Control device and control method

Abstract

To provide a control method capable of maintaining production of products with stable quality.SOLUTION: A control method includes steps of: monitoring statistics obtained by performing multivariate analysis on a plurality of parameters S1; extracting a predetermined number of high-rank parameters having a large degree of influence on the variation of the statistic among the plurality of parameters S2; generating a plurality of experimental patterns according to the experimental design method S3; acquiring measurement results of a specific parameter that indicates product quality when controlling one or more instruments according to each of the plurality of experimental patterns S4; setting new target values for the predetermined number of high-rank parameters for stabilizing the value of the specific parameter within a control range based on the measurement results S5; controlling one or more devices so that the predetermined number of high-rank parameters approach the new target values S6.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a control device and a control method. [Background technology]

[0002] "Manabu Kano, "Multivariate Statistical Process Control," [online], June 2005, [Retrieved January 4, 2022], Internet<http: / / manabukano.brilliant-future.net / research / report / Report2005_MSPC.pdf> (Non-Patent Document 1) discloses a technique for monitoring the operating status of a production line. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Manabu Kano, "Multivariate Statistical Process Control", [online], June 2005, [Retrieved January 4, 2022], Internet<http: / / manabukano.brilliant-future.net / research / report / Report2005_MSPC.pdf> [Non-patent document 2] Kishio Tamura, "The Origin of the 2nd MT System - MT Method," Standardization and Quality Control, 2008, Vol. 61, No. 11 [Non-patent document 3] Kishio Tamura, "5th MT System with Direction Judgment - TS Method, T Method," Standardization and Quality Control, 2009, Vol. 62, No. 2 [Non-patent document 4] Kishio Tamura, "Condition Diagnosis by the 3rd MT Method," Standardization and Quality Control, 2008, Vol. 61, No. 12 [Non-patent document 5] "Concepts and Use of Quality Engineering in the Development and Design Stage - System Evaluation and Improvement Without Prototypes or Tests -" [online] [Retrieved January 4, 2022], Internet <https: / / foundry.jp / bukai / wp-content / uploads / 2012 / 07 / e4806f10b0797ec0932d9317dd92a533.pdf> Summary of the Invention [Problem to be solved by the invention]

[0004] The operating status of a production line determines the quality of the products produced by the production line. Therefore, according to the technology disclosed in Non-Patent Document 1, when an abnormality in the operating status of the production line is detected, processing is performed to return the abnormality to normal, thereby preventing the production of poor-quality products. However, in the past, processing to return the abnormality to normal required time and effort, such as manual analysis. In other words, it was time-consuming to maintain the production of products with stable quality.

[0005] The present disclosure has been made in view of the above problems, and its purpose is to provide a control device and a control method that are capable of maintaining the production of products with stable quality. [Means for solving the problem]

[0006] According to an example of the present disclosure, a control device for controlling one or more devices included in a production line that produces 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 statistics obtained by performing multivariate analysis on multiple parameters related to the operation of one or more devices. The extraction unit extracts a predetermined number of parameters from the multiple parameters that have a significant impact on the variation of the statistics when the variation of the statistics is greater than a reference value. The generation unit generates multiple experimental patterns using an experimental design method, each of which has a different combination of target values for the predetermined number of parameters. For each of the multiple experimental patterns, the experiment execution unit obtains measurement results of specific parameters that indicate product quality when one or more devices are controlled according to the experimental pattern. The setting unit sets new target values for the predetermined number of parameters based on the measurement results, in order to stabilize the values of the specific parameters within a control range. The control unit controls the one or more devices so that the predetermined number of parameters approach the new target values.

[0007] According to this disclosure, when some abnormality occurs in the operating state of one or more devices, the target value of a parameter that is a candidate cause of the abnormality among multiple parameters related to the operation of the one or more devices is changed so that the quality of the product is stabilized, thereby maintaining the production of products with stable quality.

[0008] In the above disclosure, the experimental design is preferably a method using an orthogonal array, which increases the efficiency of experiments using multiple experimental patterns.

[0009] In the above disclosure, for example, PCA, PLS, MT method, and T method are used as multivariate analysis techniques.

[0010] In the above disclosure, the setting unit performs a variance analysis on the measurement results, calculates the contribution rate of each of the top predetermined number of parameters, selects one or more target parameters from the top predetermined number of parameters based on the contribution rate, and sets new target values for the one or more target parameters.

[0011] According to the above disclosure, a parameter that contributes significantly to the fluctuation of a specific parameter is determined as a target parameter. By setting a new target value for the target parameter, it becomes easier to stabilize the value of the specific parameter within a control range.

[0012] In the above disclosure, the generation unit generates a plurality of experimental patterns so that the values of each of a predetermined number of top parameters are one of a first level, a second level, and a third level. The second level is a current target value. The first level is smaller than the second level. The third level is larger than the second level. The first and third levels are determined so that the values of specific parameters fall within a control range.

[0013] According to the above disclosure, even if experiments based on multiple experimental patterns are performed during the production of a product, it is possible to prevent the production of defective products, thereby enabling the continuous production of products with stable quality without stopping the production line.

[0014] According to another example of the present disclosure, a control method for controlling one or more devices included in a production line that produces products includes first to sixth steps. The first step is a step of monitoring statistics obtained by performing multivariate analysis on multiple parameters related to the operation of one or more devices. The second step is a step of extracting a predetermined number of parameters from the multiple parameters that have a significant impact on the variation of the statistics, based on whether the variation of the statistics is greater than a reference value. The third step is a step of generating multiple experimental patterns, each of which has a different combination of target values for the predetermined number of parameters, according to an experimental design method. The fourth step is a step of acquiring, for each of the multiple experimental patterns, measurement results of specific parameters that indicate product quality when one or more devices are controlled according to the experimental pattern. The fifth step is a step of setting new target values for the predetermined number of parameters based on the measurement results, in order to stabilize the values of the specific parameters within a control range. The sixth step is a step of controlling one or more devices so that the predetermined number of parameters approach the new target values. The above disclosure also maintains the production of products with stable quality. [Effects of the Invention]

[0015] According to the present disclosure, production of products with stable quality is maintained. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 2 is a schematic diagram showing the flow of a control method according to an embodiment. [Figure 2] 1 is a diagram showing a configuration of a system to which a control device according to an embodiment of the present invention is applied; [Figure 3] FIG. 2 is a block diagram illustrating an example of a hardware configuration of a control device according to the present embodiment. [Figure 4] 2 is a block diagram showing an example of a functional configuration of a control device according to the present embodiment; FIG. [Figure 5] FIG. 10 is a diagram showing transitions in the values of parameters collected by the collection process. [Figure 6]FIG. 10 is a diagram showing an example of statistics obtained by performing multivariate analysis. [Figure 7] FIG. 10 is a diagram showing an example of the transition of statistics when an abnormality occurs in the operation of one or more devices. [Figure 8] FIG. 1 is a diagram showing an example of a cause-and-effect diagram. [Figure 9] FIG. 10 is a diagram showing an example of the SN ratio and proportionality constant β of each parameter calculated using the T method. [Figure 10] FIG. 10 is a diagram showing an example of an experiment pattern generated according to an experimental design method. [Figure 11] FIG. 10 is a diagram showing an example of measurement results acquired by the experiment execution unit. [Figure 12] FIG. 10 is a diagram illustrating an example of a result of an analysis of variance performed by a setting unit. [Figure 13] FIG. 10 is a factor-effect diagram created by a setting unit. [Figure 14] FIG. 10 is a diagram illustrating an example of a method for identifying the direction of fluctuation of a specific parameter caused by an abnormality in the operation of one or more devices. [Figure 15] FIG. 10 is a diagram illustrating another example of a method for identifying the direction of fluctuation of a specific parameter caused by an abnormality in the operation of one or more devices. [Figure 16] FIG. 10 is a diagram illustrating an example of a transition of a specific parameter. DETAILED DESCRIPTION OF THE INVENTION

[0017] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail with reference to the accompanying drawings, in which the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will not be repeated.

[0018] §1 Application Examples An example of a situation in which the present invention is applied will be described with reference to Fig. 1. Fig. 1 is a schematic diagram showing the flow of a control method according to an embodiment. Fig. 1 shows the flow of a control method for controlling one or more devices included in a production line that produces products. The control method shown in Fig. 1 is executed by one or more processors. The one or more processors may be incorporated into a single control device, or may be distributed and arranged in multiple control devices that can communicate with each other.

[0019] 1, the control method includes steps S1 to S6. Steps S1 to S6 are repeatedly executed in this order.

[0020] Step S1 is a step of monitoring statistics obtained by performing multivariate analysis on multiple parameters related to the operation of one or more devices included in the production line. The statistics represent the operating status of one or more devices. Therefore, the operating status of one or more devices is monitored.

[0021] Step S2 is a step of extracting a predetermined number of parameters from among the multiple parameters that have a large influence on the fluctuation of the statistical quantity, depending on whether the fluctuation of the statistical quantity is greater than the reference value. When some abnormality occurs in the operational state of one or more devices, the statistical quantity fluctuates. Therefore, when the fluctuation of the statistical quantity is greater than the reference value, it means that some abnormality has occurred in the operational state of one or more devices. The predetermined number of parameters extracted in step S2 are considered as candidate factors for the abnormality in the operational state of one or more devices.

[0022] Step S3 is a step of generating multiple experimental patterns with different combinations of target values for a predetermined number of top parameters according to an experimental design method. Step S4 is a step of acquiring, for each of the multiple experimental patterns, measurement results of specific parameters that indicate product quality when one or more devices are controlled according to the experimental pattern. By executing steps S3 and S4, fluctuations in product quality when the values of the predetermined number of top parameters are changed are confirmed.

[0023] Step S5 is a step of setting new target values for a predetermined number of top parameters based on the measurement results in order to stabilize specific parameters within a specified range. Step S6 is a step of controlling one or more devices so that the predetermined number of top parameters approach the new target values. Steps S5 and S6 stabilize the quality of the product.

[0024] In this way, by executing steps S1 to S6, when some abnormality occurs in the operating state of one or more devices, the target values of the parameters that are candidates for the cause of the abnormality among the multiple parameters related to the operation of one or more devices are changed so that the quality of the product is stabilized, thereby maintaining the production of products with stable quality.

[0025] Steps S1 to S6 can be performed during the production of products, so that the production of products with stable quality can be continued without stopping the production line.

[0026] §2 Specific examples <System configuration> Fig. 2 is a diagram showing the configuration of a system to which a control device according to this embodiment is applied. As shown in Fig. 2, 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 communicatively connected via an information network 6. The control device 100 and one or more devices included in the production line 300 are communicatively connected via a control network 4.

[0027] The control device 100 is typically a programmable logic controller (PLC), and controls one or more devices included in the production line 300.

[0028] The HMI 200 has a function of presenting information to a user and a function of accepting operations from the user. In this embodiment, the HMI 200 provides the user with time-varying statistics monitored by the control device 100.

[0029] The production line 300 includes one or more devices for producing products. The production line 300 illustrated in Figure 2 produces painted metal plates 400. The production line 300 includes a paint preparation process 310, a painting process 320, a drying process 330, and an inspection process 340.

[0030] The equipment installed in the paint preparation process 310 includes a raw material feeder 311, a mixer 312, and a paint storage container 313. The raw material feeder 311 is a device that feeds 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 storage container 313 stores the paint at an appropriate temperature (paint storage temperature) to maintain the paint viscosity within a predetermined range. Parameters related to the operation of the equipment installed in the paint preparation process 310 include the paint dilution ratio, stirring speed, and paint storage temperature.

[0031] The equipment installed in the painting process 320 includes an applicator 321 and a belt conveyor 322. The applicator 321 uses compressed air to spray paint supplied from a paint storage container 313 onto the metal plates 400 being transported by the belt conveyor 322. Parameters related to the operation of the equipment installed in the painting process 320 include the speed at which the metal plates 400 are transported by the belt conveyor 322, the pressure of the compressed air (air pressure), the amount of paint discharged from the applicator 321, the distance between the applicator 321 and the metal plates 400 (spraying distance), and the room temperature around the applicator 321.

[0032] The equipment installed in the drying process 330 includes a dryer 331 and a belt conveyor 332. The dryer 331 dries the paint on the metal plate 400 by heating the metal plate 400 being transported by the belt conveyor 332. Parameters related to the operation of the equipment installed in the drying process 330 include the transport speed of the metal plate 400 by the belt conveyor 332 and the drying temperature.

[0033] Parameter values related to the operation of the equipment installed in the paint preparation process 310, the painting process 320, and the drying process 330 are collected by the control device 100. The parameter values may be measured by sensors included in the equipment, or may be calculated based on data (such as command values) output from the control device 100 to the equipment.

[0034] The inspection process 340 is a process for measuring the thickness of the coating on the metal plate 400. The coating thickness may be measured by an operator or automatically using a coating thickness measuring device. The coating thickness measured in the inspection process 340 corresponds to a specific parameter that indicates the quality of the products produced by the production line 300.

[0035] <Control device hardware configuration> Fig. 3 is a block diagram showing an example of the hardware configuration of a control device according to this embodiment. The control device 100 is typically a PLC (Programmable Logic Controller). As shown in Fig. 3, the control device 100 includes a processor 102 such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit), a chipset 104, a main memory 106, storage 110, a control network controller 120, an information network controller 122, a USB controller 124, and a memory card interface 126.

[0036] The processor 102 reads various programs stored in the storage 110, expands them in the main memory 106, and executes them to perform control calculations for controlling the control target. The chipset 104 controls data transmission between the processor 102 and each component.

[0037] The storage 110 stores a system program 112 for implementing basic processing, a user program 114 for implementing control calculations, and a monitoring program 116 for monitoring the operating status of one or more devices included in the production line 300.

[0038] The control network controller 120 controls data exchange with devices via the control network 4 .

[0039] The information network controller 122 controls data exchange with the HMI 100 and the like via the information network 6 .

[0040] The USB controller 124 controls the transfer of data to and from external devices (eg, support devices) via a USB connection.

[0041] The memory card interface 126 is configured to allow a memory card 128 to be attached and detached, and is capable of writing data to the memory card 128 and reading various data (such as a user program) from the memory card 128.

[0042] 3 shows an example of a configuration in which the processor 102 executes a program to provide the necessary processing, but some or all of the provided processing may be implemented using dedicated hardware circuits (e.g., ASIC or FPGA). Alternatively, the main part of the control device 100 may be realized using hardware that conforms to a general-purpose architecture (e.g., an industrial PC based on a general-purpose PC). In this case, virtualization technology may be used to run multiple operating systems with different purposes in parallel, and necessary applications may be executed on each operating system.

[0043] <Controller functional configuration> Fig. 4 is a block diagram showing an example of the functional configuration of a control device according to this embodiment. As shown in Fig. 4, the control device 100 includes an IO 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 IO processing unit 10 is realized by a processor 102 executing a user program 114. The monitoring unit 11, the extraction unit 12, the generation unit 13, the experiment execution unit 14, and the setting unit 15 are realized by a processor 102 executing a monitoring program 116.

[0044] The IO processing unit 10 executes a collection process, a control arithmetic process, and an output process. The collection process is a process of collecting data from one or more devices included in the production line 300. The control arithmetic process is a process for controlling one or more devices included in the production line 300, and uses the collected data. The output process is a process of outputting data obtained by the control arithmetic process to one or more devices included in the production line 300.

[0045] The data collected by the collection process includes data indicating the values of multiple parameters (paint dilution ratio, stirring speed, paint storage temperature, feed speed (painting process), air pressure, room temperature, discharge amount, spray distance, feed speed (drying process), drying temperature) related to the operation of one or more devices included in the production line 300. Furthermore, the data collected by the collection process includes data indicating the value of a specific parameter (paint film thickness) that indicates the quality of the products produced by the production line 300.

[0046] A target value is set in advance for each of a plurality of parameters related to the operation of one or more devices included in the production line 300. The target value of each parameter is set so that a specific parameter, "paint film thickness," falls within a control range. The IO processing unit 10 executes control arithmetic processing so that each of the plurality of parameters related to the operation of one or more devices approaches the target value, and outputs data obtained by the control arithmetic processing to one or more devices.

[0047] FIG. 5 shows the transition of parameter values collected by the collection process. Graphs (a) to (k) in FIG. 5 show the time-dependent changes in the values of the parameters "paint dilution ratio," "stirring speed," "paint storage temperature," "feed speed (painting process)," "air pressure," "room temperature," "discharge rate," "spray distance," "feed speed (drying process)," "drying temperature," and "paint film thickness." In each graph, the horizontal axis represents the product ID, and the vertical axis represents the parameter value. In the example shown in FIG. 5, the product ID is represented by a serial number. Each graph also shows the upper and lower limits that define the control range.

[0048] The IO processing unit 10 manages the collected data in association with a product ID. Multiple devices included in the production line 300 perform processes on the same product or paint to be applied to the product at different times. The time lag between the processes is caused by the time it takes to transport the product between processes, the time it takes to prepare the paint, etc., and is measured in advance through experiments, etc. Therefore, by managing the collected data in association with a product ID, multiple parameters when processes are performed on the same product or paint to be applied to the product are associated with each other. The IO processing unit 10 may manage the collected data using a timestamp instead of a product ID.

[0049] In the example shown in FIG. 5, for products with a product ID greater than "3500," the values of four parameters—"paint dilution ratio," "paint storage temperature," "room temperature," and "spray distance"—fluctuate. Specifically, the values of these four parameters increase. Subsequently, for products with a product ID near "4600," the specific parameter "paint film thickness" falls below its lower limit. Thus, before the product quality falls outside the control range, an abnormality is observed in the operational status of one or more devices included in the production line 300 that produces the product. Therefore, the control device 100 according to this embodiment monitors the operational status of one or more devices included in the production line 300 and, based on the monitoring results, sets new target values for parameters related to the operation of the one or more devices.

[0050] The monitoring unit 11 (see FIG. 4) monitors the operating status of one or more devices included in the production line 300. Specifically, the monitoring unit 11 monitors statistics obtained by performing multivariate analysis on multiple parameters related to the operation of one or more devices. The statistics represent the interrelationships between the multiple parameters. Therefore, when the interrelationships between the multiple parameters are stable, the statistics are also stable, and when there is a fluctuation in the interrelationships between the multiple parameters, the statistics also fluctuate.

[0051] Multivariate analysis is a method for statistically handling multivariate data consisting of multiple explanatory variables. The monitoring unit 11 may use a known multivariate analysis method. For example, the monitoring unit 11 may monitor statistics obtained using the Mahalanobis-Taguchi (MT) method.

[0052] FIG. 6 is a diagram illustrating an example of a statistical quantity obtained by performing multivariate analysis. FIG. 6 shows the trend of the Mahalanobis distance, a statistical quantity obtained by applying the MT method to 10 parameters: "paint dilution ratio," "stirring speed," "paint storage temperature," "feed speed (painting process)," "air pressure," "room temperature," "discharge rate," "spray distance," "feed speed (drying process)," and "drying temperature." The Mahalanobis distance indicates the distance from the unit space. The unit space is generated in advance from a group of sample data obtained when the production line 300 is operating normally. Each of the multiple sample data sets included in the sample data group indicates the values of the 10 parameters. As shown in FIG. 6, when one or more pieces of equipment included in the 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 Non-Patent Document 2 (Tamura Kishio, "Origin of the 2nd MT System - MT Method," Standardization and Quality Control, 2008, Vol. 61, No. 11).

[0053] FIG. 7 is a diagram showing an example of the transition of a statistical quantity when an abnormality occurs in the operation of one or more devices. FIG. 7 shows the transition of the Mahalanobis distance. As shown in FIG. 7, when an abnormality occurs in the operation of one or more devices included in the production line 300, the statistical quantity fluctuates. Therefore, the monitoring unit 11 notifies the extraction unit 12 of the occurrence of an abnormality when the fluctuation of the statistical quantity is greater than a reference value. Specifically, the monitoring unit 11 determines whether the value of the statistical quantity is within a predetermined range (a range of 4 or less in the example shown in FIG. 7), and notifies the extraction unit 12 of the occurrence of an abnormality when the value of the statistical quantity falls outside the predetermined range.

[0054] The monitoring unit 11 may monitor statistics obtained using multivariate analysis other than the MT method, such as PCA (Principal Component Analysis), PLS (Partial Least Squares), and T method (Taguchi method).

[0055] For example, the monitoring unit 11 may monitor Q statistics obtained using PCA. The Q statistics are calculated using the method disclosed in Non-Patent Document 1.

[0056] Alternatively, the monitoring unit 11 may monitor a predicted value obtained using the T-method. The predicted value is the value of a specific parameter, "coating film thickness," predicted from all or some of the ten parameters, "coating dilution ratio," "stirring speed," "coating storage temperature," "feed speed (coating process)," "air pressure," "room temperature," "discharge rate," "spray distance," "feed speed (drying process)," and "drying temperature." The predicted value is calculated using the method disclosed in Non-Patent Document 3.

[0057] When the fluctuation of the statistics is greater than the reference value, the extraction unit 12 (see FIG. 4) extracts a predetermined number of parameters that have the highest influence on the fluctuation of the statistics from among the multiple parameters.

[0058] For example, when the MT method is used as the multivariate analysis, the extraction unit 12 extracts a predetermined number of top-ranking parameters using the technology disclosed in Non-Patent Document 4 (Tamura Kishio, "3rd State Diagnosis by MT Method," Standardization and Quality Control, 2008, Vol. 61, No. 12). Specifically, the extraction unit 12 acquires an abnormal data set indicating the values of 10 parameters (paint dilution ratio, stirring speed, paint storage temperature, feed speed (painting process)), air pressure, room temperature, discharge rate, spray distance, feed speed (drying process)), and drying temperature) when the fluctuation of the statistical quantity is determined to be greater than the reference value. For each combination of one or more parameters selected from the 10 parameters, the extraction unit 12 calculates the Mahalanobis distance of the abnormal data set using a unit space corresponding to the combination. The unit space corresponding to each combination is generated in advance using a group of sample data obtained when the production line 300 is operating normally. The extraction unit 12 creates a factorial effect diagram for the 10 parameters based on the Mahalanobis distance calculated for each combination, and extracts a predetermined number of parameters with the highest factorial effect. The factorial effect diagram shows the degree of influence of each parameter on the Mahalanobis distance.

[0059] For example, if the Mahalanobis distance calculated using the unit space corresponding to a combination that includes the parameter "paint dilution rate" is larger than the Mahalanobis distance calculated using the unit space corresponding to a combination that does not include the parameter "paint dilution rate," then the influence of the parameter "paint dilution rate" on the Mahalanobis distance is large.

[0060] Fig. 8 is a diagram showing an example of a cause-and-effect diagram. In Fig. 8, the horizontal axis represents the degree of influence on the Mahalanobis distance. The extraction unit 12 may extract a predetermined number (e.g., seven) of parameters with the highest degree of influence from the cause-and-effect diagram.

[0061] When PCA is used as the multivariate analysis, the extraction unit 12 may extract a predetermined number of parameters that have a large influence on the fluctuation of the Q statistic based on a contribution plot of each parameter to the Q statistic. The extraction unit 12 calculates the contribution plot of each parameter to the Q statistic using the calculation method disclosed in Non-Patent Document 1. Specifically, the extraction unit 12 calculates the square of the difference between the average value of the parameter in the sample data group and the value of the parameter in the abnormal data set as the contribution plot. The larger the contribution plot, the greater the influence on the fluctuation of the Q statistic. Therefore, the extraction unit 12 extracts a predetermined number of parameters that have a large contribution plot.

[0062] When the T-method is used as the multivariate analysis, the extraction unit 12 may extract a predetermined number of parameters that have the highest degree of influence on the fluctuation of the statistics based on the S / N ratio and proportionality constant β of each parameter.

[0063] FIG. 9 is a diagram showing an example of the SN ratio and proportionality constant β of each parameter calculated using the T-method. The extraction unit 12 calculates the SN ratio and proportionality constant β of each parameter using the calculation method disclosed in Non-Patent Document 3. The proportionality constant β indicates the slope of simple regression. The SN ratio represents the linearity between the parameter value and the true value. A parameter with a larger SN ratio value indicates a higher contribution to estimation accuracy. Therefore, the extraction unit 12 extracts a predetermined number of parameters with the highest SN ratios. Alternatively, the extraction unit 12 may correct the SN ratio using a weighting coefficient corresponding to the proportionality constant β, and extract a predetermined number of parameters with the highest SN ratios after the correction.

[0064] The generation unit 13 (see FIG. 4 ) generates multiple experimental patterns, each with a different combination of target values for a predetermined number of top parameters extracted by the extraction unit 12, according to an experimental design. The generation unit 13 uses, for example, an orthogonal array as the experimental design. 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 seven parameters, the generation unit 13 generates 18 experimental patterns by using an L18 orthogonal array and setting the value of each parameter to one of three levels. 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 smaller than the second level, and the third level is larger than the second level. The first and third levels are limited to a range in which the specific parameter “paint film thickness” falls within the control range. This prevents the production of defective products, even if experiments based on multiple experimental patterns are performed during product production. As a result, production of stable-quality products can be continued without stopping the production line. Specifically, if you want a response equivalent to nσ, the difference between the second level and the first level and the difference between the third level and the second level are: (3 / 2) 1 / 2 n σ is determined in accordance with

[0065] For example, when the tolerance Δ=3σ, the generation unit 13 1st level = 2nd level - (3 / 2) 1 / 2 σ 3rd level = 2nd level + (3 / 2) 1 / 2 σ The first and third levels are set according to the following.

[0066] Fig. 10 shows an example of an experimental pattern generated according to the experimental design method. Fig. 10 shows 18 experimental patterns in which the parameters "feed speed (drying process)," "spray distance," "drying temperature," "feed speed (painting process)," "stirring speed," and "paint storage temperature" are set to one of three levels.

[0067] The experiment execution unit 14 (see FIG. 4) acquires, for each of a plurality of experiment patterns, measurement results of a specific parameter when one or more devices included in the production line 300 are controlled in accordance with the experiment pattern.

[0068] Specifically, the experiment execution unit 14 sequentially selects one experiment pattern from the multiple experiment patterns generated by the generation unit 13. The experiment execution unit 14 changes the target value of each parameter according to the selected experiment pattern. As a result, the IO processing unit 10 executes control arithmetic processing so that the value of each parameter approaches the changed target value, and outputs data obtained by the control arithmetic processing to one or more devices. The experiment execution unit 14 acquires the value of a specific parameter that indicates the quality of products produced by the production line 300 operating according to the changed target value. When the experiment execution unit 14 acquires the values of the specific parameter corresponding to multiple products, it calculates a representative value (e.g., an average value).

[0069] 11 is a diagram showing an example of measurement results acquired by the experiment execution unit 14. As shown in FIG. 11, the experiment execution unit 14 acquires measurement results of specific parameters for each of a plurality of experiment patterns.

[0070] Based on the measurement results obtained by the experiment execution unit 14, the setting unit 15 (see FIG. 4) sets new target values for the specific parameters to stabilize them within a specified range.

[0071] Specifically, the setting unit 15 determines, from among the predetermined number of parameters extracted by the extraction unit 12, target parameters to which new setting values are to be set.

[0072] The setting unit 15 performs a variance analysis on the measurement results of a plurality of experimental patterns, and calculates the variation S, variance V, and contribution rate p of each parameter. The setting unit 15 is based on Non-Patent Document 5 ("Concept and Utilization of Quality Engineering in the Development and Design Stage - System Evaluation and Improvement Without Prototypes and Tests" [online], [searched January 4, 2022], Internet<https: / / foundry.jp / bukai / wp-content / uploads / 2012 / 07 / e4806f10b0797ec0932d9317dd92a533.pdf> ) can be used to calculate the fluctuation S, variance V, and contribution rate p.

[0073] Furthermore, the setting unit 15 calculates the average value of the specific parameter value for each level for each parameter, and creates a factor-effect diagram based on the calculation results.

[0074] Fig. 12 is a diagram showing an example of the result of the analysis of variance performed by the setting unit, and Fig. 13 is a factorial effect diagram created by the setting unit.

[0075] The contribution rate p of each parameter is the ratio of the variation S of that parameter to the total variation S of all parameters. As shown in FIGS. 12 and 13, the larger the contribution rate p of a parameter, the greater the influence on the specific parameter when the target value is changed. Therefore, the setting unit 15 determines, as the target parameters, parameters whose contribution rate p exceeds a predetermined threshold value or a predetermined number of parameters with the highest contribution rates p. For example, in the example shown in FIG. 12, the setting unit 15 determines, as the target parameters, parameters whose contribution rates exceed 15%, namely, "spray distance," "feed speed (painting process)," and "paint storage temperature."

[0076] On the horizontal axis of Fig. 13, "1," [2], and [3] indicate the first level, second level, and third level, respectively. As shown in Fig. 13, the cause-effect diagram allows us to understand the direction of change (either positive or negative) in the value of a specific parameter when the target value of each parameter is changed from the second level to the first level or the third level.

[0077] Next, the setting unit 15 identifies the direction of fluctuation of the specific parameter caused by an abnormality in the operation of one or more devices included in the production line 300.

[0078] FIG. 14 is a diagram illustrating an example of a method for identifying the direction of fluctuation of a specific parameter caused by an abnormal operation of one or more devices. The setting unit 15 monitors the moving average of the specific parameter "paint film thickness" and identifies the direction of fluctuation of the specific parameter according to changes in the moving average. In the example shown in FIG. 14, the average value of the specific parameter in the second section (the section of product IDs "2216-2764") after the abnormal operation of one or more devices has occurred is lower than that in the first section (the section of product IDs "572-1120") before the abnormal operation of one or more devices has occurred. Therefore, the setting unit 15 identifies the direction of fluctuation of the specific parameter caused by the abnormal operation of one or more devices as the "negative direction."

[0079] The setting unit 15 preferably verifies whether the difference in the average value between the first section and the second section is significant using a T-test. If the result of the T-test verification indicates that the difference is significant, the setting unit 15 may identify the direction of variation in the specific parameter caused by an abnormality in the operation of one or more devices.

[0080] There are cases where no fluctuation is observed in the value of the specific parameter at the timing when an abnormality occurs in the operation of one or more devices included in the production line 300. In such cases, the setting unit 15 cannot identify the direction of fluctuation of the specific parameter caused by the abnormality in the operation of one or more devices. Therefore, the setting unit 15 monitors the moving average of the specific parameter until it can identify the direction of fluctuation of the specific parameter caused by the abnormality in the operation of one or more devices.

[0081] FIG. 15 is a diagram showing another example of a method for identifying the direction of fluctuation of a specific parameter caused by an abnormal operation of one or more devices. FIG. 15 shows the transition of the moving average of the predicted value of the specific parameter "paint film thickness" predicted using the T-method. When the T-method is used as the multivariate analysis, the setting unit 15 monitors the moving average of the predicted value of the specific parameter "paint film thickness" and identifies the direction of fluctuation of the specific parameter caused by an abnormal operation of one or more devices. That is, the setting unit 15 identifies the direction of fluctuation of the specific parameter at timing T1 when the fluctuation of the moving average becomes significant using a T-test.

[0082] The predicted value may be calculated from the parameters obtained before the drying process, such as "paint dilution ratio," "stirring speed," "paint storage temperature," "feed speed (paint process)," "air pressure," "room temperature," "discharge rate," and "spray distance." This allows the direction of change in a specific parameter to be identified earlier.

[0083] The setting unit 15 sets a new target value for the target parameter according to the direction and amount of variation of the specific parameter.

[0084] Specifically, the setting unit 15 determines the shift direction (either the direction toward the first level or the direction toward the third level) of the target value of the target parameter to change the value of the specific parameter in the direction opposite to the fluctuation direction of the specific parameter. The setting unit 15 determines the shift direction based on the cause-effect diagram shown in FIG. 13. In the example shown in FIG. 13, by shifting the target value of the target parameter toward the first level, the value of the specific parameter fluctuates toward the positive side. Therefore, when the fluctuation direction of the specific parameter is negative as shown in FIG. 14, the setting unit 15 determines the shift direction to be the direction toward the first level.

[0085] Furthermore, the setting unit 15 specifies a multiple N of the variation amount of the specific parameter relative to the standard deviation of the specific parameter. The setting unit 15 considers each parameter to be linear, and shifts the setting value of the target parameter in the shift direction by an amount obtained by multiplying the difference between the level corresponding to the shift direction and the second level by N. For example, assume that the standard center of the specific parameter "paint film thickness" is 17.5 μm and the width of the control range is 4.0 μm (=4σ). In this case, the variation amount of 0.2 μm shown in FIG. 14 corresponds to 0.2σ. That is, the setting unit 15 specifies that the multiple N is 0.2. If the shift direction is toward the first level and the difference between the first level and the second level is (3 / 2), 1 / 2 σ, the setting unit 15 sets 0.2×(3 / 2) 1 / 2 A value that is smaller than the second level by σ is determined as a new target value for the target parameter.

[0086] 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 executes a control calculation process so that the target parameter approaches the new target value, and outputs data obtained by the control calculation process to one or more devices. This stabilizes the value of the specific parameter that indicates the quality of the product.

[0087] Fig. 16 is a diagram showing an example of the transition of a specific parameter. In Fig. 16, timing T2 is the timing when a new target value of the target parameter is set by the setting unit 15. As shown in Fig. 16, after timing T2, the value of the specific parameter falls within the management range.

[0088] Graphs showing the transition of the statistical quantities monitored by the monitoring unit 11 (see FIGS. 6 and 7), information showing a predetermined number of parameters extracted by the extraction unit 12 (see FIG. 8), information showing the experiment results acquired by the experiment execution unit 14 (see FIG. 11), and a factor-effect diagram generated by the setting unit 15 (see FIG. 13) may be displayed on the HMI, thereby allowing the user to understand the monitoring situation using multivariate analysis.

[0089] §3 Supplementary Note As described above, the present embodiment includes the following disclosures.

[0090] (Configuration 1) A control device (100) for controlling one or more devices (311 to 313, 321, 322, 331, 332) included in a production line (300) for producing a product, a monitoring unit (11, 102) that monitors statistics obtained by performing multivariate analysis on a plurality of parameters related to the operation of the one or more devices (311 to 313, 321, 322, 331, 332); an extraction unit (12, 102) that extracts a predetermined number of parameters from the plurality of parameters that have a large influence on the variation of the statistical quantity when the variation of the statistical quantity is greater than a reference value; a generation unit (13, 102) that generates a plurality of experimental patterns in which the combinations of target values of the predetermined number of top parameters are different from each other in accordance with an experimental design method; an experiment execution unit (14, 102) that acquires measurement results of specific parameters that indicate the quality of the product when the one or more devices (311 to 313, 321, 322, 331, 332) are controlled according to each of the plurality of experiment patterns; a setting unit (15, 102) that sets new target values for the predetermined number of top parameters based on the measurement results to stabilize the values of the specific parameters within a control range; A control device (100) comprising: a control unit (10, 102) that controls the one or more devices (311 to 313, 321, 322, 331, 332) so that the top predetermined number of parameters approach the new target value.

[0091] (Configuration 2) 2. The control device (100) according to configuration 1, wherein the experimental design is a method using an orthogonal table.

[0092] (Configuration 3) The setting unit (15, 102) performing a variance analysis on the measurement results and calculating the contribution rate of each of the top predetermined number of parameters; selecting one or more target parameters for which the target value is to be set from among the predetermined number of top parameters based on the contribution rate; 3. The control device (100) according to configuration 1 or 2, wherein the new target value of the one or more target parameters is set.

[0093] (Configuration 4) the generating unit (13, 102) generates the plurality of experimental patterns so that the values of each of the top predetermined number of parameters are at any one of a first level, a second level, and a third level; the second level is a current target value, the first level is less than the second level; the third level is greater than the second level; 4. The control device (100) according to any one of configurations 1 to 3, wherein the first level and the third level are determined so that the value of the specific parameter falls within the control range.

[0094] (Configuration 5) A control method for controlling one or more devices (311 to 313, 321, 322, 331, 332) included in a production line (300) that produces a product, A step of monitoring statistics obtained by performing multivariate analysis on a plurality of parameters related to the operation of the one or more devices (311 to 313, 321, 322, 331, 332); extracting a predetermined number of parameters from the plurality of parameters that have a large influence on the variation of the statistical quantity when the variation of the statistical quantity is greater than a reference value; generating a plurality of experimental patterns each having a different combination of target values of the top predetermined number of parameters according to an experimental design method; For each of the plurality of experimental patterns, a step of acquiring a measurement result of a specific parameter indicating the quality of the product when the one or more devices (311 to 313, 321, 322, 331, 332) are controlled according to the experimental pattern; setting new target values for the predetermined number of top parameters based on the measurement results to stabilize the values of the specific parameters within a control range; and controlling the one or more devices (311 to 313, 321, 322, 331, 332) so that the top predetermined number of parameters approach the new target value.

[0095] Although the embodiments of the present invention have been described, the embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, and it is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0096] 1 System, 4 Control system network, 6 Information system network, 10 IO processing unit, 11 Monitoring unit, 12 Extraction unit, 13 Generation unit, 14 Experiment execution unit, 15 Setting unit, 100 Control device, 102 Processor, 104 Chipset, 106 Main memory, 110 Storage, 112 System program, 114 User program, 116 Monitoring program, 120 Control system network controller, 122 Information system network controller, 124 USB controller, 126 Memory card interface, 128 Memory card, 300 Production line, 310 Paint preparation process, 311 Raw material input machine, 312 Mixer, 313 Paint storage container, 320 Painting process, 321 Coating device, 322, 332 Belt conveyor, 330 Drying process, 331 Dryer, 340 Inspection process, 400 Metal plate.

Claims

1. A control device that controls one or more devices included in a production line that produces products, a monitoring unit that monitors statistics obtained by performing multivariate analysis on a plurality of parameters related to the operation of the one or more devices; an extracting unit that extracts a top predetermined number of parameters that have a large influence on the variation of the statistical quantity from among the plurality of parameters when the variation of the statistical quantity is greater than a reference value; a generation unit that generates a plurality of experimental patterns each having a different combination of target values of the predetermined number of top parameters according to an experimental design method; an experiment execution unit that acquires, for each of the plurality of experiment patterns, measurement results of a specific parameter that indicates quality of the product when the one or more devices are controlled in accordance with the experiment pattern; a setting unit that sets new target values for the predetermined number of top parameters based on the measurement results in order to stabilize the values of the specific parameters within a control range; a control unit that controls the one or more devices so that the top predetermined number of parameters approach the new target value.

2. The control device according to claim 1 , wherein the experimental design method is a method using an orthogonal table.

3. The setting unit performing a variance analysis on the measurement results and calculating the contribution rate of each of the top predetermined number of parameters; selecting one or more target parameters for which the target value is to be set from among the predetermined number of top parameters based on the contribution rate; The control device according to claim 1 or 2, further comprising: a controller configured to set the new target values of the one or more target parameters.

4. the generating unit generates the plurality of experimental patterns so that each value of the top predetermined number of parameters is at one of a first level, a second level, and a third level; the second level is a current target value, the first level is less than the second level; the third level is greater than the second level; The control device according to claim 1 , wherein the first level and the third level are determined so that the value of the specific parameter falls within the control range.

5. A control method for controlling one or more devices included in a production line that produces a product, comprising: monitoring statistics obtained by performing a multivariate analysis on a plurality of parameters related to the operation of the one or more devices; extracting a predetermined number of parameters from the plurality of parameters that have a large influence on the variation of the statistical quantity when the variation of the statistical quantity is greater than a reference value; generating a plurality of experimental patterns each having a different combination of target values of the top predetermined number of parameters according to an experimental design method; a step of acquiring, for each of the plurality of experimental patterns, measurement results of a specific parameter indicating quality of the product when the one or more devices are controlled according to the experimental pattern; setting new target values for the predetermined number of top parameters based on the measurement results to stabilize the values of the specific parameters within a control range; and controlling the one or more devices so that the top predetermined number of parameters approach the new target value.

Images