Control device and control method

Abstract

Control device and control method. The control method has the steps of monitoring a statistic obtained by performing multivariate analysis on a plurality of parameters, extracting a predetermined number of parameters of a higher rank having a large degree of influence on a change in the statistic from among the plurality of parameters, generating a plurality of experimental patterns in accordance with an experimental design method, acquiring measurement results of a specific parameter indicating the quality of a product when one or more devices are controlled in accordance with the plurality of experimental patterns, setting new target values of the predetermined number of parameters of the higher rank for stabilizing the value of the specific parameter within a management range on the basis of the measurement results, and controlling the one or more devices in such a manner that the predetermined number of parameters of the higher rank approach the new target values.

Description

Technical Field

[0001] This disclosure relates to control devices and control methods. Background Technology

[0002] "Ghana Studies, Multivariate Statistical Process Management", [online], June 2005, [retrieved January 4, 2022], Internet<http: / / manabukano.brilliant-future.net / research / report / Report2005_MSPC.pdf> "The technology for monitoring the operation of the production line has been made public."

[0003] The operational status of the production line determines the quality of the products it produces. Therefore, based on "Ghanaian Studies," "Multivariate Statistical Process Management," [online], June 2005, [retrieved January 4, 2022], and the internet...<http: / / manabukano.brilliant-future.net / research / report / Report2005_MSPC.pdf> "The disclosed technology, when an abnormality is detected in the operation of the production line, executes a process to restore the abnormality to normal, thereby preventing the production of products with poor quality. However, in the past, the process to restore the abnormality to normal required human analysis and other effort. In other words, maintaining stable product quality production required considerable effort." Summary of the Invention

[0004] This disclosure was made in view of the above-mentioned problems, and its purpose is to provide a control device and control method that can maintain stable product production quality.

[0005] According to one example of this disclosure, a control device for controlling one or more pieces of equipment included in a production line for manufacturing products includes a monitoring unit, an extraction unit, a generation unit, an experiment execution unit, a setting unit, and a control unit. The monitoring unit monitors a statistic obtained through multivariate analysis of multiple parameters associated with the operation of one or more pieces of equipment. The extraction unit extracts a predetermined number of parameters from among the multiple parameters whose influence on the statistic's change is greater than a baseline, based on the case where the change in the statistic exceeds a baseline. The generation unit generates multiple experimental patterns with different combinations of target values ​​for the predetermined number of parameters, according to an experimental planning method. The experiment execution unit obtains measurement results for specific parameters representing product quality when one or more pieces of equipment are controlled according to each of the multiple experimental patterns. Based on the measurement results, the setting unit sets new target values ​​for the predetermined number of parameters to stabilize the value of the specific parameters within a manageable range. The control unit controls one or more pieces of equipment in a manner that brings the predetermined number of parameters close to the new target values.

[0006] According to this disclosure, when certain abnormalities occur in the operating state of one or more devices, the target value of a parameter among multiple parameters associated with the operation of the one or more devices that are candidate factors for the abnormality is changed to stabilize the product quality. As a result, the production of products with stable quality is maintained.

[0007] In the above disclosure, the preferred experimental design method is one that uses orthogonal arrays. This improves the efficiency of experiments using multiple experimental modes.

[0008] In the above disclosure, methods for multivariate analysis include, for example, PCA, PLS, MT method, and T method.

[0009] In the above disclosure, the setting unit performs variance analysis on the measurement results, calculates the contribution rate of each of the specified number of parameters, selects one or more object parameters from the specified number of parameters as the setting object of the target value based on the contribution rate, and sets a new target value for one or more object parameters.

[0010] Based on the above disclosure, parameters that significantly contribute to changes in specific parameters are identified as object parameters. Therefore, by setting new target values ​​for object parameters, it is easier to stabilize the values ​​of specific parameters within a manageable range.

[0011] In the aforementioned disclosure, the generation unit generates multiple experimental patterns by using the values ​​of a predetermined number of parameters as arbitrary levels among the first, second, and third levels. The second level is the current target value. The first level is less than the second level. The third level is greater than the second level. The first and third levels are determined such that the values ​​of specific parameters converge within a controlled range.

[0012] According to the above disclosure, even when experiments are conducted according to multiple experimental modes during product manufacturing, defective products can be suppressed. As a result, it is possible to continue producing products with stable quality without stopping the production line.

[0013] According to another example of this disclosure, a control method for controlling one or more pieces of equipment included in a production line for manufacturing products comprises steps 1 to 6. Step 1 is a step of monitoring a statistic, which is obtained through multivariate analysis of multiple parameters associated with the actions of one or more pieces of equipment. Step S2 is as follows: based on the condition that the variation of the statistic is greater than a baseline, a predetermined number of parameters with a high degree of influence on the variation of the statistic are extracted from among the multiple parameters. Step S3 is as follows: according to the experimental planning method, multiple experimental patterns with different combinations of target values ​​for the predetermined number of parameters are generated. Step S4 is as follows: for each of the multiple experimental patterns, measurement results of a specific parameter representing the quality of the product are obtained when one or more pieces of equipment are controlled according to that experimental pattern. Step S5 is as follows: based on the measurement results, a new target value is set for the predetermined number of parameters to stabilize the value of the specific parameter within a managed range. Step S6 is as follows: one or more pieces of equipment are controlled in a manner that brings the predetermined number of parameters close to the new target value. According to the above disclosure, it is also possible to maintain stable product production in terms of quality.

[0014] The above and other objects, features, aspects and advantages of the present invention will become clear from the following detailed description, which is understood in conjunction with the accompanying drawings and is relevant to the present invention. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating the flow of the control method in the implementation method.

[0016] Figure 2 This is a diagram showing the structure of a system using the control device described in this embodiment.

[0017] Figure 3 This is a block diagram illustrating an example of the hardware structure of the control device in this embodiment.

[0018] Figure 4 This is a block diagram illustrating an example of the functional structure of the control device in this embodiment.

[0019] Figure 5 It is a graph representing the shift in the values ​​of the parameters collected through the collection process.

[0020] Figure 6 This is a graph representing an example of a statistic obtained through multivariate analysis.

[0021] Figure 7 This is a graph illustrating the shift in statistics when the operation of one or more devices malfunctions.

[0022] Figure 8 This is an example of a factor effect diagram.

[0023] Figure 9 This is a graph showing an example of the SN ratio and proportionality constant β of the parameters calculated using the T method.

[0024] Figure 10 This is a diagram representing an example of an experimental pattern generated according to the experimental planning method.

[0025] Figure 11 This is a diagram showing an example of measurement results obtained by the experimental execution department.

[0026] Figure 12 This is a graph showing an example of the results of an analysis of variance performed by the setting department.

[0027] Figure 13 It is a factor effect diagram created by the setting department.

[0028] Figure 14 This is a diagram illustrating an example of a method for determining the direction of change of a specific parameter caused by an abnormal operation of one or more devices.

[0029] Figure 15 This is another example of a method for determining the direction of change of a specific parameter caused by an abnormality in the operation of one or more devices.

[0030] Figure 16 This is a diagram illustrating an example of the shift in a specific parameter. Detailed Implementation

[0031] Embodiments of the present invention will be described in detail with reference to the accompanying drawings. Furthermore, the same or corresponding parts in the drawings are labeled with the same reference numerals and their descriptions are not repeated.

[0032] §1 Application Examples

[0033] Reference Figure 1 An example of a scenario in which the present invention is applied will be described. Figure 1 This is a schematic diagram illustrating the flow of the control method implemented. Figure 1 The diagram illustrates a control method for controlling one or more pieces of equipment within a production line that manufactures products. Figure 1 The control method shown is executed by one or more processors. These processors can be integrated into a single control device or distributed across multiple control devices capable of communicating with each other.

[0034] like Figure 1 As shown, the control method has steps S1 to S6. Steps S1 to S6 are executed repeatedly in sequence.

[0035] Step S1 involves monitoring a statistic obtained through multivariate analysis of multiple parameters associated with the operation of one or more pieces of equipment within the production line. The statistic represents the operating status of one or more pieces of equipment. Therefore, the operating status of one or more pieces of equipment is monitored.

[0036] Step S2 is as follows: Based on the condition that the change in the statistic is greater than the baseline, extract a specified number of parameters from multiple parameters that have a significant impact on the change in the statistic. When the operating status of one or more devices experiences certain anomalies, the statistic changes. Therefore, when the change in the statistic is greater than the baseline, it means that the operating status of one or more devices has experienced certain anomalies. The specified number of parameters extracted in step S2 are listed as candidate factors for the abnormal operating status of one or more devices.

[0037] Step S3 is as follows: Following the experimental planning method, generate multiple experimental modes with distinct combinations of target values ​​for a specified number of parameters. Step S4 is as follows: For each of the multiple experimental modes, obtain the measurement results of specific parameters representing product quality when one or more devices are controlled according to that experimental mode. Through the execution of steps S3 and S4, confirm the product quality changes when the values ​​of the specified number of parameters are altered.

[0038] Step S5 is as follows: Based on the measurement results, a new target value is set for a specified number of parameters above the upper limit to stabilize a specific parameter within a specified range. Step S6 is as follows: Control one or more devices in a manner that brings the specified number of parameters above the upper limit close to the new target value. Through steps S5 and S6, the product quality is stabilized.

[0039] Thus, by executing steps S1 to S6, in the event that an anomaly occurs in the operating state of one or more devices, the target value of the parameter among the multiple parameters associated with the operation of one or more devices that are candidate factors for the anomaly is changed, so as to stabilize the product quality. As a result, product production with stable quality is maintained.

[0040] Furthermore, steps S1 to S6 can be performed during product manufacturing. Therefore, it is possible to continue producing products with consistent quality without stopping the production line.

[0041] §2Concrete examples

[0042] <System Architecture>

[0043] Figure 2 This is a diagram showing the structure of a system using the control device described in this embodiment. For example... Figure 2As shown, system 1 includes a control device 100, an HMI (Human Machine Interface) 200, and a production line 300. The control device 100 and the HMI 200 are connected in a communicable manner via an information system network 6. The control device 100 is connected in a communicable manner to one or more devices included in the production line 300 via a control system network 4.

[0044] The control device 100 is typically a PLC (Programmable Logic Controller) that controls one or more pieces of equipment included in the production line 300.

[0045] HMI 200 includes functions for providing user prompts and handling user actions. In this embodiment, HMI 200 provides the user with changes over time in statistics monitored by control device 100.

[0046] Production line 300 includes one or more pieces of equipment for producing products. Figure 2 The illustrated production line 300 produces coated metal sheets 400. Production line 300 includes a coating preparation process 310, a coating process 320, a drying process 330, and an inspection process 340.

[0047] The equipment installed in the paint preparation step 310 includes a raw material feeder 311, a mixer 312, and a paint container 313. The raw material feeder 311 is used to feed the paint raw materials (pigments, resins, additives, solvents, etc.) into a container. The mixer 312 stirs and mixes the raw materials in the container. The paint container 313 holds the paint at an appropriate temperature (paint storage temperature) to maintain the paint viscosity within a specified range. Parameters related to the operation of the equipment installed in the paint preparation step 310 include paint dilution rate, stirring speed, and paint storage temperature.

[0048] The equipment installed in the coating process 320 includes a coating device 321 and a belt conveyor 322. The coating device 321 uses compressed air to spray paint supplied from the paint holder 313 onto the metal sheet 400 conveyed by the belt conveyor 322. Parameters related to the operation of the equipment installed in the coating process 320 include the feed speed of the belt conveyor 322 to the metal sheet 400, the pressure of the compressed air, the amount of paint sprayed from the coating device 321, the distance between the coating device 321 and the metal sheet 400 (spraying distance), and the ambient temperature around the coating device 321.

[0049] The equipment installed in the drying process 330 includes a dryer 331 and a belt conveyor 332. The dryer 331 dries the coating on the metal sheet 400 conveyed by the belt conveyor 332 by heating it. Parameters associated with the operation of the equipment installed in the drying process 330 include the feed speed of the metal sheet 400 carried by the belt conveyor 332 and the drying temperature.

[0050] The values ​​of parameters associated with the operation of the equipment installed in the paint preparation process 310, the coating process 320, and the drying process 330 are collected by the control device 100. In addition, the values ​​of the parameters can be measured by sensors included in the equipment or calculated based on data (command values, etc.) output from the control device 100 to the equipment.

[0051] Inspection step 340 is the process of measuring the coating thickness of the metal sheet 400. The coating thickness can be measured by an operator or automatically using a coating thickness measuring machine. The coating thickness measured in inspection step 340 corresponds to specific parameters that indicate the quality of the products produced by production line 300.

[0052] <Hardware Structure of the Control Device>

[0053] Figure 3 This is a block diagram illustrating an example of the hardware structure of the control device in this embodiment. The control device 100 is typically a PLC (Programmable Logic Controller). For example... Figure 3 As shown, the control device 100 includes a processor 102 such as a CPU (Central Processing Unit) or MPU (Micro-Processing Unit), a chipset 104, a main memory 106, a storage device 110, a control system network controller 120, an information system network controller 122, a USB controller 124, and a memory card interface 126.

[0054] The processor 102 reads various programs stored in the memory 110, expands and executes them in the main memory 106, thereby realizing control operations for controlling the controlled object. The chipset 104 controls data transmission between the processor 102 and various components.

[0055] The storage device 110 stores a system program 112 for implementing basic processing, a user program 114 for implementing control operations, and a monitoring program 116 for monitoring the working status of one or more devices included in the production line 300.

[0056] The control system network controller 120 controls the data exchange between the control system network 4 and the equipment.

[0057] The information system network controller 122 controls the data exchange between the information system network 6 and the HMI 200, etc.

[0058] USB controller 124 controls data exchange with external devices (e.g., support devices) via a USB connection.

[0059] The memory card interface 126 is configured to be able to load and unload the memory card 128, write data to the memory card 128, and read various data (user programs, etc.) from the memory card 128.

[0060] exist Figure 3 The diagram illustrates a structural example of the processing provided by the processor 102 executing a program, but dedicated hardware circuitry (such as an ASIC or FPGA) can also be used to install some or all of this processing. Alternatively, the main part of the control device 100 can be implemented using hardware following a general architecture (such as an industrial personal computer based on a general-purpose personal computer). In this case, virtualization technology can also be used to execute multiple operating systems with different purposes in parallel, and to execute the required applications on each operating system.

[0061] <Functional Structure of Control Device>

[0062] Figure 4 This is a block diagram illustrating an example of the functional structure of the control device in this embodiment. For example... Figure 4 As shown, the control device 100 includes an I / O processing unit 10, a monitoring unit 11, an extraction unit 12, a generation unit 13, an experiment execution unit 14, and a setting unit 15. The I / O processing unit 10 is implemented by executing a user program 114 via a processor 102. The monitoring unit 11, the extraction unit 12, the generation unit 13, the experiment execution unit 14, and the setting unit 15 are implemented by executing a monitoring program 116 via a processor 102.

[0063] The IO processing unit 10 performs collection processing, control calculation processing, and output processing. Collection processing involves collecting data from one or more devices included in the production line 300. Control calculation processing is used to control the one or more devices included in the production line 300, utilizing the collected data. Output processing is the process of outputting the data obtained through control calculation processing to the one or more devices included in the production line 300.

[0064] The data collected through the collection and processing includes values ​​representing multiple parameters (coating dilution rate, stirring speed, paint storage temperature, feed speed (coating process), air pressure, room temperature, spray volume, spraying distance, feed speed (drying process), and drying temperature) related to the operation of one or more pieces of equipment included in the production line 300. Furthermore, the data collected through the collection and processing includes values ​​representing specific parameters (coating film thickness), which indicate the quality of the product produced by the production line 300.

[0065] For multiple parameters related to the operation of one or more devices included in the production line 300, target values ​​are predetermined for each. The target value for each parameter is determined in such a way that the specific parameter "coating film thickness" converges within a manageable range. The IO processing unit 10 performs control calculations to bring the multiple parameters related to the operation of one or more devices closer to their target values, and outputs the data obtained through the control calculations to one or more devices.

[0066] Figure 5 It is a graph representing the shift in the values ​​of the parameters collected through the collection process. Figure 5 Figures (a) through (k) show graphs illustrating the changes over time for the parameters "coating dilution rate," "stirring speed," "coating storage temperature," "feed speed (coating process)," "air pressure," "room temperature," "spray volume," "spraying distance," "feed speed (drying process)," "drying temperature," and "coating film thickness." In each graph, the horizontal axis represents the product ID, and the vertical axis represents the parameter value. Figure 5 In the examples shown, product IDs are represented by consecutive numbers. Additionally, each chart shows the upper and lower limits defining the scope of management.

[0067] The IO processing unit 10 manages the collected data by associating it with product IDs. Multiple devices within the production line 300 process the same product or the coating applied to it at different time points. Timing deviations (delays) in processing occur due to factors such as product transport time between processes and coating preparation time, which are pre-measured through experiments. Therefore, by managing the collected data by associating it with product IDs, multiple parameters related to the processing of the same product or the coating applied to it are linked together. Furthermore, the IO processing unit 10 can also manage the collected data using timestamps instead of product IDs.

[0068] exist Figure 5In the example shown, for products with a product ID exceeding "3500", the values ​​of four parameters—"coating dilution rate", "coating storage temperature", "room temperature", and "spraying distance"—change. Specifically, the values ​​of these four parameters increase. Subsequently, for products with a product ID around "4600", the specific parameter "coating film thickness" falls below the lower limit. Thus, before the product quality falls out of control, the operating status of one or more pieces of equipment included in the production line 300 producing this product becomes abnormal. Therefore, the control device 100 of this embodiment monitors the operating status of one or more pieces of equipment included in the production line 300 and, based on the monitoring results, sets new target values ​​for parameters associated with the operation of one or more pieces of equipment.

[0069] Monitoring Department 11 (Reference) Figure 4 The monitoring unit 11 monitors the operating status of one or more pieces of equipment within the production line 300. Specifically, the monitoring unit 11 monitors statistics obtained through multivariate analysis of multiple parameters associated with the operation of one or more pieces of equipment. The statistics represent the interrelationships between the multiple parameters. Therefore, when the interrelationships between the multiple parameters are stable, the statistics are also stable; when the interrelationships between the multiple parameters change, the statistics also change.

[0070] Multivariate analysis is a method for statistically processing multivariate data consisting of multiple explanatory variables. Monitoring unit 11 can use well-known multivariate analysis methods. For example, monitoring unit 11 monitors the statistics obtained using the MT (Mathieu Taguchi) method.

[0071] Figure 6 This is a graph representing an example of a statistic obtained through multivariate analysis. Figure 6 The diagram shows the shift in Mahalanobis distance, a statistical measure obtained by applying the MT method to 10 parameters: "coating dilution rate," "stirring speed," "coating storage temperature," "feed speed (coating process)," "air pressure," "room temperature," "spray volume," "spraying distance," "feed speed (drying process)," and "drying temperature." Mahalanobis distance represents the distance per unit space. The unit space is pre-generated based on a sample data set obtained during normal operation of the production line 300. The sample data set contains multiple sample datasets, each representing a value for one of the 10 parameters. For example... Figure 6 As shown, when one or more devices in production line 300 are operating normally, the Mahalanobis distance is distributed around 1. The monitoring unit 11 calculates the Mahalanobis distance using, for example, the method disclosed in "Tamura Kishiomi, Origin-MT Method of the Second MT System", Standardization and Quality Management, 2008, Vol. 61, No. 11.

[0072] Figure 7This is a graph illustrating the shift in statistics when the operation of one or more devices malfunctions. Figure 7 This illustrates the shift in Mahalanobis distance. For example... Figure 7 As shown, when the operation of one or more devices included in production line 300 malfunctions, the statistical value changes. Therefore, the monitoring unit 11 notifies the extraction unit 12 of the occurrence of the anomaly based on the fact that the change in the statistical value is greater than the baseline. Specifically, the monitoring unit 11 determines whether the value of the statistical value is within a predetermined range (in... Figure 7 In the example shown, the value is within the range of 4 or below. If the value of the statistic changes to outside the predetermined range, the extraction unit 12 is notified of the occurrence of the anomaly.

[0073] The monitoring unit 11 can also monitor statistics obtained using multivariate analysis other than the MT method. Examples of multivariate analysis other than the MT method include PCA (Principal Component Analysis), PLS (Partial Least Squares), and the T method (Taguchi method).

[0074] For example, monitoring unit 11 can also monitor the Q statistic obtained using PCA. The Q statistic is obtained using "Ghanaian Studies," "Multivariate Statistical Process Management," [online], June 2005, [retrieved January 4, 2022], and the internet.<http: / / manabukano.brilliant-future.net / research / report / Report2005_MSPC.pdf> "The method disclosed is used for calculation."

[0075] Alternatively, monitoring unit 11 can also monitor the predicted values ​​obtained using the T-method. The predicted value is the value of a specific parameter, "coating film thickness," predicted based on all or some of the 10 parameters: "coating dilution rate," "stirring speed," "coating storage temperature," "feed speed (coating process)," "air pressure," "room temperature," "spray volume," "spraying distance," "feed speed (drying process)," and "drying temperature." The predicted value is calculated using the method disclosed in "Tamura Kijimi, 'The 5th MT system capable of direction determination - TS method, T-method, Standardization and Quality Management, 2009, Vol. 62, No. 2'."

[0076] Extraction unit 12 (refer to) Figure 4 Based on the case where the change in the statistic is greater than the baseline, extract the specified number of parameters from among multiple parameters that have a greater impact on the change in the statistic.

[0077] For example, when using the MT method for multivariate analysis, the extraction unit 12 uses the technique disclosed in "Kiji Tamura, 'The Third State Diagnosis Based on the MT Method,' Standardization and Quality Management, 2008, Vol. 61, No. 12" to extract a specified number of parameters. Specifically, the extraction unit 12 obtains anomaly datasets representing the values ​​of 10 parameters—"coating dilution rate," "stirring speed," "coating storage temperature," "feed speed (coating process)," "air pressure," "room temperature," "spray volume," "spraying distance," "feed speed (drying process)," and "drying temperature"—that indicate a change in a statistic greater than the baseline. For each combination of parameters selected from the 10 parameters, the extraction unit 12 calculates the Mahalanobis distance of the anomaly dataset using the unit space corresponding to that combination. The unit space corresponding to each combination is pre-generated using sample data sets obtained during normal operation of production line 300. Based on the Mahalanobis distance calculated for each combination, the extraction unit 12 creates a factor effect diagram for the 10 parameters and extracts the specified number of parameters with large factor effects. The factor effect diagram represents the degree of influence of each parameter on the Mahalanobis distance.

[0078] For example, if the Mahalanobis distance calculated per unit space using a combination that includes the parameter "coating dilution rate" is larger than the Mahalanobis distance calculated per unit space using a combination that does not include the parameter "coating dilution rate", then the parameter "coating dilution rate" has a greater influence on the Mahalanobis distance.

[0079] Figure 8 This is an example of a factor-effects diagram. In Figure 8 In the diagram, the horizontal axis represents the degree of influence on Mahalanobis distance. The extraction unit 12 can extract a specified number (e.g., 7) of parameters with high influence from the factor effect diagram.

[0080] When using PCA for multivariate analysis, Extraction Unit 12 extracts a specified number of parameters with a large influence on the variation of the Q statistic based on the contribution rate of each parameter to the plotting of the Q statistic. Extraction Unit 12 uses the following resources: "Ghanaian Studies," "Multivariate Statistical Process Management," [online], June 2005, [searched January 4, 2022], and the Internet.<http: / / manabukano.brilliant-future.net / research / report / Report2005_MSPC.pdf> The disclosed calculation method calculates the contribution rate of each parameter to the plotting of the Q statistic. Specifically, the extraction unit 12 calculates the square of the difference between the average value of the parameter in the sampled data set and the value of the parameter in the outlier data set as the plotting contribution rate. The larger the plotting contribution rate, the greater the influence on the variation of the Q statistic. Therefore, the extraction unit 12 extracts a predetermined number of parameters with large plotting contribution rates.

[0081] When using the T-method for multivariate analysis, the extraction unit 12 extracts a specified number of parameters that have a significant impact on the variation of the statistic, based on the SN ratio of each parameter and the proportionality constant β.

[0082] Figure 9 This is a graph showing an example of the SN ratio and proportionality constant β of each parameter calculated using the T-method. Extraction unit 12 calculates the SN ratio and proportionality constant β of each parameter using the calculation method disclosed in "Tamura Kijimi, 'The 5th MT System Capable of Direction Determination - TS Method, T-method, Standardization and Quality Management, 2009, Vol. 62, No. 2'". The proportionality constant β represents the slope of the univariate regression. The SN ratio represents the linearity between the parameter value and the true value. The larger the SN ratio, the greater the contribution to the estimation accuracy. Therefore, extraction unit 12 extracts a predetermined number of parameters with large SN ratios. Alternatively, extraction unit 12 may use a weighting coefficient corresponding to the proportionality constant β to correct the SN ratio and extract a predetermined number of parameters with large corrected SN ratios.

[0083] Generation section 13 (refer to) Figure 4 According to the experimental design method, multiple experimental patterns are generated, each with a different combination of target values ​​for a predetermined number of parameters extracted by the extraction unit 12. The generation unit 13 uses, for example, an orthogonal array as the experimental design method. The size of the orthogonal array is determined by the number of parameters extracted by the extraction unit 12. For example, if the extraction unit 12 extracts 7 parameters, the generation unit 13 uses an L18 orthogonal array, setting the value of each parameter to one selected from three levels, thereby generating 18 experimental patterns. The three levels include a first level, a second level, and a third level. The second level is the current target value. The first level is less than the second level, and the third level is greater than the second level. The first and third levels are limited to a range that causes the specific parameter "coating film thickness" to converge within a manageable range. Therefore, even if experiments according to multiple experimental patterns are performed during product production, defective products can be suppressed. As a result, stable product production can continue without stopping the production line. Specifically, in cases where a response equivalent to nσ is required, the difference between the second level and the first level, and the difference between the third level and the second level, are calculated according to (3 / 2). 1 / 2 The value is determined by ·n·σ.

[0084] For example, when the allowable error Δ = 3σ, the generation unit 13 operates according to the principle that the first level = the second level - (3 / 2). 1 / 2 σ and the third level = the second level + (3 / 2) 1 / 2 σ is used to set the first and third levels.

[0085] Figure 10This is a diagram representing an example of an experimental pattern generated according to the experimental planning method. Figure 10 The paper shows 18 experimental modes in which the values ​​of the parameters “feed speed (drying process)”, “spraying distance”, “drying temperature”, “feed speed (coating process)”, “stirring speed”, and “coating storage temperature” were set to any of the three levels.

[0086] Experimental Execution Unit 14 (refer to) Figure 4 For each of the multiple experimental modes, the measurement results of specific parameters were obtained when one or more devices included in the production line 300 were controlled according to the experimental mode.

[0087] Specifically, the experiment execution unit 14 sequentially selects one experiment mode from the multiple experiment modes generated by the generation unit 13. The experiment execution unit 14 changes the target values ​​of each parameter according to the selected experiment mode. Consequently, the IO processing unit 10 performs control operation processing to bring the values ​​of each parameter closer to the changed target values, and outputs the data obtained through this control operation processing to one or more devices. The experiment execution unit 14 obtains the values ​​of specific parameters representing the quality of products produced by the production line 300 operating according to the changed target values. Having obtained the values ​​of specific parameters corresponding to multiple products, the experiment execution unit 14 calculates representative values ​​(e.g., average values).

[0088] Figure 11 This is a diagram illustrating an example of measurement results obtained by the experimental execution department. For example... Figure 11 As shown, the experimental execution unit 14 obtains measurement results of specific parameters for multiple experimental modes.

[0089] Setting section 15 (reference) Figure 4 Based on the measurement results obtained by the experimental execution unit 14, a new target value for the parameter is set to stabilize a specific parameter within a specified range.

[0090] Specifically, the setting unit 15 determines the object parameter of the setting object that becomes the new setting value from the predetermined number of parameters extracted by the extraction unit 12.

[0091] The setting unit 15 performs variance analysis on the measurement results of multiple experimental modes, calculating the variation S, variance V, and contribution rate p of each parameter. The setting unit 15 only needs to use the concepts and flexible application of quality engineering in the development and design phase - system evaluation and improvement without prototype / experimentation - [online], [searched January 4, 2022], and the internet.<https: / / foundry.jp / bukai / wp-content / uploads / 2012 / 07 / e4806f10b0797ec0932d9317dd92a533.pdf> The method disclosed in the document can be used to calculate the variation S, variance V, and contribution rate p.

[0092] Furthermore, the setting unit 15 calculates the average value of the specific parameter for each level for each parameter, and creates a factor effect diagram based on the calculation results.

[0093] Figure 12 This is a diagram showing an example of the results of an analysis of variance performed by the setting department. Figure 13 It is a factor effect diagram created by the setting department.

[0094] The contribution rate p of each parameter is the proportion of the change S of that parameter relative to the total change S of all parameters. For example... Figure 12 , 13 As shown, the larger the contribution rate p, the greater the influence of a parameter on a specific parameter when the target value is changed. Therefore, the setting unit 15 determines parameters whose contribution rate p exceeds a predetermined threshold, or parameters with a large contribution rate p, as target parameters. For example, in Figure 12 In the example shown, the setting unit 15 determines the parameters "spraying distance", "feed speed (coating process)" and "paint storage temperature" that have a contribution rate of more than 15% as target parameters.

[0095] exist Figure 13 In the horizontal axis, "1", "2", and "3" represent the 1st level, 2nd level, and 3rd level, respectively. For example... Figure 13 As shown in the factor effect diagram, we can understand the direction of change (either positive or negative) of the value of a specific parameter when the target value of each parameter is changed from level 2 to level 1 and level 3.

[0096] Next, the setting unit 15 determines the direction of change of a specific parameter caused by an abnormal operation of one or more devices included in the production line 300.

[0097] Figure 14This diagram illustrates an example of a method for determining the direction of change of a specific parameter caused by malfunctions in one or more devices. The setting unit 15 monitors the moving average value of the specific parameter "coating film thickness" and determines the direction of change of the specific parameter based on the change in the moving average value. Figure 14 In the example shown, compared to the first interval (the interval of product IDs "572 to 1120") before the operation of one or more devices malfunctioned, the average value of a specific parameter in the second interval (the interval of product IDs "2216 to 2764") after the operation of one or more devices malfunctioned decreased. Therefore, the setting unit 15 determined the direction of change of the specific parameter caused by the operation malfunction of one or more devices as "negative direction".

[0098] The setting unit 15 preferably uses a T-test to verify whether the difference between the average values ​​of the first interval and the second interval is significant. If the verification result of the T-test is significant, the setting unit 15 can determine the direction of change of a specific parameter caused by the abnormal operation of one or more devices.

[0099] Furthermore, there exists a situation where, at the precise timing of an abnormal operation of one or more devices included in the production line 300, the value of a specific parameter does not show any change. In such cases, the setting unit 15 cannot determine the direction of change of the specific parameter caused by the abnormal operation of one or more devices. Therefore, the setting unit 15 monitors the moving average of the specific parameter until it can determine the direction of change of the specific parameter caused by the abnormal operation of one or more devices.

[0100] Figure 15 This is another example of a method for determining the direction of change of a specific parameter caused by an abnormal operation of one or more devices. Figure 15 This represents the shift of the moving average of the predicted value of the specific parameter "coating film thickness" using the T-test. When using the T-test for multivariate analysis, the setting unit 15 monitors the moving average of the predicted value of the specific parameter "coating film thickness" to determine the direction of change of the specific parameter caused by an operational anomaly of one or more devices. In other words, the setting unit 15 determines the direction of change of the specific parameter at a time T1 when the change of the moving average becomes significant using the T-test.

[0101] Furthermore, predicted values ​​can also be calculated based on parameters obtained before the drying process, such as "coating dilution rate," "stirring speed," "coating storage temperature," "feed speed (coating process)," "air pressure," "room temperature," "spray volume," and "spraying distance." This allows for earlier determination of the direction of change for specific parameters.

[0102] The setting unit 15 sets a new target value for the object parameter based on the direction and amount of change of a specific parameter.

[0103] Specifically, the setting unit 15 determines the shift direction (either towards the first level or towards the third level) of the target value of the object parameter used to change the value of a specific parameter in a direction opposite to the direction of change of the specific parameter. The setting unit 15 is based on... Figure 13 The factor effect diagram shown is used to determine the direction of displacement. Figure 13 In the example shown, by shifting the target value of the object parameter towards the first level, the value of a specific parameter changes towards the positive side. Therefore, as... Figure 14 As shown, when the direction of change of a specific parameter is negative, the setting unit 15 determines the displacement direction to be toward the first level.

[0104] Furthermore, the setting unit 15 determines the multiple N of the variation of a specific parameter relative to the standard deviation of that specific parameter. The setting unit 15 treats each parameter as linear, shifting the set value of the target parameter in the shift direction by multiplying the difference between the level corresponding to the shift direction and the second level by N. For example, the specification center of the specific parameter "coating film thickness" is 17.5 μm, and the width of the control range is 4.0 μm (=4σ). In this case, Figure 14 The variation of 0.2 μm shown corresponds to 0.2σ. That is, the setting unit 15 determines the multiplier N to be 0.2. The shift direction is towards the first level, and the difference between the first level and the second level is (3 / 2). 1 / 2 In the case of σ, the setting unit 15 will be 0.2 × (3 / 2) smaller than the second level. 1 / 2 The value of σ determines the new target value for the object parameter.

[0105] The new target value set by the setting unit 15 is reflected in the control calculation process of the IO processing unit 10. Therefore, the IO processing unit 10 performs control calculations to bring the target parameter closer to the new target value, and outputs the data obtained through the control calculations to one or more devices. As a result, the value of a specific parameter indicating product quality becomes stable.

[0106] Figure 16 This is a graph illustrating an example of the shift in a specific parameter. In Figure 16 In this context, timing T2 is a timing function where the setting unit 15 sets a new target value for the object parameters. For example... Figure 16 As shown, after timer T2, the value of a specific parameter converges to the management range.

[0107] A chart showing the shift in statistics monitored by Monitoring Department 11 (see reference) Figure 6 , Figure 7 ), representing information about the specified number of parameters extracted by the extraction unit 12 (see reference). Figure 8 ), representing information about the experimental results obtained by the experimental execution unit 14 (see Figure 11), and the factor effect diagram generated by the setting unit 15 (refer to) Figure 13 Data such as multivariate analysis can also be displayed in the HMI. This allows users to monitor the status of data collected using multivariate analysis.

[0108] §3 Notes

[0109] As described above, this embodiment includes the following disclosures.

[0110] (Structure 1)

[0111] A control device (100) controls one or more pieces of equipment (311-313, 321, 322, 331, 332) included in a production line (300) for producing products, wherein the control device (100) has:

[0112] The monitoring unit (11, 102) monitors statistics, which are obtained by performing multivariate analysis on multiple parameters associated with the operation of the one or more devices (311-313, 321, 322, 331, 332).

[0113] The extraction unit (12, 102) extracts a specified number of parameters from the plurality of parameters that have a greater influence on the change of the statistic, based on the case where the change of the statistic is greater than the baseline.

[0114] The generation unit (13, 102) generates multiple experimental modes with different combinations of target values ​​for the specified number of parameters in accordance with the experimental planning method.

[0115] The experimental execution unit (14, 102) obtains measurement results of specific parameters representing the quality of the product when one or more devices (311-313, 321, 322, 331, 332) are controlled according to the experimental mode for the plurality of experimental modes respectively.

[0116] The setting unit (15, 102), based on the measurement results, sets a new target value for a predetermined number of parameters to stabilize the value of the specific parameter within a managed range; and

[0117] The control unit (10, 102) controls one or more devices (311-313, 321, 322, 331, 332) in a manner that makes the specified number of parameters from the upper level approach the new target value.

[0118] (Structure 2)

[0119] In the control device (100) described in structure 1,

[0120] The experimental design method described above uses orthogonal arrays.

[0121] (Structure 3)

[0122] In the control device (100) described in structure 1 or 2,

[0123] The setting unit (15, 102) performs variance analysis on the measurement results, calculates the contribution rate of each of the specified number of parameters, selects one or more object parameters from the specified number of parameters as the setting object of the target value based on the contribution rate, and sets the new target value of the one or more object parameters.

[0124] (Structure 4)

[0125] In any of the control devices (100) described in structures 1 to 3,

[0126] The generation unit (13, 102) generates the plurality of experimental modes by using the values ​​of the specified number of parameters as any one of the first, second, and third levels.

[0127] The second level is the current target value.

[0128] The first level is lower than the second level.

[0129] The third level is greater than the second level.

[0130] The first level and the third level are determined such that the value of the specific parameter converges within the management range.

[0131] (Structure 5)

[0132] A control method for controlling one or more pieces of equipment (311-313, 321, 322, 331, 332) included in a production line (300) for producing products, wherein the control method comprises the following steps:

[0133] The monitoring statistics are obtained by performing multivariate analysis on multiple parameters associated with the actions of the one or more devices (311-313, 321, 322, 331, 332);

[0134] Based on the case where the change in the statistic is greater than the baseline, extract a specified number of parameters from the plurality of parameters that have a greater impact on the change in the statistic.

[0135] According to the experimental planning method, generate multiple experimental modes with different combinations of target values ​​for the specified number of parameters at the higher level;

[0136] For each of the multiple experimental modes, measurement results of specific parameters representing the quality of the product are obtained when one or more devices (311-313, 321, 322, 331, 332) are controlled according to the experimental mode.

[0137] Based on the measurement results, a new target value is set for the specified number of parameters above the management level to stabilize the value of the specific parameter within the management range; and

[0138] The more than one device (311-313, 321, 322, 331, 332) is controlled in such a way that the specified number of parameters are close to the new target value.

[0139] Embodiments of the present invention have been described, but should be considered as illustrative rather than restrictive in all respects. The scope of the invention is defined by the technical solutions of the invention and is intended to include all modifications equivalent to or within the scope of the technical solutions of the invention.

Claims

1. A control device for controlling one or more pieces of equipment included in a production line for producing products, wherein, The control device has: The monitoring unit monitors statistics, which are obtained by performing multivariate analysis on multiple parameters associated with the actions of the one or more devices. The extraction unit extracts a specified number of parameters from the plurality of parameters that have a greater impact on the change of the statistic, based on the case where the change of the statistic is greater than the baseline. The generation unit generates multiple experimental modes, each with a different combination of target values ​​for a specified number of parameters, according to the experimental planning method. The experimental execution unit obtains measurement results of specific parameters representing the quality of the product when one or more devices are controlled according to the experimental mode for each of the multiple experimental modes. The setting unit, based on the measurement results, sets a new target value for the parameter to stabilize the value of the specific parameter within a specified number of parameters that can suppress the production of defective products; and The control unit controls one or more devices in a manner that brings the predetermined number of parameters from the higher-level unit closer to the new target value. The statistic is any square of the first Mahalanobis distance from the unit space, the Q statistic obtained using principal component analysis, and the predicted value obtained using the T-method, wherein the unit space is generated based on sample data obtained during normal operation of the production line. When the statistic is the first Mahalanobis distance, the extraction unit extracts the specified number of parameters from the higher-order parameters based on the second Mahalanobis distance calculated for each combination of one or more parameters selected from the plurality of parameters. When the statistic is the Q statistic, the extraction unit extracts the specified number of parameters based on the contribution rate of each of the multiple parameters to the plotting of the Q statistic. When the statistic is the predicted value, the extraction unit extracts the specified number of parameters based on the SN ratios of the plurality of parameters calculated using the T-method. The generation unit generates the multiple experimental modes by using the values ​​of the specified number of parameters from the higher level as any one of the first, second, and third levels. The second level is the current target value. The first level is lower than the second level. The third level is greater than the second level. The first and third levels are determined such that the value of the specific parameter converges within the management range. Based on the measurement results, the setting unit determines the shift direction of the target value of a predetermined number of upper-level parameters that will cause the value of the specific parameter to change in the opposite direction to the change direction of the specific parameter, and sets a new target value for the predetermined number of upper-level parameters according to the amount of change of the specific parameter and the shift direction of the target value.

2. The control device according to claim 1, wherein, The experimental design method described above uses orthogonal arrays.

3. The control device according to claim 1 or 2, wherein, The setting unit performs variance analysis on the measurement results, calculates the contribution rate of each of the specified number of parameters, and based on the contribution rate, selects one or more object parameters from the specified number of parameters to be the setting object of the target value, and sets the new target value of the one or more object parameters.

4. A control method for controlling one or more pieces of equipment included in a production line for producing products, wherein, The control method has the following steps: Monitoring statistics, which are obtained by performing multivariate analysis on multiple parameters associated with the actions of the one or more devices; Based on the case where the change in the statistic is greater than the baseline, extract a specified number of parameters from the plurality of parameters that have a greater impact on the change in the statistic. According to the experimental planning method, generate multiple experimental modes with different combinations of target values ​​for the specified number of parameters at the higher level; For each of the multiple experimental modes, measurement results of specific parameters representing the quality of the product are obtained when one or more devices are controlled according to the experimental mode. Based on the measurement results, a new target value is set for the parameter to be stabilized within a specified number of parameters that can suppress the production of defective products; and To control one or more devices in a manner that brings the specified number of parameters from the higher-level controller close to the new target value. The statistic is any square of the first Mahalanobis distance from the unit space, the Q statistic obtained using principal component analysis, and the predicted value obtained using the T-method, wherein the unit space is generated based on sample data obtained during normal operation of the production line. The extraction step includes the following steps: If the statistic is the first Mahalanobis distance, then based on the second Mahalanobis distance calculated for each combination of one or more parameters selected from the plurality of parameters, extract the specified number of parameters from the higher-order parameters. The extraction step includes the following steps: when the statistic is the Q statistic, based on the contribution rate of each of the multiple parameters to the plotting of the Q statistic, extract the specified number of parameters at the higher level. The extraction step includes the following steps: when the statistic is the predicted value, based on the SN ratios of the multiple parameters calculated using the T-method, extract the specified number of parameters from the upper dimensional range. The generation process includes the following steps: generating the multiple experimental modes by using the values ​​of the specified number of parameters as any one of the first, second, and third levels. The second level is the current target value. The first level is lower than the second level. The third level is greater than the second level. The first and third levels are determined such that the value of the specific parameter converges within the management range. The setting steps include the following steps: based on the measurement results, determining the shift direction of the target value of the specified number of parameters above the specified level, which is used to change the value of the specific parameter in the direction opposite to the change direction of the specific parameter; and setting a new target value of the specified number of parameters above the specified level according to the amount of change of the specific parameter and the shift direction of the target value.

Images