has the ability to learn the first model. Furthermore, the difference between this embodiment and Embodiment 1 is that the learning unit 16 further divides the setup environment into a fixed environment fixed in the factory 100 and a variable environment that may change over time, and learns the second model.

[0069] like Figure 9 As shown, the processing apparatus 10 of this embodiment includes: a characteristic information acquisition unit 18, which acquires characteristic information representing inherent characteristics of the processing apparatus 10; and a model generation unit 19, which generates a first model 41. Figure 9 In the image, thick arrows indicate the flow of information in production site 100a.

[0070] The characteristic information acquisition unit 18 is implemented by at least one of the processor 101, the input unit 104, and the communication unit 106. The characteristic information acquisition unit 18 acquires characteristic information representing the inherent characteristics of the processing device 10 in the production site 100a. The characteristic information acquisition unit 18 can read the characteristic information from the auxiliary storage unit 103 or an external recording medium, can acquire characteristic information directly input by the user, and can receive characteristic information via a communication line or network. The characteristic information, for example, represents the results of quality checks performed on multiple processing devices 10 in the production site 100a. The characteristic information acquisition unit 18 is an example of a characteristic information acquisition unit that acquires characteristic information representing the inherent characteristics of the processing device 10.

[0071] The model generation unit 19 is primarily implemented by the processor 101. The model generation unit 19 acquires the setting data input to the input unit 11, and obtains error information from the error information acquisition unit 15, which represents the error of the object 20 after processing by the processing unit 12 without using the first model 41, based on the setting data. This error information is equivalent to an example of the second error information representing the first error mentioned above. Furthermore, the model generation unit 19 acquires environmental information representing the production environment of the production site of the production processing unit 10 from the environmental information acquisition unit 14, and characteristic information representing the characteristics of the processing unit 10 itself from the characteristic information acquisition unit 18. Then, based on this acquired information, the model generation unit 19 generates the first model 41 through learning and provides it to the model information acquisition unit 13. For example, the model generation unit 19 generates the first model 41 for obtaining setting data that minimizes the error by using the setting data as the target variable and the error information, environmental information, and characteristic information as explanatory variables through regression analysis or supervised learning. Furthermore, when the correction values ​​for the set data corresponding to the environmental conditions are predetermined statistically, the model generation unit 19 can also use the characteristic information to learn a first model 41 to reduce such deviations when deviations occur due to the characteristics of the processing device. Additionally, the model generation unit 19 can also generate the first model 41 through regression analysis or supervised learning, using error information as the target variable and set data, environmental information, and characteristic information as explanatory variables. The model generation unit 19 is an example of a model generation unit that generates a first model based on characteristic information, environmental information of the production environment, and second error information.

[0072] Furthermore, the method by which the model generation unit 19 generates the first model 41 can be arbitrarily changed. For example, one or both of the characteristic information and environmental information can be omitted from the information used to generate the first model 41. Even if one or both of the characteristic information and environmental information are omitted, the first model 41 generated by the model generation unit 19 ultimately becomes a model used to suppress errors caused by the characteristics of the processing device 10. In addition, the model generation unit 19 can also generate the first model 41 by applying characteristic information and environmental information to a template model provided from an external source at the production site 100a.

[0073] Next, the first model generation process executed by the processing device 10 at the production site 100a and the model application process executed at the factory 100 will be described in turn.

[0074] In the first model generation process, such as Figure 10As shown, the characteristic information acquisition unit 18 acquires characteristic information (step S11). Then, the input unit 11 accepts the input setting data (step S12), the processing unit 12 processes the object 20 according to the setting data (step S13), the error information acquisition unit 15 acquires error information (step S14), and the environmental information acquisition unit 14 acquires environmental information of the production environment (step S15). The environmental information of the production environment can be the same type of information as the setting environment, or it can be different information. The environmental information of the production environment only needs to be information about environmental factors that affect the quality of the processing device 10. The environmental information of the production environment is equivalent to an example of production environment information.

[0075] Next, the model generation unit 19 determines whether the amount of data obtained in steps S12 to S15 exceeds a predetermined threshold (step S16). If it is determined that the amount of data does not exceed the threshold (step S16: No), the processing device 10 repeatedly performs the processing after step S12. As a result, the data required to generate the first model 41 is accumulated.

[0076] If the data volume is determined to exceed a threshold (step S16: Yes), the model generation unit 19 generates a first model 41 by learning based on the information obtained in steps S11 to S15 (step S17). Next, the processing unit 12 applies the generated first model 41 to the set data obtained in step S12 and processes the object 21 (step S18), and the error information acquisition unit 15 acquires error information (step S19). Then, the model generation unit 19 determines whether the error when processing the object 21 by applying the first model 41 to the set data is within a predetermined range (step S20).

[0077] If the error is determined to be outside the acceptable range (step S20: No), the process returns to step S12, and the addition of data based on steps S12-S16 and the generation of the first model based on steps S17-S19 are repeated. Thus, the model generation unit 19 continues learning the first model 41. Alternatively, during repeated learning, the first model 41 can be learned based on newly collected data without using previously collected data. Furthermore, in the learning of the first model 41 in step S17, if predetermined processing such as weight updates of each layer in deep learning is repeatedly rounded up, and the determination in step S20 is negative (step S20: No), the process can return to step S17 without adding new data, and the learning process can continue. If the error is determined to be within the acceptable range in step S20 (step S20: Yes), the processing device 10 ends the first model generation process.

[0078] like Figure 11As shown, in the model application processing of this embodiment, after performing step S1 in the same manner as in Embodiment 1, the environmental information acquisition unit 14 acquires environmental information representing the fixed environment of the installation environment (step S21). The fixed environment is the environment that is fixed after the installation of the processing device 10 or the adjustment during installation. Specifically, the fixed environment is the type of installation component selected based on space, whether or not A / D conversion is performed for voltages applied from the outside.

[0079] Next, after performing steps S2 to S4 in the same manner as in Embodiment 1, the processing device 10 obtains environmental information representing the changing environment of the setting environment (step S22). The changing environment is the environment that may change each time the objects 21 and 22 are processed, such as temperature or humidity. Then, the learning unit 16 determines whether the amount of data exceeds a threshold (step S6). If it is determined that the amount of data does not exceed the threshold (step S6: No), the processing after step S2 is repeated.

[0080] On the other hand, if it is determined that the amount of data exceeds the threshold (step S6: Yes), the learning unit 16 learns the second model 42 based on the fixed environment represented by the environmental information obtained in step S11 (step S23). Here, the learning unit 16 learns the second model 42 without considering the changing environment represented by the environmental information obtained in step S22. That is, the learning unit 16 learns the second model 42 without including the changing environment as a parameter, or learns the second model 42 with the parameters of the changing environment fixed, regardless of the information obtained in step S22 (step S23).

[0081] Then, the learning unit 16 determines whether the error when the second model 42 learned in step S23 is applied to the set data is within a predetermined first range (step S24). The learning unit 16 can obtain this error by applying it to the newly input set data, or by cross-validating the data accumulated in the repeatedly executed steps S2 to S4 and S22.

[0082] If the error is determined to be outside the first range (step S24: No), return to step S2 and repeat the data addition based on the re-execution of steps S2-S4, S22, and S6, and the learning of the second model 42 in step S23. Alternatively, during the learning of the second model 42 in step S23, if the predetermined processing iterations are rounded up, and the determination in step S24 is negative (step S24: No), the process can return to step S23 without adding new data collection and continue the learning iterations. If the error is determined to be within the first range in step S24 (step S24: Yes), the learning unit 16 learns the second model 42 based on the changing environment represented by the environmental information obtained in step S22 (step S25).

[0083] Then, the learning unit 16 determines whether the error when applying the second model 42 learned in step S25 to the set data is within a predetermined second range (step S26). The second range is defined as a range larger than the first range. Since there is no change for the fixed environment, the first range is defined as a relatively narrow range so that a second model 42 that fits better for the fixed environment can be obtained in steps S23-S24. In contrast, considering changes in the changing environment, the second range is defined as a relatively wide range.

[0084] If the error is determined to be outside the second range (step S26: No), return to step S2 and re-execute the data addition based on the re-execution of steps S2~S4, S22, and S6, and the learning of the second model 42 in steps S23 and S25. Alternatively, during the learning of the second model 42 in step S25, if the predetermined processing iterations are rounded up, and the determination in step S26 is negative (step S26: No), it is also possible to return to step S25 without adding more data and continue the iterations in the learning process.

[0085] If the error is determined to be within the second range in step S26 (step S26: Yes), the learning unit 16 determines whether the error has shrunk (step S27). Specifically, the learning unit 16 determines whether the magnitude of the error determined to be within the second range in step S26 has shrunk from the magnitude of the error determined to be within the first range in step S24.

[0086] If the error is determined not to have shrunk (step S27: No), return to step S2 and re-execute the data addition based on the re-execution of steps S2~S4, S22, and S6, and the learning of the second model 42 in steps S23 and S25. Alternatively, during the learning of the second model 42 in steps S23 and S25, if the predetermined processing iterations are rounded up, and the determination in step S27 is negative (step S27: No), the process can return to step S23 without adding more data and continue the learning iterations. If the error is determined to have shrunk in step S27 (step S27: Yes), the model application processing ends.

[0087] As explained above, after performing learning based on a fixed environment, the learning unit 16 performs learning based on a changing environment. Therefore, it further distinguishes factors that may affect the capabilities of the processing device 10, and sequentially performs learning of models that absorb capability deficiencies caused by these factors. This further reduces the computational load caused by model learning.

[0088] Implementation method 3.

[0089] Next, Embodiment 3 will be described focusing on its differences from Embodiment 1 described above. Furthermore, structures that are the same as or equivalent to those in Embodiment 1 described above will be represented by the same reference numerals. For example... Figure 12 As shown, the difference between this embodiment and embodiment 1 is that the information used in learning is stored in advance and the information is used when processing new objects.

[0090] The processing apparatus 10 of this embodiment includes: a storage unit 110 that stores information used by the learning unit 16; a providing unit 111 that provides information from the storage unit 110; and a determining unit 112 that determines determining data corresponding to the setting data that should be newly input based on the information from the storage unit 110.

[0091] The storage unit 110 is primarily implemented through the auxiliary storage unit 103. The storage unit 110 repeatedly acquires the setting data input to the input unit 11 and stores it in conjunction with the information used by the learning unit 16. Specifically, the storage unit 110 stores information such as… Figure 13 The recipe library shown.

[0092] The formula library is a database that stores records that correlate setting data, target values ​​when the setting data is input, environmental information indicating the environment in which objects 21 and 22 are processed based on the setting data, the first model and the second model applied to the setting data, and error information. The target value in the formula library may be included as part of the error information, or it may be input to the input unit 11 as part of the setting data or as data different from the setting data. The storage unit 110 is equivalent to an example of an accumulation unit, which accumulates multiple sets of data as accumulation data, respectively correlated with the target value, and the third processing result of the processing device that processes the object by applying the first model and the second model to the setting data, and the third error of the target value.

[0093] The providing unit 111 is implemented by at least one of the processor 101 and the output unit 105. The providing unit 111 can also read the recipe library from the storage unit 110 according to a user's request and provide the recipe library to the user. The user can also refer to the provided recipe library to determine the setting data corresponding to the user's desired target value. Furthermore, if the setting data corresponding to the user's desired target value is associated with error information indicating a large error, the user can input the setting data into the input unit 11 after making slight changes. Also, if the record containing the user's desired target value is not included in the recipe library, the user can refer to multiple records to deduce the appropriate setting data. In addition, the providing unit 111 can also provide recipe library information to the determining unit 112.

[0094] The determination unit 112 is primarily implemented by the processor 101. When a target value is specified by the user via the input unit 11, the determination unit 112 determines determination data corresponding to the specified target value based on the recipe library. Specifically, the determination unit 112 can also determine determination data as the setting data associated with the user-specified target value in the recipe library. Furthermore, when the setting data associated with the user-specified target value in the recipe library is also associated with error information indicating a large error, or when records including the user-specified target value are not included in the recipe library, the determination unit 112 can also determine determination data corresponding to the setting data that should be input corresponding to the target value using an estimation method such as linear interpolation of multiple records. The determination unit 112 is an example of a determination unit that determines determination data corresponding to the specified target value based on multiple setting data and the third processing result of a processing device that processes the object by applying a first model and a second model to each setting data.

[0095] As explained above, using a formula library makes determining new setting data easier. Input unit 11 is an example of an input unit that accepts input of new target values ​​when processing a new object, and determination unit 112 is an example of a determination unit that determines setting data corresponding to the new target value. The processing device processes the new object based on the results of applying the first model and the second model to the determination data determined corresponding to the new target value.

[0096] Implementation method 4.

[0097] Next, Embodiment 4 will be described focusing on its differences from Embodiment 3 described above. Furthermore, the same reference numerals will be used for structures that are the same as or equivalent to those in Embodiment 3 described above. The difference between this embodiment and Embodiment 3 is that, as... Figure 14 As shown, the determination unit 112 generates an estimation model for estimating the set data.

[0098] The determination unit 112 in this embodiment includes: an estimation model generation unit 1121, which generates an estimation model 43 for estimating appropriate set data based on a target value; and a target value acquisition unit 1122, which acquires the target value input to the input unit 11. Figure 14 In the diagram, a thicker dashed line illustrates the flow of information when using the estimation model.

[0099] The estimated model generation unit 1121 reads out as follows Figure 13 The formula library shown is used to generate an estimation model 43. Here, as... Figure 15 As shown, estimation model 43 is the following model: Based on the known target values ​​contained in the formula library, set data for achieving the target value with a small error is determined as the determinant data, and for unknown target values, determinant data for achieving the target value with a small error is also determined. Here, a small error refers to an error smaller than a predetermined threshold.

[0100] For example, even if the machining time in a cutting process is set to twice its original length, the cutting amount depends on the shape of the cutting tool and therefore may not necessarily be doubled. Thus, when the relationship between the set data and the target value is not obvious to the user, an estimation model can be used. Figure 16 A simple example of an estimation model is shown. The estimation model generation unit 1121 generates, for example, an estimation model for obtaining the set data that minimizes the error with the target value by performing regression analysis or supervised learning with the set data as the target variable and the target value, environmental information, the first model 41, the second model, and error information as explanatory variables.

[0101] The determination unit 112 applies the estimation model 43 to the target value obtained by the target value acquisition unit 1122 to determine the set data used to achieve the target value, and inputs it into the input unit 11. If the user inputs a target value without being aware of the selection of set data, the processing of the object 23 that achieves the target value is performed.

[0102] The estimation model is an example of a model that estimates the third error between the third processing result of the processing device that processes the object using the first and second models and the target value, and is equivalent to the set data corresponding to the specified target value. The determination unit 112 is an example of a determination unit that generates an estimation model based on accumulated data, applies the estimation model to a new target value, and thereby determines the set data corresponding to the new target value.

[0103] Next, refer to Figures 17-18 The estimation model application process performed by the processing device 10 will be described. This estimation model application process, in such cases... Figure 6 The model application processing shown is executed after the model 42 has been learned. That is, the estimation model application processing is executed after the model 42 has been learned.

[0104] exist Figure 17 In the estimated model application process shown, the determination unit 112 performs the estimated model generation process (step S41). In the estimated model generation process, as... Figure 18 As shown, the determining unit 112 fixes the first model 41 and the second model 42 (step S411), and reads data from the recipe library of the storage unit 110 (step S412). Here, fixing the first model 41 and the second model 42 in step S411 means that neither the first model 41 nor the second model 42 is newly learned. Therefore, even if different first models 41 or second models 42 are registered in the recipe library according to conditions such as setting data or environmental information, records representing different first models 41 and second models 42 can be read in step S412.

[0105] Next, the estimation model generation unit 1121 generates an estimation model 43 by learning based on the data read in step S412 (step S413). Then, the estimation model generation unit 1121 determines whether the error when applying the estimation model 43 generated in step S413 to the new target value is within a predetermined range (step S414). The estimation model generation unit 1121 can obtain the error in step S414 using the target value newly input by the user, or it can obtain it through cross-validation of the data read in step S412.

[0106] If the error is determined to be outside the range (step S414: No), the estimation model generation unit 1121 repeatedly performs the processing after step S413 to continue learning the estimation model. Specifically, in the learning of the estimation model 43 in the previous step S413, the estimation model generation unit 1121 starts repeating the predetermined processing iterations from the point when the iterations are rounded up. Furthermore, if not all data is read from the formula library in step S412, and the determination in step S414 is negative (step S414: No), the process returns to step S412, and the determination unit 112 can also read new data. If the error is determined to be within the range in step S414 (step S414: Yes), the processing of the processing device 10 returns from the estimation model generation process. Figure 17 The estimation model is applied for processing.

[0107] return Figure 17 Next, in the estimation model generation process of step S41, the target value acquisition unit 1122 acquires a new target value input to the input unit 11 (step S42), and the determination unit 112 applies the estimation model 43 to the new target value to determine the determination data (step S43).

[0108] Next, the processing unit 12 processes the object 23 using the first model 41 and the second model 42 based on the determined data (step S44). Then, a new record of the processing of the object 23 is registered in the formula library of the storage unit 110 (step S45). The registration of the record can be performed by the learning unit 16 or by the processing unit 12.

[0109] As explained above, according to the determination unit 112, the user can perform processing on the object 23 based on appropriate determination data simply by inputting a target value. Furthermore, by enriching the recipe library in the storage unit 110 by adding information, it is expected that the determination unit 112 can determine more appropriate determination data.

[0110] Implementation method 5.

[0111] Next, Embodiment 5 will be described focusing on its differences from Embodiment 4 described above. Furthermore, the same reference numerals will be used for structures that are the same as or equivalent to those in Embodiment 4. In this embodiment, the learning of the second model 42 and the learning of the estimation model 43 are performed alternately and repeatedly.

[0112] exist Figure 19 The diagram illustrates the flow of model improvement processing performed by the processing apparatus 10 of this embodiment. In the model improvement processing, the following are performed: Figure 6 The model shown is processed (step S51). Thus, the second model 42 is initialized.

[0113] Next, data is stored in the formula library (step S52). Specifically, based on the newly input setting data, the processing of object 22 using the second model 42 is performed multiple times, and the information of object 22 being processed is stored in the storage unit 110.

[0114] Next, execute as follows Figure 17 The estimated model shown is then processed (step S53). This initializes the estimated model 43.

[0115] Next, data is stored in the formula library (step S54). Specifically, by applying estimation model 43 to the newly input target value, certain data is determined, and based on the determined data, the processing of object 23 using the second model 42 is performed multiple times. Then, the information of object 23 processed is stored in storage unit 110.

[0116] Next, the learning unit 16 learns the second model 42 based on the data accumulated in step S54 (step S55). This improves the second model 42. Then, the estimation model generation unit 1121 generates an estimation model 43 (step S56). This improves the estimation model 43. Then, the processing device 10 repeatedly performs the processing after step S54.

[0117] Therefore, as Figure 20 As illustrated, the search seeks a combination of the second model 42 and the estimated model 43 that minimizes the error with a specific target value. Figure 20 In the diagram, point P2 represents the combination of the initial values ​​of the second model 42 and the estimated model 43. This point P2 is moved by improvements to the second model 42, as indicated by the arrow along the horizontal axis, and then by improvements to the estimated model 43, as indicated by the arrow along the vertical axis. By repeatedly performing such movements, it is expected that the second model 42 and the estimated model 43 will approach the point P3 corresponding to the combination with the smallest error.

[0118] The embodiments of this disclosure have been described above; however, this disclosure is not limited to the embodiments described above.

[0119] For example, an example where the processing system 1000 is the same as the processing device 10 has been described, but it is not limited thereto. For example, such as Figure 21 As shown, the processing system 1000 can also be composed of a terminal 50 having an input unit 11, a processing unit 12, a model information acquisition unit 13, an environmental information acquisition unit 14, an error information acquisition unit 15, and a learning unit 16, and a processing device 10 serving as a machine tool. Here, the terminal 50 is a UI (User Interface) terminal for operating the processing device 10, such as an industrial PC (Personal Computer). Figure 21In the example, the processing device 10 also processes the object based on the results of applying the first model 41 and the second model 42 to the set data.

[0120] In addition, such as Figure 22 As shown, the formula library can also be stored in an external storage device 110a, instead of in the storage unit 110 of the processing device 10.

[0121] The above-described embodiments can also be combined in any way. For example, it could also be... Figure 21 The terminal 50 shown has a feature information acquisition unit 18 and a model generation unit 19 as in Embodiment 2, and a storage unit 110, a provisioning unit 111 and a determination unit 112 as in Embodiments 3 and 4.

[0122] Furthermore, examples of how Model 41 and Model 42 obtain correction values ​​for set data based on set data have been described, but the examples are not limited to these. For example, such as... Figure 23 As shown, in a simple process where a control output value different from the set data is provided to the machining unit 17 based on the set data, the first model 41 can also output a control output value more appropriate than the simple process based on the set data. Here, the set data is, for example, the tool's movement speed during cutting, and the control output value is the current flowing into the motors of the spindle that rotates the tool and the moving axis that moves the worktable. The second model 42 can obtain the output value of the first model 41 and output a correction value of that output value, or it can obtain the set data and output a correction value of the output of the first model 41. Figure 23 In the example, the model generation unit 19, which generates the first model 41, can utilize learning data, including the history of control output values. Similarly, the learning unit 16, which learns the second model 42, can also utilize learning data, including the history of control output values.

[0123] The first model 41 and the second model 42 have inputs and outputs that include a process that starts with setting data input by the user and ends with the processing of an object based on that setting data. The processing device 10 processes the object based on the results of applying the first model 41 and the second model 42 to the setting data.

[0124] The functions of the processing system 1000 described above can be implemented using dedicated hardware, or they can be implemented using a conventional computer system.

[0125] For example, by storing program P1 on a computer-readable recording medium such as floppy disk, CD-ROM (Compact Disk Read-Only Memory), DVD (Digital Versatile Disk), or MO (Magneto-Optical disk), and distributing it, and by installing program P1 on a computer, an apparatus for performing the above-mentioned processing can be constructed.

[0126] Alternatively, program P1 can be pre-stored on a disk device of a server device on a communication network such as the Internet, and then downloaded to a computer, for example, by superimposing it with a carrier wave.

[0127] Furthermore, the above processing can also be achieved by simultaneously executing the startup via a network transfer program P1, such as the Internet.

[0128] Furthermore, the aforementioned processing can be achieved by executing all or part of program P1 on a server device, while the computer is sending and receiving information related to the processing via a communication network.

[0129] In addition, when the above functions are implemented by the OS (Operating System) or through the cooperation between the OS and the application, only the parts other than the OS can be stored on the medium for distribution. Alternatively, they can be downloaded to a computer.

[0130] Furthermore, the unit that implements the functions of the processing system 1000 is not limited to software; it may also implement part or all of these functions through dedicated hardware or circuitry.

[0131] This disclosure allows for various implementations and modifications without departing from its broad spirit and scope. Furthermore, the above-described embodiments are illustrative of this disclosure and do not limit its scope. That is, the scope of this disclosure is shown through the claims, not through the embodiments. Moreover, various modifications implemented within the scope of the claims and their equivalents are considered to be within the scope of this disclosure.

[0132] Industrial availability

[0133] This disclosure is suitable for reducing the performance deviation of devices installed at the FA site.

[0134] Label Explanation

[0135] 10, 10a, 10b: Processing unit; 11: Input unit; 12: Processing unit; 13: Model information acquisition unit; 14: Environmental information acquisition unit; 15: Error information acquisition unit; 16: Learning unit; 17: Processing unit; 18: Characteristic information acquisition unit; 19: Model generation unit; 20-23: Object; 30: Measuring device; 41: First model; 42: Second model; 43: Estimation model; 50: Terminal; 100: Factory; 100a: Production site; 101: Processor; 102: Main storage unit; 103: Auxiliary storage unit; 104: Input unit; 105: Output unit; 106: Communication unit; 107: Internal bus; 110: Storage unit; 110a: Storage device; 111: Provision unit; 112: Determination unit; 1000: Processing system; 1121: Estimation model generation unit; 1122: Target value acquisition unit; P1: Program; P2, P3: Points.

Claims

1. A processing system, wherein, This processing system has: Input unit, into which set data is input; A processing device, located in a factory, processes objects according to the set data; A model information acquisition unit acquires model information representing a first model, which is applied to the setting data to suppress a first error contained in the first processing result of the processing device due to the inherent characteristics of the processing device before it is set to the factory. The first processing result is generated as follows: the processing device processes the object according to the setting data input before it is set to the factory to generate the first processing result. An environmental information acquisition unit acquires factory environmental information representing the environment of the processing device located in the factory; An error information acquisition unit acquires first error information, which represents a second error. The second error is the error between the second processing result of the processing device and the target value. The second processing result is generated as follows: the first model is applied to the set data input in the factory, and the object is processed in the factory according to the output of the first model to generate the second processing result. as well as The learning unit learns a second model based on the factory environment information and the first error information to suppress the second error related to the factory environment information. The processing device processes the object based on the results of applying the first model and the second model to the set data.

2. The processing system according to claim 1, wherein, Information representing the inherent characteristics is excluded from the information used for learning the second model by the learning unit.

3. The processing system according to claim 1, wherein, The processing system further includes a determining unit, which determines determining data corresponding to the specified target value based on a plurality of the specified data and a third processing result of the processing device that processes the object by applying the first model and the second model to each of the specified data. The input unit accepts the input of new target values ​​when processing new objects. The determining unit determines the determined data, which is equivalent to the set data corresponding to the new target value. The processing device processes the new object based on the results of applying the first model and the second model to the determined data corresponding to the new target value.

4. The processing system according to claim 2, wherein, The processing system further includes a determining unit, which determines determining data corresponding to the specified target value based on a plurality of the specified data and a third processing result of the processing device that processes the object by applying the first model and the second model to each of the specified data. The input unit accepts the input of new target values ​​when processing new objects. The determining unit determines the determined data, which is equivalent to the set data corresponding to the new target value. The processing device processes the new object based on the results of applying the first model and the second model to the determined data corresponding to the new target value.

5. The processing system according to claim 3, wherein, The processing system also includes an accumulation unit, which accumulates multiple sets of preset data as accumulated data, each associated with the target value and the third error between the third processing result and the target value. The determining unit generates an estimation model of the determined data that suppresses the third error and is equivalent to the set data corresponding to the specified target value based on the accumulated data, and applies the estimation model to the new target value, thereby determining the determined data that is equivalent to the set data corresponding to the new target value.

6. The processing system according to claim 4, wherein, The processing system also includes an accumulation unit, which accumulates multiple sets of preset data as accumulated data, each associated with the target value and the third error between the third processing result and the target value. The determining unit generates an estimation model of the determined data that suppresses the third error and is equivalent to the set data corresponding to the specified target value based on the accumulated data, and applies the estimation model to the new target value, thereby determining the determined data that is equivalent to the set data corresponding to the new target value.

7. The processing system according to claim 5, wherein, The learning unit learns the second model and the determining unit generates the estimated model alternately and repeatedly.

8. The processing system according to claim 6, wherein, The learning unit learns the second model and the determining unit generates the estimated model alternately and repeatedly.

9. The processing system according to any one of claims 1 to 8, wherein, The first model is a model used to obtain a value by correcting the set data based on the set data.

10. The processing system according to any one of claims 1 to 8, wherein, The second model is a model used to obtain a value by correcting the output value based on the output value of the first model and the factory environment information.

11. The processing system according to claim 9, wherein, The second model is a model used to obtain a value by correcting the output value based on the output value of the first model and the factory environment information.

12. The processing system of any one of claims 1-8, wherein, The processing system also features: A characteristic information acquisition unit acquires characteristic information representing the inherent characteristics; and The model generation unit generates the first model. The environmental information acquisition unit acquires production environment information representing the environment of the production site where the processing device is manufactured. The error information acquisition unit acquires second error information representing the first error. The model generation unit generates the first model based on the characteristic information, the production environment information, and the second error information.

13. The processing system of claim 9, wherein, The processing system also features: A characteristic information acquisition unit acquires characteristic information representing the inherent characteristics; and The model generation unit generates the first model. The environmental information acquisition unit acquires production environment information representing the environment of the production site where the processing device is manufactured. The error information acquisition unit acquires second error information representing the first error. The model generation unit generates the first model based on the characteristic information, the production environment information, and the second error information.

14. The processing system according to claim 10, wherein, The processing system also features: A characteristic information acquisition unit acquires characteristic information representing the inherent characteristics; and The model generation unit generates the first model. The environmental information acquisition unit acquires production environment information representing the environment of the production site where the processing device is manufactured. The error information acquisition unit acquires second error information representing the first error. The model generation unit generates the first model based on the characteristic information, the production environment information, and the second error information.

15. The processing system according to claim 11, wherein, The processing system also features: A characteristic information acquisition unit acquires characteristic information representing the inherent characteristics; and The model generation unit generates the first model. The environmental information acquisition unit acquires production environment information representing the environment of the production site where the processing device is manufactured. The error information acquisition unit acquires second error information representing the first error. The model generation unit generates the first model based on the characteristic information, the production environment information, and the second error information.

16. A processing method, wherein, This processing method includes the following processes: The input unit accepts the set data. The model information acquisition unit acquires model information representing a first model, which is applied to the set data to suppress a first error in the first processing result of the processing device, which is installed in the factory and processes an object according to the set data, due to inherent characteristics before being installed in the factory. The first processing result is generated as follows: the processing device processes the object according to the set data input before being installed in the factory, and generates the first processing result. The environmental information acquisition unit acquires environmental information representing the environment of the processing device located in the factory. The error information acquisition unit acquires error information, which represents a second error. This second error is the error between the second processing result of the processing device and the target value. The second processing result is generated as follows: the first model is applied to the set data input in the factory; based on the output of the first model, the object is processed in the factory to generate the second processing result. The learning unit learns a second model based on the environmental information and the error information to suppress the second error related to the environmental information. The processing device processes the object based on the results of applying the first model and the second model to the set data.

17. A computer program product comprising a computer program, wherein, This program is used to make the computer perform the following processes: Accept setting data, Model information representing a first model is obtained and applied to the set data to suppress a first error in the first processing result of the processing device, which is installed in the factory and processes an object according to the set data, due to inherent characteristics before being installed in the factory. The first processing result is generated as follows: the processing device processes the object according to the set data input before being installed in the factory, and generates the first processing result. Obtain environmental information representing the environment of the processing unit located in the factory. Error information is obtained, which represents a second error. This second error is the error between the second processing result of the processing device and the target value. The second processing result is generated as follows: the first model is applied to the set data input in the factory; based on the output of the first model, the object is processed in the factory to generate the second processing result. Based on the environmental information and the error information, a second model is learned to suppress the second error related to the factory environmental information. The object is processed based on the results of applying the first model and the second model to the set data.

Images