JP5441824B2 - Manufacturing condition determination system for metal strip materials - Google Patents

Description

  The present invention provides, for example, metal strip material production conditions that determine production conditions such as line speed and temperature setting from the viewpoint of productivity and quality improvement for metal strip materials produced by rolling, annealing, cleaning, degreasing, etc. Regarding the decision system.

Conventionally, a metal strip material is manufactured through a plurality of processing steps such as rolling, annealing, cleaning, and degreasing. In each of these processing steps, the work procedure is specified as a work standard, but there are often no standard production conditions related to numerical values such as plate feed speed (line speed), temperature, pressure, etc. It is.
For this reason, the setting of the manufacturing conditions such as the plate passing speed and the temperature setting has been manually performed using the past experience and know-how by the operator (skilled worker). That is, the operator for each equipment takes into consideration the productivity of the equipment in charge, and tries to pass through as quickly as possible so as to bring out the equipment performance. On the other hand, it is practically difficult to consider how the manufacturing conditions set by the operator himself have an influence on the lower process in terms of quality, and in terms of quality by increasing the plate passing speed. There is no denying that there is a possibility that negative effects have also occurred. In addition, when a plurality of operators are in charge of the same equipment, it is a fact that the manufacturing conditions vary depending on the operators.

For example, Patent Document 1 discloses a technique for improving quality by suppressing variations in manufacturing conditions of personnel such as operators as much as possible.
In Patent Document 1, manufacturing history information is stored in association with a manufacturing lot number and an identification number for a product manufactured through a plurality of work processes, and the specified defective phenomenon and the specified manufacturing condition item are defective. The cause of manufacturing defects can be analyzed by displaying the frequency distribution of products with defects and products without defects.

  In the technique of this Patent Document 1, the specified defect phenomenon, the frequency distribution with and without the defect phenomenon for the specified manufacturing condition are displayed in a graph, the frequency distribution with the defect, and the defect If the shape of the distribution is different from the frequency distribution in the case of no occurrence, it is determined that the manufacturing conditions have some influence on the defect of the product. Similarly, if the frequency distribution when there is a defect and the frequency distribution when there is no defect are similar in shape, it is determined that the manufacturing conditions do not affect the cause of the defect. Yes.

JP 2008-181341 A

  In the technique of Patent Document 1, although it is possible to graphically display the frequency distribution of the case where there is a defective phenomenon and the case where there is no defect with respect to the designated manufacturing condition, it can be determined whether or not the manufacturing condition is appropriate. It is not specifically disclosed what kind of manufacturing conditions should be selected in order to reduce product defects from such graphs. That is, even if the technique disclosed in Patent Document 1 is used, it is actually difficult to set appropriate manufacturing conditions from the displayed graph.

  Therefore, in view of the above-described problems, the present invention provides a metal strip that can easily determine an appropriate manufacturing condition for manufacturing conditions for manufacturing the metal strip material while passing through and processing the metal strip material. An object of the present invention is to provide a material manufacturing condition determination system.

In order to achieve the above-described object, the present invention takes the following technical means.
The metal strip material manufacturing condition determination system according to the present invention is a metal strip material manufacturing condition determination system for determining a manufacturing condition for manufacturing a metal strip material, and stores manufacturing performance information including the manufacturing conditions of the metal strip material. Manufacturing history storage unit for storing, manufacturing history storage unit for storing the manufacturing history for each metal band material generated based on the manufacturing result information of the manufacturing result storage unit, and quality information including quality defect information of the metal band material A quality information storage unit to store, a process classification database unit that stores a process in the manufacturing process of the metal strip material and a process classification to which the process belongs, and stores quality defect information related to the process classification The quality information storage unit searches the quality information storage unit for quality defect information related to the metal strip material whose manufacturing history is stored in the manufacturing history storage unit, and the searched quality defect And quality information retrieval means for appending a broadcast in the production history of the production history storage unit, based on the quality information of the production history of the production history storage unit the quality information storage unit, using the process classification database unit In addition to obtaining the process classification of the specified process, and using the quality defect database section to obtain quality defect information related to the process classification, the relevance between the manufacturing conditions and the quality of the metal strip material is calculated. And a manufacturing condition calculation unit that calculates a new manufacturing condition based on the relationship between the manufacturing condition obtained by the quality condition calculating unit and the quality.

Here, the quality condition calculation unit, the production result information and quality defect information of the production history manufacturing history acquired from the storage unit, the grouping unit for grouping the occurrence of quality defects of the quality defect information indicates And a frequency distribution counting unit that counts the quality defect occurrence status in a predetermined manufacturing condition using the manufacturing performance information and the quality defect information grouped by the grouping unit, the manufacturing condition calculation unit, It is preferable that a new manufacturing condition is determined by using the quality defect occurrence state in the predetermined manufacturing condition that is counted by the frequency distribution counting unit.

Here, the quality status calculation unit calculates the quality defect occurrence rate for the quality defect occurrence rate in a predetermined manufacturing condition using the quality defect occurrence status tabulated by the frequency distribution totaling unit. The manufacturing condition calculation unit calculates an estimated value of the quality defect occurrence rate obtained from the regression analysis based on the manufacturing condition and the quality defect occurrence rate, and determines a new manufacturing condition using the calculated estimated value. It is preferable to be configured to do so.

Here, the manufacturing condition calculation unit is configured to obtain a threshold value of the occurrence rate of quality failure from the distribution of the production conditions and the quality failure occurrence rate, and to determine a new production condition using the obtained threshold value. And preferred.
According to such a metal band material manufacturing condition determination system, by using the process classification database and the quality defect database, the process related to the quality defect occurring in the metal band material can be relied on for each operator depending on different experiences. You can know without.

For example, when determining manufacturing conditions for a certain process, the process classification database is referred to, the process classification to which the process belongs is acquired, and the quality defect database related to the process classification is acquired. To do.
Based on the quality defect information acquired in this way, the manufacturing history recorded in the manufacturing history storage unit is written with the group in which the quality defect information is written (where the quality defect has occurred) and the quality defect information. It is possible to analyze the manufacturing conditions of each group by dividing them into groups that are not.

  ADVANTAGE OF THE INVENTION According to this invention, an appropriate manufacturing condition can be easily determined about the manufacturing conditions for manufacturing a metal strip material, processing and passing a metal strip material.

It is the figure which showed an example of the manufacturing process of a metal strip. It is the figure which showed an example of the manufacture performance data preserve | saved in the manufacture performance database. It is the figure which showed an example of the data preserve | saved in the manufacture history database. It is the figure which added the flow of data in the manufacturing process shown in FIG. It is a figure which shows a process classification database. It is a figure which shows a quality defect database. It is a figure which shows the frequency distribution of the number of non-defective goods by the boarding speed. It is the figure which showed the frequency distribution of the number of inferior goods by the boarding speed. It is the figure which showed the quality defect incidence distribution by the boarding speed. It is a figure which shows the regression calculation result of a plate-feeding speed and a quality defect incidence rate. It is a figure which shows the operation | movement flow of the manufacturing condition determination system of a metal strip. It is a figure which shows the operation | movement flow of the manufacturing condition determination system of a metal strip.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.
FIG. 1 shows an example of a manufacturing process (manufacturing line) of a metal strip. In the present embodiment, description will be made using a metal strip as an example of the metal strip material.

  As shown in FIG. 1, for example, a metal strip used for manufacturing automobiles, home appliances and the like is hot-rolled in a hot rolling process 1 and rolled in a cold state to a thin plate having a predetermined thickness in a cold-rolling process 2. . Further, the metal strip is subjected to continuous annealing in the annealing step 3, and then pickled in the pickling step 4 to remove scales and the like, surface-treated in the surface treatment step 5, and final inspection step 6 After being inspected whether or not it has quality as a product, it is shipped. Thus, the metal strip is passed through a plurality of processes such as a hot rolling process 1, a cold rolling process 2, an annealing process 3, a pickling process 4, a surface treatment process 5, and a final inspection process 6 at a predetermined speed. To be manufactured.

Various data for performing each process, and data (manufacturing performance data) as a result of actually performing each process are support devices 10 (for example, computers) that manage the production line of the metal strip. The processing conditions (manufacturing conditions for the metal strip) of each process are determined by processing various data by the support device 10.
Hereinafter, the determination system that determines the manufacturing conditions of the metal strip by the support device 10 will be described in detail.

  The support device 10 includes a manufacturing result database (manufacturing result storage unit) 210. This production record database 210 is a production record such as equipment information of equipment used in a process, process information for specifying a process, a metal strip identification number (manufacture number), and a metal strip passing speed for each process. Data is recorded. The production performance database 210 records production performance data that changes from time to time by associating equipment, processes, and production numbers.

  As shown in FIG. 2, the manufacturing performance database 210 records the time when the manufacturing performance data was recorded, the sheet passing speed, the roll current, and the load as the manufacturing performance data. When the strip is rolled, an equipment number (S001) such as rolling equipment is recorded as equipment information, and a process number (K001) indicating the hot rolling process 1 is recorded as process information. In addition, the manufacturing performance database 210 records the manufacturing number (X0001) of the metal strip in order to identify the metal strip individually, and the time when the manufacturing performance data is recorded and the communication of the metal strip at the time. The sheet speed, the roll current for rotating the rolling roll that rolls the metal strip at the time, and the load of the rolling roll at the time are recorded. As shown in FIG. 2, the manufacturing performance database 210 records manufacturing performance data every second.

  As described above, the production result data every moment in each process is stored in the production result database 210, but the support device 10 stores the production result data for each attribute (for example, each production number) of the metal strip. As a manufacturing history database (manufacturing history storage unit) 400. This manufacturing history database 400 collects manufacturing performance data from a hot rolling process 1 to a final inspection process 6 in a time series and records it as manufacturing history data. ing.

  As shown in FIG. 3, the manufacturing history database 400 includes, for example, the manufacturing history data of the metal strip having the manufacturing number “X0001”, the start time (work start) and end time (work end) of each process, The maximum value and median value of the sheet feeding speed in the process, and the temperature of the metal strip are recorded. The manufacturing history database 400 also records quality information (for example, quality failure information “defective 1” indicating failure) linked to a quality information storage unit described later. That is, the manufacturing history database 400 not only records manufacturing performance data classified and aggregated for each attribute of the metal strip, but also records quality defect information associated with the manufacturing performance data.

The support device 10 will be described in detail with reference to FIG. FIG. 4 is obtained by adding a broken line so that the flow of each data can be understood to the manufacturing process of the metal strip of FIG.
Although the support apparatus 10 includes the manufacturing performance database 210 as described above, the manufacturing performance data recorded by the manufacturing performance database 210 is each process of the hot rolling process 1 to the surface treatment process 5 as indicated by a broken line A. To the manufacturing result database 210.

Further, the support apparatus 10 includes a quality information database (quality information storage unit) 310. The quality information database 310 records data indicating quality defects such as surface defects and dirt on the metal strip in each process, in other words, quality information including quality defect information indicating defective items (defect types) of the metal strip. To do.
Specifically, a surface inspection apparatus 7 for inspecting the surface of the metal strip is provided in the cold rolling process 2 and the surface treatment process 5, and data acquired by the surface inspection apparatus 7 (for example, the surface of the metal strip) The data obtained by imaging () is output to the support device 10. The support device 10 receives and analyzes the data output from the surface inspection device 7 to detect the presence / absence of defects such as surface flaws and dirt on the metal strip (hereinafter, sometimes simply referred to as defects).

  When a defect is detected in the surface inspection apparatus 7 or the final inspection process 6, the support apparatus 10 converts the detected defect item (quality defect data) into the surface inspection apparatus 7 and the final as shown by a broken line B in FIG. The quality defect data received from the inspection process 6 is recorded as quality information in the quality information database 310 in association with the process of detecting the defect and the metal strip production number (X0001).

  Thereby, it can be understood in which process the defect of the metal strip is detected. In the present embodiment, the surface inspection apparatus 7 is provided in the cold rolling process 2 and the surface treatment process 5 and is also provided in the final inspection process 6 although not shown. Therefore, the defect (quality defect data) of the metal strip is assumed to have been detected in any of the cold rolling process 2, the surface treatment process 5 or the final inspection process 6 in which the surface inspection apparatus 7 is provided. It is recorded in the information database 310.

The surface inspection device 7 described above may be provided in all the steps of the hot rolling step 1 to the final inspection step 6, but if there are restrictions on equipment, the surface inspection device 7 may be provided only in the final inspection step 6. Good.
The support device 10 further includes a quality information search unit 100. The quality information search unit 100 searches the quality information database 310, associates the detected defect (quality defect data) with the above-described manufacturing history data, and adds the result to the manufacturing history database 400.

As a result, as shown in FIG. 3, in the manufacturing history data of the metal strip with the manufacturing number (X0001), “defect 1”, which is quality information indicating a quality defect item (type), is recorded in the manufacturing history of the process K020. Is done.
The support apparatus 10 includes a process classification database 500 that is a process master. As shown in FIG. 5, the process classification database 500 is a database indicating which process classification each process belongs to. For example, it is understood that the process “K001” belongs to “rolling” in the process classification.

Further, the support device 10 includes a quality defect master, in other words, a quality defect database 600.
As shown in FIG. 6, the quality defect database 600 is a database that indicates a relationship with defective items that are likely to occur for each process classification in the process classification database 500 of FIG. 5.
For example, it is understood that defect 1, defect 2, and defect 5 are likely to occur in the process classification “rolling”. Conversely, it can be seen that the defects 1 and 2 are likely to occur in the annealing process and the cleaning process, and the defect 5 is likely to occur in all processes. The quality defect database 600 is maintained based on the experience of operators on the production line and theoretical predictions.

Further, the support device 10 includes a quality status calculation unit 50.
The quality status calculation unit 50 is stored in the manufacturing history database 400 and has a predetermined quality based on the manufacturing history data shown in FIG. 3 to which the quality information is added, the process classification database 500, and the quality defect database 600. The relevance between manufacturing conditions and quality is calculated.
The quality status calculation unit 50 uses, for example, predetermined manufacturing conditions such as a sheet feeding speed and quality information as a key (main information), and the quality situation (non-defective product of the metal strip) when the sheet feeding speed is a predetermined value. Number, number of defects, defect occurrence rate, etc.).

Specifically, the quality status calculation unit 50 includes a grouping unit 60, a frequency distribution totaling unit 70, and a quality defect occurrence rate calculating unit 80.
The grouping unit 60 acquires the process classification of the process designated by the operator or the like from the process classification database 500 shown in FIG. 5, and relates to the acquired process classification based on the quality defect database 600 shown in FIG. Acquire quality defect information (defective items), group the manufacturing history data in which the same defective items are recorded from the manufacturing history database 400 using the acquired defective items as a key, and group them. Manufacturing history data in which no items are recorded is grouped.

  For example, when the operator designates the process K002, the support apparatus 10 refers to the process classification database 500 illustrated in FIG. 5 and acquires the process classification “rolling” corresponding to the process K002. Next, the support apparatus 10 refers to the quality defect database 600 illustrated in FIG. 6 and acquires defect 1, defect 2, and defect 5, which are defect items corresponding to the acquired process classification “rolling”. Thereafter, the support device 10 groups the manufacturing history data in which the defect 1, defect 2, or defect 5 is recorded in the quality information as one group, and the defect 1, defect 2, or defect 5 is not recorded in the quality information. Manufacturing history data is grouped as another group.

  The grouping unit 60 accesses the manufacturing history database 400 and refers to the quality information. Instead, the grouping unit 60 accesses the quality information database 310 and refers to the quality information to obtain the defect 1 and defect 2. , Or the manufacturing number of the metal strip on which the defect 5 is recorded may be detected, and the manufacturing result data corresponding to the detected manufacturing number may be grouped as a group with a defect, and the others may be grouped as a group without a defect. .

The frequency distribution totaling unit 70 uses the manufacturing result data (group with defect, group without defect) grouped by the grouping unit 60 to total the quality status under predetermined manufacturing conditions.
For example, the frequency distribution counting unit 70 calculates and counts the number of defective products for each predetermined sheeting speed from the defective group when counting using the sheeting speed as a key in the manufacturing conditions, and the predetermined number of sheets from the defect-free group. The number of non-defective products for each plate speed is calculated and tabulated, and these are stored as tabulated data.

  As described above, when the quality distribution (the number of non-defective products or the number of defective products) is totaled by the frequency distribution totaling unit 70 using the sheet feeding speed as a key in the manufacturing conditions, graphs as shown in FIGS. 7 and 8 can be obtained. These graphs are the results of dividing the sheet feeding speed (median value) every 50 [mpm], and counting and counting how many metal strips were passed for each section.

As shown in FIG. 7 and FIG. 8, regardless of whether the product is non-defective or defective, the frequency of the metal strip with a plate passing speed (median) of about 500 [mpm] is the highest, and the plate passing speed is the highest. The frequency decreases as the speed becomes lower or higher. However, the frequency distribution of non-defective products has a narrower distribution peak width than the frequency distribution of defective products.
The quality defect occurrence rate calculation unit 80 obtains the number of non-defective products and the number of defective products under a predetermined production condition from the frequency distribution totaling unit 70, and obtains the quality defect occurrence rate with respect to the predetermined production condition. For example, the quality defect occurrence rate calculation unit 80 includes a frequency distribution of the number of non-defective products for each predetermined sheet feeding speed as shown in FIG. Using the frequency distribution of defective products for each predetermined sheet passing speed (numerical values counted as defective products by the frequency distribution totaling unit 70), the defective product occurrence rate is obtained for each sheet passing speed.

As described above, when the quality defect occurrence rate is obtained by the quality defect occurrence rate calculation unit 80 using the sheet feeding speed as a key in the manufacturing conditions, the graph shown in FIG. 6 is obtained.
The support device 10 further includes a manufacturing condition calculation unit 90. The manufacturing condition calculation unit 90 is based on the relationship between the manufacturing conditions and the quality obtained by the quality status calculation unit 50 including the grouping unit 60, the frequency distribution totaling unit 70, and the quality defect occurrence rate calculating unit 80 described above. In addition, new manufacturing conditions for improving the quality (conditions for reducing the number of defective products or conditions for improving the quality) are calculated.

The manufacturing condition calculation unit 90, for example, the frequency distribution of the number of non-defective products shown in FIG. 7 obtained from the manufacturing performance data and the quality defect data, or the manufacturing history data, the frequency distribution of the number of defective products shown in FIG. A new manufacturing condition is obtained using the indicated quality defect occurrence rate and the like.
Specifically, the manufacturing condition calculation unit 90 performs a polynomial regression analysis using the quality defect occurrence rate with respect to the sheet passing speed as an explanatory variable and the quality defect occurrence rate as an objective variable, as shown in FIG. An estimated value (also referred to as an estimated value curve) is obtained by polynomial regression between the plate passing speed and the quality defect occurrence rate. FIG. 10 shows a change in the estimated value of the defect occurrence rate with a solid line (estimated value curve), and the original quality defect occurrence rate data with a broken line for reference. As shown in FIG. 10, it can be confirmed that the quality defect occurrence rate is high when the plate passing speed is low and high, and the quality defect occurrence rate is low near the plate passing speed of 500 [mpm].

  Then, when manufacturing the metal strip using the estimated value curve of the quality defect occurrence rate with respect to a predetermined sheet passing speed, the manufacturing condition calculating unit 90 estimates that the actual sheet passing speed is obtained from the estimated value curve. A plate passing speed (new manufacturing condition) lower than the value is obtained, and this is applied to the plate passing speed of the process designated by the operator. That is, the manufacturing condition calculation unit 90 calculates an estimated value of the quality defect occurrence rate obtained from the regression analysis based on the sheet feeding speed and the quality defect occurrence rate, and uses the calculated estimated value for the production condition ( Manufacturing conditions for reducing the number of defective products) are determined.

  11 and 12 show an operation flow of the metal strip manufacturing condition determination system. The flow shown in FIG. 11, that is, steps S101 and S102, shows a maintenance procedure of the production history database 400, and is continuously executed during operation of the metal strip production line. The flow shown in FIG. 12, that is, steps S201 to S207, shows the procedure for determining the manufacturing conditions of the metal strip, and after the operations of steps S101 and S102 are executed and the manufacturing history database 400 is prepared, It is executed when considering the production conditions of the metal strip.

First, the maintenance procedure of the manufacturing history database 400 will be described with reference to FIGS.
In step S101, as indicated by a broken line A in FIG. 4, the support device 10 produces a manufacturing result that is sent every second from each step of the production line while the metal strip is passing through the production line. Data is recorded in the manufacturing performance database 210. Thereby, the manufacturing result database 210 shown in FIG. 2 is maintained.

At the same time, as shown by a broken line B in FIG. 4, the support device 10 uses the quality inspection data of the metal strip detected through the surface inspection device 7 and the final inspection process 6, the process of detecting the failure and the metal strip of the metal strip. Recorded in the quality information database 310 in association with the production number. As a result, the quality information database 310 is maintained.
Thereafter, the support apparatus 10 creates manufacturing history data by using the manufacturing number data of each metal strip (for each metal strip) as a key using the manufacturing record data accumulated in the manufacturing record database 210. Data is stored in the manufacturing history database 400. The manufacturing history database 400 at this time is in a state where no quality information is recorded in the manufacturing history database 400 shown in FIG.

In step S102, when the operation in step S101 proceeds and manufacturing history data is stored in the manufacturing history database 400, the quality information search unit 100 of the support apparatus 10 records the quality defect recorded with reference to the quality information database 310. Search for information.
When quality defect information is detected from the quality information database 310, the quality information search unit 100 adds the detected quality defect information to the corresponding manufacturing history data in the manufacturing history database 400. As a result, the manufacturing history database 400 is maintained as a database having quality information as shown in FIG.

After preparing the manufacturing history database 400 as described above, when the manufacturing conditions of the metal strip are examined, the manufacturing process recorded in the manufacturing history database 400 in the flow of FIG. 12, that is, the procedure shown in steps S201 to S207. A new manufacturing condition is calculated using the history data. The calculation of the manufacturing conditions will be described with reference to FIG.
In step S <b> 201, first, an operator or the like designates the process and the type of manufacturing condition to be examined in the process to the quality status calculation unit 50 of the support apparatus 10. Thereafter, the grouping unit 60 of the quality status calculation unit 50 acquires the process classification of the designated process from the process classification database 500 shown in FIG.

In step S202, the grouping unit 60 acquires quality defect information, which is a defect item related to the process classification acquired in step S201, based on the quality defect database 600 shown in FIG.
Subsequently, in step S203, the grouping unit 60 uses the acquired defective item as a key, acquires manufacturing history data in which the same defective item is recorded in the quality information column from the manufacturing history database 400, and groups them. Manufacturing history data in which the defective items are not recorded is grouped. Then, the numerical value of the manufacturing condition corresponding to the process designated by the operator or the like in step S201 is acquired from the manufacturing history data belonging to each group.

  In step S204, after grouping in step S203, for example, when the production conditions in each group are acquired and aggregated using the maximum sheet feeding speed value as the designated production condition as a key, the frequency distribution aggregation unit 70 The frequency distribution of the number of defective products with respect to the maximum threading speed in the process specified in S201 is obtained from the group of manufacturing history data in which defective items are recorded, and the frequency distribution of the number of non-defective products is not recorded in the defective items. Obtained from the group of manufacturing history data. These frequency distributions are graphs as shown in FIGS.

In step S205, after obtaining the frequency distribution in step S204, the quality defect occurrence rate calculation unit 80 obtains the number of non-defective products and the number of defective products for each of the divisions of the frequency distribution, and calculates the quality defect occurrence rate. The quality defect occurrence rate is a graph as shown in FIG.
In step S206, after calculating the quality defect occurrence rate, the manufacturing condition calculation unit 90 performs a polynomial regression analysis using the sheet feeding speed as an explanatory variable and the quality defect occurrence rate as an objective variable, and results as shown in the graph of FIG. Get.

Finally, in step S207, the manufacturing condition calculation unit 90 obtains the maximum plate speed at which the estimated value of the quality defect occurrence rate is below the threshold, and presents the numerical value as a recommended manufacturing condition. The operator sets the numerical value presented in step S207 as a new production condition in the designated process in the metal strip production line.
The operation from step S201 to step S207 has been described above as being performed after the maintenance of the manufacturing history database 400 is completed. However, even during the maintenance of the manufacturing history database 400, even if the operations from step S201 to step S207 are executed using only the maintained manufacturing history data at a stage where the manufacturing history is maintained to some extent, the recommended manufacturing conditions. Can be presented.

As described above, since the support device 10 includes the process classification database 500 and the quality defect database 600, it is possible to know the processes related to the quality defects that have occurred in the metal strip without depending on different experiences for each operator. .
For example, when determining the manufacturing conditions of a certain process, the process classification database 500 is referred to obtain the process classification to which the process belongs, and the quality defect database 600 is referred to to determine the quality defect related to the process classification. Get the type.

The production history data recorded in the production history database 400 is divided into a group in which the acquired quality failure has occurred and a group in which the quality failure has not occurred, based on the type of quality failure thus acquired, The manufacturing conditions of each group can be analyzed.
In the above embodiment, the regression analysis is used when newly determining the production conditions from the quality defect occurrence rate. For example, if the relationship between the production conditions and the quality defect occurrence rate is a gentle curve, Alternatively, a threshold value that is not an estimated value of the quality defect occurrence rate may be simply set to calculate a manufacturing condition equal to or less than the threshold value. The manufacturing condition calculation unit 90 may calculate a threshold value of the quality defect occurrence rate using the distribution of the manufacturing condition and the quality defect occurrence rate, and determine a new manufacturing condition using the calculated threshold value.

As described above, in the present invention, based on the manufacturing performance data and the quality information data resulting from the inspection of the metal strip, the non-defective product and the defective product are grouped, and the frequency distribution due to the difference in manufacturing conditions In addition, the manufacturing conditions are calculated using the quality defect occurrence rate for each manufacturing condition.
Therefore, a method for determining manufacturing conditions based on data analysis that excludes personality has been established for manufacturing conditions that have been determined from experience by skilled workers or have been varied by workers, and variations in manufacturing conditions have been reduced. Can be standardized. Further, in determining the manufacturing conditions, it is possible to set conditions so that defective products are not generated as much as possible, so that the quality of the metal strip can be improved and the yield can be improved.

In said embodiment, although the attribute of the metal strip was distinguished based on a manufacturing number, it is not restricted to this, The magnitude | size (length, width) of a metal strip, and the steel type (kind) of a metal strip You may distinguish by etc.
The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. In particular, in the embodiment disclosed this time, matters that are not explicitly disclosed, for example, operating conditions and operating conditions, various parameters, dimensions, weights, volumes, and the like of a component deviate from a range that a person skilled in the art normally performs. Instead, values that can be easily assumed by those skilled in the art are employed.

DESCRIPTION OF SYMBOLS 1 Hot rolling process 2 Cold rolling process 3 Annealing process 4 Pickling process 5 Surface treatment process 6 Final inspection process 7 Surface inspection apparatus 10 Support apparatus 50 Quality condition calculation part 60 Grouping part 70 Frequency distribution totaling part 80 Quality defect occurrence rate calculation Unit 90 Manufacturing condition calculation unit 100 Quality information search unit 210 Manufacturing result storage unit 310 Quality information storage unit 400 Manufacturing history storage unit 500 Process classification database 600 Quality defect database

Claims (4)

A metal strip material manufacturing condition determination system for determining manufacturing conditions for manufacturing a metal strip material,
A production record storage unit that stores production record information including the production conditions of the metal strip material;
A manufacturing history storage unit for storing a manufacturing history for each metal strip material generated based on the manufacturing result information of the manufacturing result storage unit;
A quality information storage unit for storing quality information including quality defect information of the metal strip material;
A process classification database unit that stores the process in the manufacturing process of the metal strip material and the process classification to which the process belongs, in association with each other;
A quality defect database part storing quality defect information related to the process classification;
Quality information that searches the quality information storage unit for quality defect information related to the metal strip material whose manufacturing history is stored in the manufacturing history storage unit, and adds the searched quality defect information to the manufacturing history of the manufacturing history storage unit Search means;
Based on the manufacturing history of the manufacturing history storage unit and the quality information of the quality information storage unit, the process classification database unit is used to obtain the process classification of the designated process, and the quality defect database unit is Using the quality status calculation unit that calculates the relationship between the manufacturing conditions and the quality of the metal strip material by acquiring quality defect information related to the process classification ,
A manufacturing condition determination system for a metal strip material, comprising: a manufacturing condition calculation unit that calculates a new manufacturing condition based on the relationship between the manufacturing condition obtained by the quality status calculation unit and the quality . The quality status calculation unit
A grouping unit for grouping the occurrence of said manufacturing performance information and quality defect information of the production history manufacturing history acquired from the storage unit, the quality the quality defect information indicates failure,
A frequency distribution totaling unit that totalizes the quality defect occurrence status in a predetermined manufacturing condition using the manufacturing performance information and the quality defect information grouped by the grouping unit,
The said manufacturing condition calculation part is comprised so that a new manufacturing condition may be determined using the quality defect occurrence condition in the predetermined manufacturing condition totaled by the said frequency distribution total part. The manufacturing condition determination system of the metal strip material described in 1. The quality status calculation unit includes a quality defect occurrence rate calculation unit for obtaining a quality defect occurrence rate in a predetermined manufacturing condition using the quality defect occurrence status tabulated by the frequency distribution counting unit ,
The manufacturing condition calculation unit is configured to calculate an estimated value of the quality defect occurrence rate obtained from the regression analysis based on the manufacturing condition and the quality defect occurrence rate, and to determine a new manufacturing condition using the calculated estimated value. The manufacturing condition determination system for a metal strip material according to claim 2, wherein the manufacturing condition is determined. The manufacturing condition calculation unit is configured to obtain a threshold value of a quality defect occurrence rate from the distribution of the manufacturing condition and the quality defect occurrence rate, and to determine a new production condition using the obtained threshold value. A manufacturing condition determination system for a metal strip material according to claim 3.

Images