TWI889144B - Methods of width reduction and width control - Google Patents

Abstract

A width reduction method suitable for a rolling process performed on a steel coil is disclosed. The width reduction method is executed by a computing device and includes the following steps: obtaining basic data of the steel coil; performing a pre-processing operation on the basic data of the steel coil; establishing a width reduction algorithm and using the width reduction algorithm to calculate a width reduction amount of the steel coil; and setting the width reduction amount in the trimming equipment that performs the rolling process to cut the steel coil.

Description

Width reduction and width control methods

The present invention relates to a method for width reduction and width control, and in particular to a method for width reduction and width control of a steel coil used in a rolling process.

In the current rolling process production line, the method of controlling the output width of steel coils is to use a statistical model combined with a table lookup method derived from actual production experience to calculate the steel coil width reduction. However, the table lookup method cannot effectively classify various steel types and distinguish the differences between steel types. In addition, basic steel coil data such as steel coil feed thickness and steel coil reduction are not calculated using actual values but range values, which will lead to a certain error between the table lookup result and the actual steel coil output width.

The purpose of the present invention is to provide a width reduction and width control method, which is respectively applicable to a rolling process with and without trimming equipment. In a rolling process with trimming equipment, a width reduction algorithm can be established to calculate the width reduction of the steel coil so that the trimming width of the steel coil meets the required output width of the steel coil; on the other hand, in a rolling process without trimming equipment, a tension control algorithm can be established to calculate the width contraction of the steel coil so that the contraction width of the steel coil meets the required output width of the steel coil.

One aspect of the present invention is to provide a width reduction method suitable for a rolling process of a steel coil. The width reduction method is executed by a computing device and includes the following steps: obtaining basic data of the steel coil; performing pre-processing operations on the basic data of the steel coil; establishing a width reduction algorithm and using the width reduction algorithm to calculate the width reduction amount of the steel coil; and setting the width reduction amount in the trimming device to perform a trimming process on the steel coil.

In some embodiments, the basic data of the steel coil includes the steel coil feed material width, the steel coil feed material thickness, the steel coil output width, the steel coil output thickness, the steel coil alloy composition, the steel coil steel grade code and the steel coil process code.

In some embodiments, the pre-processing operation includes calculating the reduction rate of the steel coil and the width-to-height ratio of the steel coil.

In some embodiments, the width reduction is obtained by substituting the steel coil reduction rate, steel coil width-to-height ratio, steel coil carbon content, steel coil silicon content, steel coil manganese content, steel coil chromium content, and steel coil molybdenum content into the width reduction algorithm for calculation.

In some embodiments, the width reduction algorithm includes the following steps: obtaining at least one process information of a rolling process performed on a steel coil; using a machine learning method to establish a width reduction model using at least one process information as training data; and verifying and updating the width reduction model.

Another aspect of the present invention is to provide a width control method, which is applicable to a rolling process for a steel coil. The width control method is executed by a computing device and includes the following steps: obtaining basic data of the steel coil; performing pre-processing operations on the basic data of the steel coil; establishing a tension control algorithm and calculating the width compression of the steel coil using the tension control algorithm; and setting the width compression in the tension control equipment for the rolling process to perform stretching and compression processing on the steel coil.

In some embodiments, the basic data of the steel coil includes the steel coil feed material width, the steel coil feed material thickness, the steel coil output width, the steel coil output thickness, the steel coil alloy composition, the steel coil steel grade code and the steel coil process code.

In some embodiments, the pre-processing operation includes calculating the reduction rate of the steel coil and the width-to-height ratio of the steel coil.

In some embodiments, the width reduction is calculated by substituting the coil reduction rate, coil width-to-height ratio, coil carbon content, coil silicon content, coil manganese content, coil chromium content, and coil molybdenum content into the tension control algorithm.

In some embodiments, the tension control algorithm includes the following steps: obtaining at least one process information of a rolling process performed on a steel coil; using a machine learning method to establish a tension control model using at least one process information as training data; and verifying and updating the tension control model.

100: Width reduction method

200: Width control method

S110,S120,S130,S140,S210,S220,S230,S240: Steps

In order to make the above and other purposes, features, advantages and embodiments of the present invention more clearly understandable, the attached drawings are described as follows: FIG. 1 is a flow chart of a width reduction method according to an embodiment of the present invention; and FIG. 2 is a flow chart of a width control method according to an embodiment of the present invention.

The following is a detailed discussion of embodiments of the present invention. It is understood that the embodiments provide many applicable concepts that can be implemented in a variety of specific contexts. The embodiments discussed and disclosed are for illustrative purposes only and are not intended to limit the scope of the present invention.

FIG. 1 is a flow chart of a width reduction method 100 according to an embodiment of the present invention. The width reduction method 100 is applicable to a rolling process for a steel coil, which is executed by a computing device and includes steps S110 to S140, which are described below.

Step S110: Obtain the basic data of the steel coil. This step is to explain that before the steel coil width is reduced, the necessary information of the steel coil must be collected, that is, the basic data of the steel coil must be read, including the steel coil feed width, steel coil feed thickness, steel coil output width, steel coil output thickness, steel coil alloy composition, steel coil steel grade code, steel coil process code or other basic data of the steel coil, but not limited to this.

Step S120: Pre-processing the basic data of the steel coil. This step is to explain that after the basic data of the steel coil is obtained in step S110, the obtained basic data of the steel coil is then pre-processed, including calculating the steel coil reduction rate and the steel coil width-to-height ratio of the rolling process. In one embodiment of the present invention, the steel coil reduction rate and the steel coil width-to-height ratio can be calculated by the following formulas (1) and (2) respectively: Steel coil reduction rate = (steel coil feed thickness - steel coil output thickness) / steel coil output thickness, (1)

Steel coil width-to-height ratio = steel coil input width/steel coil output width. (2)

In addition, the pre-processing of the basic data of the steel coil also includes calculating the alloy content of the steel coil (such as the carbon content of the steel coil, the silicon content of the steel coil, the manganese content of the steel coil, the chromium content of the steel coil, and the molybdenum content of the steel coil), calculating the tension setting value of the coiling machine, and confirming whether the steel coil has undergone an annealing process.

Step S130: Establish a width reduction algorithm and use the width reduction algorithm to calculate the width reduction of the steel coil. This step is to explain that after the processing operation before step S120, the process information including the steel coil reduction rate, the steel coil width-to-height ratio, the steel coil alloy component content, the coiling machine tension setting value, whether the steel coil has undergone an annealing process, etc. is obtained, and then the machine learning method is used to use these process information as training data (training data) to establish a width reduction model of the width reduction algorithm to calculate the width reduction of the steel coil. Specifically, the width reduction is obtained by substituting the steel coil reduction rate, the steel coil width-to-height ratio, the steel coil carbon content, the steel coil silicon content, the steel coil manganese content, the steel coil chromium content and the steel coil molybdenum content into the width reduction algorithm for calculation. It should be noted that the machine learning method used may be decision tree regression, random forest, linear regression, polynomial regression, logistic regression, support vector machine (SVM) and/or other suitable machine learning methods. After the width reduction model is established, the continuously collected new process information is then used as validation data to input into the established width reduction model for validation and obtain performance indicators. In one embodiment of the present invention, when the performance index is greater than the threshold, a new width reduction model is established based on the collected new process information as training data to replace the original width reduction model to calculate the width reduction of the steel coil, that is, to update the width reduction algorithm; conversely, when the performance index is less than or equal to the threshold, the original width reduction model is used to calculate the width reduction of the steel coil, and new process information is continuously collected.

In some embodiments, the width reduction method 100 can be executed on a computing device equipped with a processor. The processor can automatically collect process information of the rolling process after a specific time period, and verify the performance index obtained by the process information collected in each cycle to be greater than, less than or equal to the threshold value to determine whether to update the width reduction algorithm at the moment. It should be noted that the computing device can be a desktop computer, a laptop, a tablet computer, a smart phone, a cloud server or other suitable device equipped with a processor.

Step S140: Set the width reduction in the trimming equipment for the rolling process. This step is to illustrate that the width reduction method 100 is applicable to a rolling production line equipped with a trimming equipment. After the width reduction of the steel coil is calculated in step S130, the width reduction is set in the trimming equipment so that the trimming width obtained by the trimming equipment when cutting the steel coil according to the width reduction meets the required output width of the steel coil.

After the rolling process is completed, all process information including the trimming information of the width reduction during the rolling process is stored in the database for future query, access or establishment of width reduction calculation. It should be noted that the output width of the steel coil after the rolling process is completed can be measured by width measuring equipment or other appropriate measuring equipment.

FIG. 2 is a flow chart of a width control method 200 according to an embodiment of the present invention. The width control method 200 is applicable to a rolling process for a steel coil, which is executed by a computing device and includes steps S210 to S240, which are described below.

It should be noted that the width reduction method 100 is applicable to a rolling production line equipped with trimming equipment, and the width reduction amount needs to be calculated so that the width of the steel coil after cutting meets the output width of the steel coil after the rolling process; the width control method 200 is different from the width reduction method 100, and is applicable to a rolling production line without trimming equipment, and the tension setting value of the coiler needs to be changed to control the steel coil to meet the output width of the steel coil after the rolling process.

Step S210: Obtain the basic data of the steel coil. This step is to explain that before the steel coil width stretching and tightening is performed, the necessary information of the steel coil must be collected, that is, the basic data of the steel coil must be read, including the steel coil feed width, steel coil feed thickness, steel coil output width, steel coil output thickness, steel coil alloy composition, steel coil steel grade code, steel coil process code or other basic data of the steel coil, but not limited to this.

Step S220: Pre-processing the basic data of the steel coil. This step is to explain that after the basic data of the steel coil is obtained in step S210, the obtained basic data of the steel coil is then pre-processed, including calculating the steel coil reduction rate and the steel coil width-to-height ratio of the rolling process. In one embodiment of the present invention, the steel coil reduction rate can be calculated by the following formula (3): steel coil reduction rate = (steel coil feed thickness - steel coil output thickness) / steel coil output thickness, (3) and the steel coil width-to-height ratio can be calculated by the above formula (2).

In addition, the pre-processing of the basic data of the steel coil also includes calculating the alloy content of the steel coil (such as the carbon content of the steel coil, the silicon content of the steel coil, the manganese content of the steel coil, the chromium content of the steel coil, and the molybdenum content of the steel coil), calculating the tension setting value of the coiling machine, and confirming whether the steel coil has undergone an annealing process.

Step S230: Establish a tension control algorithm and use the tension control algorithm to calculate the width compression of the steel coil. This step is to explain that after the processing operation before step S220, the process information including the steel coil reduction rate, the steel coil width-to-height ratio, the steel coil alloy component content, the coiling machine tension setting value, whether the steel coil has undergone an annealing process, etc. is obtained, and then the machine learning method is used to use this process information as training data A tension control model of a tension control algorithm is established to calculate the width compression of the steel coil. Specifically, the width compression is obtained by substituting the steel coil reduction rate, steel coil width-to-height ratio, steel coil carbon content, steel coil silicon content, steel coil manganese content, steel coil chromium content and steel coil molybdenum content into the tension control algorithm for calculation. It should be noted that the machine learning method used can be decision tree regression, random forest, linear regression, polynomial regression, logical regression, support vector machine and/or other suitable machine learning methods. After the tension control model is established, the new process information collected continuously is then input as verification data into the established tension control model for verification and performance indicators are obtained. In one embodiment of the present invention, when the performance indicator is greater than the threshold, a new tension control model is established based on the new process information collected as training data to replace the original tension control model to calculate the width compression of the steel roll, that is, to update the tension control algorithm; conversely, when the performance indicator is less than or equal to the threshold, the original tension control model is used to calculate the width compression of the steel roll, and new process information is continuously collected.

In some embodiments, the width control method 200 may be executed on a computing device equipped with a processor, and the processor may automatically collect process information of the rolling process after a specific time period, and determine whether to update the tension control algorithm at the moment by verifying that the performance index obtained by the process information collected in each cycle is greater than/less than or equal to the threshold value. It should be noted that the computing device may be a desktop computer, a laptop, a tablet computer, a smart phone, a cloud server, or other suitable device equipped with a processor.

Step S240: Set the width contraction amount in the tension control device for the rolling process. This step is to illustrate that the width control method 200 is applicable to the rolling production line without trimming equipment. After the width contraction amount of the steel roll is calculated in step S230, the width contraction amount is set in the tension control device so that the tension control device stretches and contracts the steel roll according to the width contraction amount to obtain a contraction width that meets the required steel roll output width.

After the rolling process is completed, the various process information including the tension control information of the width compression during the rolling process is stored in the database for future query, access or establishment of tension control algorithm. It should be noted that the output width of the steel coil after the rolling process is completed can be measured by width measuring equipment or other appropriate measuring equipment.

In summary, the width reduction and width control methods of the present invention are respectively applicable to rolling processes with and without trimming equipment. In the rolling process with trimming equipment, a width reduction algorithm can be established to calculate the width reduction of the steel coil so that the trimming width of the steel coil meets the required output width of the steel coil; on the other hand, in the rolling process without trimming equipment, a tension control algorithm can be established to calculate the width contraction of the steel coil so that the contraction width of the steel coil meets the required output width of the steel coil.

Although the present invention has been disclosed as above by way of embodiments, it is not intended to limit the scope of the present invention. Anyone with ordinary knowledge in the relevant technical field can make various changes, substitutions and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of the patent application attached hereto.

100: Width reduction method

S110, S120, S130, S140: Steps

Claims (6)

A width reduction method is applicable to a rolling process for a steel coil. The width reduction method is executed by a computing device and comprises: obtaining basic data of the steel coil; performing a pre-processing operation on the basic data of the steel coil, wherein the pre-processing operation comprises calculating a reduction rate of the steel coil and a width-to-height ratio of the steel coil; A width reduction algorithm is established and a width reduction amount of the steel coil is calculated using the width reduction algorithm, wherein the width reduction amount is obtained by substituting the reduction rate of the steel coil, the aspect ratio of the steel coil, a carbon content of the steel coil, a silicon content of the steel coil, a manganese content of the steel coil, a chromium content of the steel coil and a molybdenum content of the steel coil into the width reduction algorithm for calculation; and the width reduction amount is set in a trimming device performing the rolling process to perform a trimming process on the steel coil. A width reduction method as described in claim 1, wherein the basic data of the steel coil includes a steel coil feed width, a steel coil feed thickness, a steel coil output width, a steel coil output thickness, a steel coil alloy composition, a steel coil steel grade code and a steel coil process code. A width reduction method as described in claim 1, wherein the width reduction algorithm comprises: obtaining at least one process information of the rolling process performed on the steel coil; using a machine learning method to use the at least one process information as training data to establish a width reduction model for the trimming equipment; and verifying and updating the width reduction model. A width control method is applicable to a rolling process for a steel coil. The width control method is executed by a computing device and comprises: obtaining basic data of the steel coil; performing a pre-processing operation on the basic data of the steel coil, wherein the pre-processing operation comprises calculating a reduction rate of the steel coil and a width-to-height ratio of the steel coil; A tension control algorithm is established and a width contraction amount of the steel coil is calculated using the tension control algorithm, wherein the width contraction amount is obtained by substituting the reduction rate of the steel coil, the aspect ratio of the steel coil, a carbon content of the steel coil, a silicon content of the steel coil, a manganese content of the steel coil, a chromium content of the steel coil and a molybdenum content of the steel coil into the tension control algorithm for calculation; and the width contraction amount is set in a tension control device for performing the rolling process so as to perform a stretching and contraction treatment on the steel coil. A width control method as described in claim 4, wherein the basic data of the steel coil includes a steel coil feed width, a steel coil feed thickness, a steel coil output width, a steel coil output thickness, a steel coil alloy composition, a steel coil steel grade code and a steel coil process code. A width control method as described in claim 4, wherein the tension control algorithm comprises: obtaining at least one process information of the rolling process performed on the steel coil; using a machine learning method to use the at least one process information as training data to establish a tension control model for the tension control device; and verifying and updating the tension control model.

Images