Method and computer system for monitoring tensile data

Abstract

This disclosure relates to a monitoring method, comprising: obtaining multiple tensile performance values ​​corresponding to a product; establishing multiple expert value ranges according to domain rules; applying the expert value ranges to at least one tensile performance value to generate multiple first values; establishing multiple statistical value ranges according to a statistical model; applying the statistical value ranges to at least one tensile performance value to generate multiple second values; and combining the first values ​​and the second values ​​to determine whether the product has an anomaly.

Description

[Technical Field]

[0001] This disclosure relates to a method for monitoring stretching data in a steel plant and a computer system for performing this method. [Previous Technology]

[0002] In steel mills, multiple sensors are installed to measure the tensile data of the products being produced, such as yield strength, elastic modulus, and elongation. To detect whether anomalies have occurred in the products, these tensile data can be statistically analyzed to determine whether they exceed the normal range. However, this method still cannot detect some abnormal phenomena. Furthermore, even if an anomaly is detected, conventional technology cannot determine whether the cause is an equipment error or a production line error. [Summary of the Invention]

[0003] The embodiments disclosed herein propose a method for monitoring tensile data, including: obtaining multiple tensile performance values ​​corresponding to a product; establishing multiple expert value ranges according to a domain rule; applying the expert value ranges to at least one tensile performance value to generate multiple first values; establishing multiple statistical value ranges according to a statistical model; applying the statistical value ranges to at least one tensile performance value to generate multiple second values; and combining the first values ​​and the second values ​​to determine whether the product has an abnormality.

[0004] In some embodiments, the step of combining the first value and the second value to determine whether the product is abnormal includes: setting multiple weights and calculating the weighted sum of the first value and the second value according to these weights; determining whether the weighted sum is less than a threshold value, and if so, determining that the product is abnormal.

[0005] In some embodiments, the monitoring method further includes: establishing multiple levels, wherein a lowest level corresponds to a product, the above weights are used as the first warning indicator corresponding to the lowest level; and summing the first warning indicators corresponding to the lowest level to generate a second warning indicator for the next higher level.

[0006] In some embodiments, the monitoring method further includes: obtaining a first tensile performance value and a second tensile performance value from the tensile performance values; generating a curve based on the first tensile performance value and the second tensile performance value; and determining whether the curve is a straight line, and if not, determining that a certain device has malfunctioned.

[0007] In some embodiments, the tensile property values ​​mentioned above include yield tensile ratio (YT), elastic modulus (E-mod), yield strength (YS), elongation (EL), thickness, R-value, or tensile strength (TS).

[0008] From another perspective, the embodiments disclosed herein provide a computer system including a memory and a processor. The memory is used to store multiple instructions, and the processor is communicatively connected to the memory to execute the instructions to complete the aforementioned monitoring method.

Implementation Method

[0010] The terms "first" and "second" used in this document do not specifically refer to order or sequence, but are used only to distinguish elements or operations described with the same technical terms.

[0011] Figure 1 is a schematic diagram illustrating a computer system for monitoring stretching data according to an embodiment. Referring to Figure 1, the computer system 100 can be a personal computer, laptop computer, server, distributed computer, cloud server, industrial computer, or various electronic devices with computing capabilities, etc., and this invention is not limited thereto. The computer system 100 includes a processor 110 and a memory 120. The processor 110 is communicatively connected to the memory 120. This communication connection can be achieved through any wired or wireless communication means, or it can also be achieved through the Internet. The processor 110 can be a central processing unit, microprocessor, microcontroller, special application integrated circuit, etc., and the memory 120 can be random access memory, read-only memory, flash memory, floppy disk, hard disk, optical disk, USB flash drive, magnetic tape, or a database accessible through the Internet, which stores multiple instructions. The processor 110 executes these instructions to complete the method for monitoring stretching data. This monitoring method will be described in detail below.

[0012] Figure 2 is a flowchart illustrating a method for monitoring tensile data according to an embodiment. Referring to Figure 2, in step 201, multiple tensile property values ​​corresponding to the product are obtained. These tensile property values ​​can be obtained using any suitable sensor. These tensile property values ​​may include yield tensile ratio (YT), elastic modulus (E-mod), yield strength (YS), elongation (EL), thickness, R-value, or tensile strength (TS), etc., and this disclosure is not limited thereto.

[0013] In step 202, multiple expert numerical ranges are established according to domain rules. Domain rules refer to the rules by which experts determine which tensile property values ​​are key values ​​based on their own experience, and set appropriate numerical ranges based on their knowledge of materials science or field practice. In some embodiments, the tensile property values ​​involved in step 202 include production tension ratio, elastic modulus, yield strength, elongation, R-value, and thickness; therefore, each tensile property value has a corresponding expert numerical range.

[0014] In step 203, the aforementioned expert value range is applied to the corresponding tensile performance value to generate multiple first values. Specifically, each expert value range can be used to form a function. Figure 3 is a schematic diagram of the function corresponding to the expert value range according to an embodiment. Referring to Figure 3, the horizontal axis represents the tensile performance value, and the vertical axis represents the converted value. This function is a discontinuous function, containing segments 301 to 303, where segment 302 represents a suitable value range, corresponding to a higher score (i.e., the first value). If the measured tensile performance value falls within segment 301, it indicates that it is less than the suitable value range, and a lower score can be set. Similarly, if the measured tensile performance value falls within segment 303, it indicates that it is greater than the suitable value range, and a lower score can also be set. Accordingly, each tensile performance value can be converted into a corresponding first value, and the lower the first value, the higher the probability of product abnormality.

[0015] In the embodiment shown in Figure 3, a discontinuous function is used, but in other embodiments, a continuous function can also be used. This function can be a linear function, a polynomial function, an exponential function, etc., and this disclosure is not limited thereto. In addition, in other embodiments, it can be set that the higher the first value, the higher the probability of product abnormality. For example, the value corresponding to segment 302 can be set to be lower, while the values ​​corresponding to segments 301 and 303 can be set to be higher, and so on.

[0016] Please refer to Figure 2. In step 204, multiple statistical value ranges are established based on the statistical model. In this embodiment, the tensile property values ​​involved in step 204 include yield strength, tensile strength, elongation, 0-degree R-value, 45-degree R-value, and 90-degree R-value. In other words, the tensile property values ​​analyzed in step 204 may partially overlap but are not entirely the same as the tensile property values ​​analyzed in step 202. Here, the statistical value range is set based on the statistical model, not based on expert experience. For example, the statistical model is used to calculate the mean and standard deviation. Here, the mean m and standard deviation s of a certain tensile property value can be calculated based on historical data, and then the statistical value range is set, where n is a real number, for example, 2.

[0017] In step 205, the aforementioned statistical value range is applied to the corresponding tensile property values ​​to generate multiple second values. Similar to step 203, a corresponding function can be set for each statistical value range, and the second value can be obtained by substituting the tensile property value into the function. In this embodiment, the function corresponding to the statistical value range may be different from the function corresponding to the expert value range. For example, the function corresponding to the expert value range may be discontinuous, while the function corresponding to the statistical value range may be continuous; this disclosure is not limited to this.

[0018] In step 206, the first and second values ​​described above are combined to determine whether the product is abnormal. Since each first and second value represents the quality of the product or the probability of an abnormality, these first and second values ​​can be summed. In some embodiments, multiple weights can be further set, and the weighted sum of the first and second values ​​can be calculated based on these weights. For example, if a certain tensile performance value is more important, a larger weight can be set, and vice versa. In some embodiments, these weights can also be normalized so that the sum of all weights equals 1. After calculating the weighted sum, it can be determined whether the weighted sum is less than a threshold value; if so, the product is determined to be abnormal. In embodiments where a larger value indicates a higher probability of an abnormality, it can be determined whether the weighted sum is greater than a threshold value; if so, the product is determined to be abnormal.

[0019] In step 207, a human-machine interface (HMI) is provided to display abnormal information. Tables, diagrams, animations, and any suitable text, numbers, colors, and patterns can be used to display the abnormal information; this disclosure is not limited to these methods. Figure 4 is a schematic diagram illustrating the HMI according to one embodiment. Referring to Figure 4, in this embodiment, a table 400 is displayed on the monitor (not shown) of the computer system 100. The columns of this table 400 are used to display information such as status, product type, steel grade, furnace number, test code, completion time, and results. Each column represents a product. When an abnormality is detected in a product, the corresponding status column is displayed in red; if there is no abnormality, it is displayed in green. In this way, the operator can learn about the relevant status of multiple products and whether there are any abnormalities from the table 400.

[0020] Please return to Figure 2. In step 208, the tensile performance values ​​are analyzed graphically. Specifically, based on the experience of on-site experts, two tensile performance values ​​can be obtained from all the tensile performance values ​​to plot a curve. A curve can be generated based on these two tensile performance values, as shown in Figure 5, where the horizontal axis is the first tensile performance value, which in this example is elongation, and the vertical axis is the second tensile performance value, which in this example is the R value. Six elongations and six R values ​​can be measured from a certain product, thereby generating six coordinate points on graph 500, which form curve 510. Since there should be a linear relationship between elongation and R value, it can be determined whether curve 510 is a straight line. If it is not a straight line, it is determined that a piece of equipment is malfunctioning. In this example, the equipment involved is a width gauge. There are several ways to determine whether curve 510 is a straight line. For example, the ratio of R value to elongation can be calculated for each coordinate point, and then the standard deviation of these ratios can be calculated. If the standard deviation is greater than a critical value, it indicates that it is not a straight line. Alternatively, a slope can be calculated for every two coordinate points; if the error between all the slopes is outside a certain range, it indicates that the line is not straight. Alternatively, regression can be used to approximate these coordinate points with a straight line; if the regression error is greater than a critical value, it indicates that the line is not straight.

[0021] In step 209 of Figure 2, warning indicators are displayed according to hierarchy. Figure 6 is a schematic diagram illustrating multiple levels according to an embodiment. Referring to Figure 6, in this example, three levels 601-603 are designed, where the lowest level 601 corresponds to the product, the next level 602 corresponds to the steel grade or production line, and the top level 603 corresponds to cold rolling or hot rolling. In other words, the design of levels 601-603 conforms to the production process in a steel plant. These processes can be drawn as a tree diagram, where the parent node's process covers all child nodes. For example, there are multiple products on a production line, and so on. The weight sum calculated in step 206 can be used as the warning indicator for the lowest level 601. When this warning indicator is less than a critical value, it indicates that the product is abnormal, which also indicates that a certain piece of equipment is abnormal. By summing up (or weighting) all the warning indicators at the lowest level 601, a warning indicator for the next higher level 602 can be generated. When the warning indicator at level 602 falls below a critical value, it indicates an anomaly in a certain production line or steel grade. Through this hierarchical analysis, operators can determine whether the anomaly is in the equipment, the production line / steel grade, or the entire cold / hot rolling process.

[0022] In some embodiments, when an abnormality is detected in the equipment, production line, steel grade, cold rolling, or hot rolling, the computer system 100 can send a warning message to relevant personnel. This warning message can be sent via speaker, screen, indicator light, SMS, email, etc., and this disclosure is not limited thereto. After receiving the warning message, relevant personnel can make corresponding adjustments.

[0023] Although the present invention has been disclosed above by way of embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims. [Simplified Explanation of the Diagram]

[0009] To make the above-mentioned features and advantages of the present invention more apparent and understandable, specific embodiments are described below in conjunction with the accompanying drawings. Figure 1 is a schematic diagram illustrating a computer system for monitoring tensile data according to an embodiment. Figure 2 is a flowchart illustrating a method for monitoring tensile data according to an embodiment. Figure 3 is a schematic diagram illustrating a function corresponding to an expert numerical range according to an embodiment. Figure 4 is a schematic diagram illustrating a human-machine interface according to an embodiment. Figure 5 is a schematic diagram illustrating a curve according to an embodiment. Figure 6 is a schematic diagram illustrating multiple levels according to an embodiment.

Claims

1. A method for monitoring stretching data, comprising: Obtain multiple tensile property values ​​corresponding to a product; Establish multiple expert numerical ranges according to a domain rule; perform a numerical generation step to apply the expert numerical ranges to at least one of the tensile performance values ​​to generate multiple first values ​​having a first score or a second score, wherein the numerical generation step includes: forming a function using one of the expert numerical ranges, wherein the function includes a first function segment and a second function segment; determining that one of the tensile performance values ​​falls within the first function segment or the second function segment; assigning the first score when the tensile performance value falls within the first function segment; and assigning the second score, which is less than the first score, when the tensile performance value falls within the second function segment; establish multiple statistical numerical ranges according to a statistical model; apply the statistical numerical ranges to at least one of the tensile performance values ​​to generate multiple second values; and combine the first values ​​and the second values ​​to determine whether the product is abnormal.

2. The monitoring method as described in Request 1, wherein the step of combining the first values ​​and the second values ​​to determine whether the product has experienced an anomaly includes: Set multiple weights, and calculate a weighted sum of the first and second values ​​based on these weights: Determine if the weighted sum is less than a threshold value; if so, determine that the product is abnormal.

3. The monitoring method as described in request item 2 further includes: Establish multiple levels, with the lowest level corresponding to the product, and the weight sum serves as the first warning indicator corresponding to the lowest level; And sum up the first warning indicator corresponding to the lowest level to generate a second warning indicator for the next higher level.

4. The monitoring method as described in Request 1 further includes: A first tensile property value and a second tensile property value are obtained from these tensile property values; A curve is generated based on the first tensile property value and the second tensile property value; and it is determined whether the curve is a straight line, and if not, it is determined that a device malfunctions.

5. The monitoring method as described in claim 1, wherein the tensile property values ​​include yield tensile ratio (YT), elastic modulus (E-mod), yield strength (YS), elongation (EL), thickness, R-value, or tensile strength (TS).

6. A computer system for monitoring stretching data, comprising: A memory module used to store multiple instructions; The system also includes a processor communicatively connected to the memory for executing instructions to perform multiple steps: obtaining multiple tensile performance values ​​corresponding to a product; establishing multiple expert value ranges according to a domain rule; performing a value generation step to apply the expert value ranges to at least one of the tensile performance values ​​to generate multiple first values ​​having a first score or a second score, wherein the value generation step includes: forming a function using one of the expert value ranges, wherein the function includes a first function segment and a second function segment; determining that one of the tensile performance values ​​falls within the first function segment or the second function segment; assigning the first score when the tensile performance value falls within the first function segment; and assigning the second score, which is less than the first score, when the tensile performance value falls within the second function segment; establishing multiple statistical value ranges according to a statistical model; applying the statistical value ranges to at least one of the tensile performance values ​​to generate multiple second values; and combining the first values ​​and the second values ​​to determine whether the product is abnormal.

7. The computer system as described in claim 6, wherein the step of combining the first values ​​and the second values ​​to determine whether the product is malfunctioning includes: Set multiple weights, and calculate a weighted sum of the first and second values ​​based on these weights: Determine if the weighted sum is less than a threshold value; if so, determine that the product is abnormal.

8. The computer system as described in claim 7, wherein the steps further include: Establish multiple levels, with the lowest level corresponding to the product, and the weight sum serves as the first warning indicator corresponding to the lowest level; And sum up the first warning indicator corresponding to the lowest level to generate a second warning indicator for the next higher level.

9. The computer system as described in claim 6, wherein the steps further include: A first tensile property value and a second tensile property value are obtained from these tensile property values; A curve is generated based on the first tensile property value and the second tensile property value; and it is determined whether the curve is a straight line, and if not, it is determined that a device malfunctions.

10. The computer system as described in claim 6, wherein the tensile property values ​​include yield tensile ratio (YT), elastic modulus (E-mod), yield strength (YS), elongation (EL), thickness, R-value, or tensile strength (TS).

Images