Method and system for generating a heat map of comprehensive quality grades of steel coils based on a rule-based decision model

Abstract

The application provides a steel coil comprehensive quality grade thermodynamic map generation method and system based on a rule judgment model. The method and system first receive real-time production line iron loss data, appearance grade data, edge loss data and steel coil specifications. Then, based on predefined rules, the iron loss grade and appearance grade classification are determined. The comprehensive quality grade judgment result of each area of the steel coil is obtained by combining the two dimensions of iron loss and appearance grade. Finally, a thermodynamic map is generated with the comprehensive quality grade as the parameter, and the steel coil comprehensive pass rate is calculated, realizing the quantitative evaluation of the overall quality of the steel coil. By using the segmented and layered, multi-dimensional judgment method, the application not only improves the accuracy, fineness and efficiency of the steel coil quality evaluation, but also intuitively displays the comprehensive quality distribution of the steel coil, which helps to analyze the production quality fluctuation, optimize the product quality, provide guidance for subsequent processing, reduce the judgment scrap rate and reduce material waste.

Description

Technical Field

[0001] This invention relates to the field of data visualization technology in the steel industry, and in particular to a method and system for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based judgment model. Background Technology

[0002] In the field of steel coil production and quality management, ensuring product quality and accurately assessing quality grades are crucial. This directly impacts a company's reputation, market competitiveness, and customer satisfaction. Therefore, product quality inspection plays a vital role in steel coil production, typically encompassing both in-process inspection and finished product inspection. Traditional finished product inspection relies primarily on manual sampling, a method that is not only inefficient but also prone to biases and omissions, failing to provide comprehensive and accurate reflection of the product's true quality status.

[0003] With the advancement of intelligent manufacturing and Industry 4.0, steel coil manufacturers are facing increasingly stringent quality control requirements, leading to the emergence of quality management systems. However, most existing quality management systems focus only on single-dimensional inspections, such as surface defects and production process parameters, lacking comprehensive analysis of quality data across multiple dimensions. In particular, they lack integrated analysis of iron loss data and surface inspection data. Iron loss is a crucial indicator reflecting the electromagnetic properties of steel coils and is of significant value for timely product quality diagnosis.

[0004] Therefore, this invention proposes a method and system for generating heat maps of comprehensive quality grades of steel coils based on a rule-based judgment model, which solves the above-mentioned problems. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based judgment model. By segmenting and layering the steel coil into several small regions, and based on preset judgment rules, comprehensively evaluating iron loss and appearance grade data, the comprehensive quality grade of each region is determined and a heat map is generated to intuitively display the quality distribution of the steel coil. At the same time, the key quantitative indicator of "comprehensive positive product rate" is used to achieve a quantitative assessment of the overall quality of the steel coil.

[0006] A method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model includes the following steps: S1. Receive steel coil data from the production line in real time. The steel coil data includes iron loss data, appearance grade data, edge loss data, and steel coil specifications. The iron loss data refers to the iron loss value of W locations continuously detected along the length of the steel coil. The appearance grade data refers to the appearance grade obtained by continuously detecting all areas after dividing the length of the steel coil into N segments and each segment into several layers in the width direction. The edge loss data includes the upper edge loss width and the lower edge loss width. The steel coil specifications include the steel coil type, steel coil number, and steel coil thickness. S2. Based on the specifications of the steel coil, predefine the judgment rules and build a set of rule judgment models. The rule judgment models include iron loss level judgment model, appearance level classification model and comprehensive quality level judgment model. S3. Iron loss level determination: First, the received iron loss value is processed to obtain the iron loss value of each length region of the steel coil. Then, the processed iron loss value is used as a parameter to input into the iron loss level determination model to obtain the iron loss level determination result of each length region. S4. Appearance grade classification: Since the appearance grade inspection proceeds from the tail to the head of the steel coil, the appearance grade data must first be reversed before being input into the appearance grade classification model to obtain the appearance grade classification results for each area. S5. Comprehensive quality grade determination: The iron loss grade and appearance grade are simultaneously input into the comprehensive quality grade determination model as parameters to obtain the comprehensive quality grade determination results and location information of each region of the steel coil. S6. Establish a two-dimensional coordinate system with the lower left corner of the steel coil head as the origin, use the comprehensive quality level as the heat map parameter, design a color configuration scheme, construct a two-dimensional matrix, and generate a heat map. S7. Map the processed iron loss value, appearance grade, and overall quality grade to the corresponding positions on the heat map, and display them on the heat map through mouse operation, so as to intuitively analyze and understand the distribution and correlation of each indicator.

[0007] Furthermore, in step S1, by comprehensively evaluating the surface insulation resistance, unevenness, color difference, appearance, and coating quality of the steel coil, the appearance grade is divided into five levels: GG01, GG02, GG03, GG04, and GG05.

[0008] Furthermore, the rule determination model in step S2 also includes a grade determination model. The grade is an identifier for the steel coil, named according to the iron loss range. The iron loss range is divided according to enterprise standards, and each iron loss range corresponds to a fixed grade, named from smallest to largest as 70, 80, 90, 100, 110, 120, 130 and above. These values ​​are called grades, and the grades can more intuitively reflect the iron loss level.

[0009] Furthermore, the method for processing the iron loss value in step S3 is as follows: take the average iron loss value of the starting and ending positions of the steel coil length region as the iron loss value of that region, and sequentially traverse the received iron loss values ​​to obtain the iron loss value of each length region.

[0010] Furthermore, the specific method for defining the determination rules in step S2 is as follows: S201. Define the iron loss level determination rule: The iron loss value is set as P, the lower threshold is set as P1, and the upper threshold is set as P2. If P≤P1, the iron loss level is "excellent"; if P1<P≤P2, the iron loss level is "medium"; if P>P2, the iron loss level is "poor". P1 and P2 are set according to the steel coil specifications and enterprise standards. S202. Define the appearance grade classification rules: appearance grades GG01 and GG02 are "genuine products", and appearance grades GG03 to GG05 are "concessions". S203. Define grade determination rules: Set multiple thresholds, compare the iron loss value with the thresholds in sequence, determine the iron loss range that falls into, and obtain the corresponding grade. S204. Define the comprehensive quality grade judgment rules: Define the judgment rules from two dimensions: iron loss grade and appearance grade. When the iron loss grade is excellent and the appearance grade is graded as genuine, the comprehensive quality grade is "comprehensive genuine product". All others are "comprehensive concession".

[0011] Furthermore, to further refine the overall quality grade of steel coils, the specific steps are as follows: First, the appearance grade classification in step S202 is refined. The judgment criteria are set as follows: appearance grade GG01 is "Genuine Product 1", GG02 is "Genuine Product 2", GG03 is "Concession 1", GG04 is "Concession 2", and GG05 is "Concession 3". Second, to align with the refined appearance grade classification, the judgment criteria in step S204 are set as follows: when the iron loss grade is excellent and the appearance grade is "Genuine Product 1", the overall quality grade is "Genuine Product 1-1"; when the iron loss grade is excellent and the appearance grade is "Genuine Product 2", the overall quality grade is "Genuine Product 1-2"; when the iron loss grade is medium and the appearance grade is "Genuine Product 1", the overall quality grade is "Concession 1-1". The overall quality grade is defined sequentially based on the iron loss grade and the appearance grade classification.

[0012] In actual steel coil production and testing, the iron loss range and the appearance grade length range are significantly inconsistent. The iron loss range is much smaller than the appearance grade length range, resulting in multiple iron loss ranges falling within the same appearance grade length range. Furthermore, the same iron loss range may span two different appearance grade length ranges. Therefore, determining the overall quality grade is a complex process. Further, the specific steps for determining the overall quality grade in step S5 are as follows:

[0013] First, define the total length of the steel coil as X and the total width as Y; For the iron loss interval, let X i-1 X i Let represent the left and right boundary lengths of the i-th iron loss interval, respectively, where i is a positive integer and satisfies 1≤i≤W-1; For the length range of appearance grade, let L j-1 L j Let represent the left and right boundary lengths of the j-th appearance level length interval, respectively, where j is a positive integer and satisfies 1≤j≤N-1; For the appearance grade width range, let Y m-1 Y m This represents the lower and upper boundary widths of the m-th appearance level width interval, where m is a positive integer, 1≤m≤M-1; During initialization, X0=L0=0 and X1<L1 are set, and it is determined that the first iron loss interval completely falls within the first appearance level length interval.

[0014] Next, set m=0 and, based on the pre-built rule-based judgment model, determine the comprehensive quality level of the first appearance grade width range under the first iron loss interval; then, increment m sequentially to complete the comprehensive quality level judgment of all layered areas under the first iron loss interval; after completing the judgment, exit the current loop and increment i sequentially to i+1, if X i-1 ≥L j-1 And X i ≤L j If the iron loss range falls entirely within the length range of the corresponding appearance grade, repeat the above process; if X i-1 <L j <X i To determine that the current iron loss interval spans different appearance grade length intervals, the current iron loss interval needs to be subdivided into two sub-intervals: (X i-1 L j ) and (L j X i Then, perform a comprehensive quality level determination on each of the two sub-intervals, and exit the current loop after completion. Repeat the above loops and iterations for i, j, and m in turn until all areas of the steel coil have been determined;

[0015] Finally, the overall quality grade results and location information of all areas of the steel coil are recorded to generate a heat map. Simultaneously, the location information of all positive product areas is output separately, including X... i-1 X i L j Y m-1 Y m It is used to calculate the overall quality rate of steel coils.

[0016] Although the overall quality grade of each region can reflect the distribution characteristics of the steel coil quality, it cannot directly reflect the overall quality level of the steel coil. Therefore, this invention introduces the key quantitative indicator of "overall good product rate" to measure the overall quality of the steel coil. The level of the overall good product rate directly reflects the quality of the overall steel coil; the higher the overall good product rate, the better the overall quality of the steel coil. The formula for calculating the overall good product rate is: Overall good product rate = Total area of ​​good products / (Total area of ​​steel coil - Total area of ​​edge damage), where the total area of ​​good products refers to the sum of the areas of all good product areas of the steel coil, the total area of ​​the steel coil refers to the area of ​​the entire steel coil, and the total area of ​​edge damage refers to the sum of the areas of all edge damage areas.

[0017] Furthermore, the area of ​​the genuine product area is related to its location. When the genuine product area involves upper and lower edge-damaged areas, the area of ​​the genuine product area needs to be subtracted accordingly. The location of the genuine product area can be divided into four categories: a. If Y m-1 >0 and Y m When =Y, it is determined that the current positive product area is located at the upper edge of the steel coil, and the area of ​​this positive product area = (Y – Y) / (Y). m-1 – D k ) * (X i – X i-1 ); b. If Y m-1 =0 and 0 < Y m When <Y, it is determined that the current genuine product area is located at the lower edge of the steel coil, and the area of ​​this genuine product area = (Y) m – E k ) * (X i – X i-1 ); c. If Y m-1 =0 and Y m When =Y, it is determined that the current genuine product area is not layered in the width direction, and the area of ​​the genuine product area = (Y – D) k – E k ) * (X i – X i-1 ); d. If Y m-1 >0 and Y m When <Y, it is determined that the current positive product area is located in the middle of the steel coil, and the area of ​​this positive product area = (Y) m – Y m-1 ) * (X i – X i-1 ); Therefore, the method for calculating the area of ​​the genuine product area is as follows: based on the location information of all genuine product areas provided in step S503, determine the location of each genuine product area, and calculate the area of ​​each genuine product area according to the location matching area calculation formula. The total area of ​​the genuine product area is obtained by summing the areas of all genuine product areas.

[0018] Edge damage detection is performed in segments along the length of the steel coil. Let the length of the steel coil be divided into K segments, and D... k E k L represents the upper and lower edge loss widths of the k-th edge loss interval, respectively. k-1 L k Let the lengths of the left and right boundaries of the k-th edge loss interval be represented respectively, where 1 ≤ k ≤ K-1, and the total area of ​​the edge loss interval = .

[0019] Furthermore, the heatmap generation method is as follows: A two-dimensional coordinate system is established with the lower left corner of the steel coil head as the origin. The length direction is set as the X-axis and the width direction as the Y-axis. The "comprehensive quality grade" is used as the parameter of the heat map. A color configuration scheme is designed, a two-dimensional matrix is ​​constructed, and a heat map is generated.

[0020] Furthermore, based on the differences in overall quality levels, a color scheme is designed, specifically as follows: (1) Differentiated color configurations are implemented for each comprehensive quality level, and corresponding color values ​​are assigned. Among them, the "comprehensive genuine product" and "comprehensive concession" are distinguished by contrasting colors, and their sub-levels are distinguished by colors of the same color family but different brightness; (2) Configure the color of the upper and lower edge damage of the steel coil: Based on the received edge damage data, calculate the location of all edge damage areas and reset the edge damage areas to "gray".

[0021] Furthermore, the grade, appearance level, overall quality level, and processed iron loss value are mapped to the corresponding positions on the heat map and displayed on the heat map through mouse operation, so as to intuitively analyze and understand the distribution and correlation of various indicators.

[0022] This invention also provides a system for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based judgment model. The system includes: a data receiving module, an iron loss grade judgment module, an appearance grade grading module, a grade judgment module, a comprehensive quality grade judgment module, a heat map generation module, and a steel coil overall quality calculation module. The data receiving module receives steel coil iron loss data, appearance grade data, edge loss data, and steel coil specifications in real time. The iron loss grade judgment module obtains the iron loss grade results for each length region of the steel coil according to preset iron loss grade judgment rules. The appearance grade grading module obtains the appearance grade grading for each region of the steel coil according to preset appearance grade grading rules. The comprehensive quality grade judgment module determines the comprehensive quality grade of each region of the steel coil based on two dimensions: iron loss and appearance grade, according to preset comprehensive quality grade judgment rules. The heat map generation module generates a heat map using the comprehensive quality grade as a parameter. The steel coil overall quality calculation module calculates the comprehensive yield of the entire steel coil according to a formula.

[0023] Furthermore, the iron loss level determination module, appearance level classification module, grade determination module, and comprehensive quality level determination module all support custom determination rules. The system allows users to select and view any area on the heat map by mouse operation, and instantly displays the grade, appearance level, comprehensive quality rating, and iron loss value corresponding to that area.

[0024] The beneficial effects of the present invention are as follows: (1) It provides a segmented, layered, multi-dimensional comprehensive quality judgment method for steel coils. This method segments the entire steel coil in the length direction and layers it in the width direction. Combining iron loss data and surface detection data, a set of rule judgment models is built through custom judgment rules. Based on the AviatorScript rule engine, the comprehensive quality level of all areas of the steel coil is judged. It not only breaks through the limitations of traditional finished product inspection and improves the accuracy, precision and efficiency of steel coil quality assessment, but also helps to establish a comprehensive quality assessment standard for the steel coil industry.

[0025] (2) A method for generating a heat map of the comprehensive quality grade of steel coils and a method for calculating the comprehensive positive product rate are provided. The heat map can intuitively display the quality grade of each region, while the comprehensive positive product rate presents the overall quality of the steel coils in a quantitative form. This can not only provide feedback for the production process, but also provide a scientific basis for product market price setting and enhance market competitiveness.

[0026] (3) A comprehensive quality grade heat map generation system for steel coils is provided. This system has the ability to comprehensively detect the quality of steel coils. It not only helps to analyze the quality fluctuations in the steel coil production process and provide real-time feedback data to the production process to help improve product quality, but also provides accurate guidance for subsequent processing, such as selecting cross-cutting or longitudinal cutting methods to reduce the scrap rate caused by substandard quality and effectively reduce material waste.

[0027] (4) The proposed rule-based judgment method supports user customization, which facilitates manual review to verify the rationality of the judgment rules. Once an unreasonable situation is found, the rule-based judgment model can be verified by modifying the judgment threshold, thereby improving the accuracy of the comprehensive quality judgment of steel coils. Attached Figure Description

[0028] Figure 1 The flowchart of the method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based judgment model proposed in this invention.

[0029] Figure 2 The flowchart of the definition and determination rule method proposed in this invention.

[0030] Figure 3 A schematic diagram of the structure of the heat map generation system for comprehensive quality grade of steel coils based on a rule-based judgment model proposed in this invention. Specific Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] This embodiment provides the following: Figure 1 The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model, as shown, specifically includes the following steps: S101. Receive steel coil data from the production line in real time, including iron loss data, appearance grade data, edge loss data, and steel coil specifications; S102. Based on the specifications of the steel coil, predefine the judgment rules and build a set of rule judgment models. The rule judgment models include iron loss level judgment model, appearance level classification model, grade judgment model and comprehensive quality level judgment model. S103, Iron loss level determination: First, the received iron loss value is processed to obtain the iron loss value of each length region of the steel coil. Then, the processed iron loss value is used as a parameter to input into the iron loss level determination model to obtain the iron loss level determination result of each length region. S104, Grade Determination: Input the processed iron loss data as parameters into the grade determination model to obtain the grade determination results for each length region; S105 Appearance Grade Classification: Since the appearance grade inspection proceeds step by step from the tail to the head of the steel coil, the appearance grade data must first be reversed before being input into the appearance grade classification model to obtain the appearance grade classification results for each region. S106. Comprehensive quality grade determination: The iron loss grade and appearance grade are simultaneously input into the comprehensive quality grade determination model as parameters to obtain the comprehensive quality grade determination results and location information of each region of the steel coil. S107. Establish a two-dimensional coordinate system with the lower left corner of the steel coil head as the origin, use the comprehensive quality level as the heat map parameter, design a color configuration scheme, construct a two-dimensional matrix, and generate a heat map. S108. Map the grade, appearance grade, comprehensive quality grade, and processed iron loss value to the corresponding positions on the heat map, and display them on the heat map through mouse operation, so as to intuitively analyze and understand the distribution and correlation of various indicators.

[0033] Specifically, attached Figure 2 The method for defining the decision rules in step S102 is given below: S2-1. Define the iron loss level determination rule: The iron loss value is set as P, the lower threshold is set as P1, and the upper threshold is set as P2. If P≤P1, the iron loss level is "excellent"; if P1<P≤P2, the iron loss level is "medium"; if P>P2, the iron loss level is "poor". The setting of P1 and P2 is related to the specifications of the steel coil. S2-2. Define the appearance grade classification rules: the appearance grade is GG01 as "Genuine Product 1", GG02 as "Genuine Product 2", GG03 as "Concession 1", GG04 as "Concession 2", and GG05 as "Concession 3". S2-3. Define the grade determination rules: Set multiple thresholds, compare the iron loss value of each length region with the thresholds in turn, determine the iron loss range that falls into the range, and obtain the corresponding grade. S2-4. Define the comprehensive quality grade judgment rules: Define the judgment rules from two dimensions: iron loss grade and appearance grade. When the iron loss grade is excellent and the appearance grade is "Genuine 1", the comprehensive quality grade is "Genuine 1-1". When the iron loss grade is excellent and the appearance grade is "Genuine 2", the comprehensive quality grade is "Genuine 1-2". When the iron loss grade is medium and the appearance grade is "Genuine 1", the comprehensive quality grade is "Concession 1-1". Define the comprehensive quality grade according to the iron loss grade and appearance grade, as shown in Table 1.

[0034] Table 1

[0035] Specifically, the method for processing the iron loss value in step S103 is as follows: Let the total length of the steel coil be X, take the average value of the iron loss at the beginning and end positions of the steel coil length region as the iron loss value of that region, and traverse the received iron loss values ​​in sequence to obtain the iron loss value of each length region, as shown in Table 2.

[0036] Table 2

[0037] Specifically, the execution process and results of steps S103 and S104 are as follows: the processed iron loss values ​​are input into the iron loss level determination model and the grade determination model respectively, and the iron loss level determination results and grade determination results of each length region of the steel coil are obtained, as shown in Table 3.

[0038] Table 3

[0039] Specifically, the execution process and results of step S105 are as follows: input the processed appearance grade data into the appearance grade grading model to obtain the appearance grade grading results of each region of the steel coil, as shown in Table 4.

[0040] Table 4

[0041] Specifically, the execution process and results of step S106 are as follows: the iron loss level and appearance level are simultaneously input into the comprehensive quality judgment model to obtain the comprehensive quality level results and location information of each region of the steel coil, and the location information of all positive product regions is output separately for the calculation of the comprehensive positive product rate of the steel coil.

[0042] To ensure the accuracy of the overall quality grade results, it is necessary to simultaneously obtain the iron loss grade determination results and appearance grade classification results for the same area of ​​the steel coil. However, due to the significant difference between the iron loss interval and the appearance grade length interval (the iron loss interval is much smaller than the appearance grade length interval), multiple iron loss intervals fall within the same appearance grade length interval, and the same iron loss interval may span two different appearance grade length intervals. Therefore, it is necessary to refine the steel coil area. The specific method is as follows: taking the iron loss interval as a unit, obtain the iron loss grade determination results and appearance grade classification results for each area along the width direction of the steel coil one by one. When the iron loss interval spans two different appearance grade length intervals, divide the iron loss interval into two sub-intervals, and repeat the above process. Finally, the overall quality grade and location information of each area of ​​the steel coil are obtained, and the location information of all positive product areas is output separately.

[0043] This invention introduces the key quantitative indicator of "overall good product rate" to measure the overall quality of steel coils. The formula for calculating the overall good product rate is: Overall good product rate = Total area of ​​good products / (Total area of ​​steel coil - Total area of ​​edge damage), where the total area of ​​good products refers to the sum of the areas of all good product areas of the steel coil, the total area of ​​the steel coil refers to the area of ​​the entire steel coil, and the total area of ​​edge damage refers to the sum of the areas of all edge damage areas.

[0044] The area of ​​the genuine product area is related to its location. When the genuine product area involves upper and lower edge damage areas, the area of ​​the edge damage area needs to be subtracted from the area of ​​the genuine product area accordingly. Specifically, based on the location information of all genuine product areas provided in step S106, the location of each genuine product area is determined one by one, and the area of ​​each genuine product area is calculated one by one according to the location matching area calculation formula. Then, the total area of ​​genuine products is calculated, and finally, the overall genuine product rate is calculated.

[0045] Specifically, the method for generating the heatmap in step S107 is as follows: A two-dimensional coordinate system is established with the lower left corner of the steel coil head as the origin, with the length direction as the X-axis and the width direction as the Y-axis. Using the "Comprehensive Quality Grade" as a parameter in the heatmap, a color scheme is designed to construct a two-dimensional matrix and generate the heatmap. The "Comprehensive Authentic" and "Comprehensive Concession" grades are set to green and red respectively, and their sub-grades are distinguished using colors of the same color family but different brightness levels. The edge-damaged areas are reset to "gray".

[0046] Appendix Figure 3 A system for generating heat maps of comprehensive quality grades of steel coils based on a rule-based judgment model is provided. The system includes: a data receiving module, an iron loss grade judgment module, an appearance grade classification module, a grade judgment module, a comprehensive quality grade judgment module, a heat map generation module, and a steel coil overall quality calculation module. The data receiving module is used to receive steel coil data from the production line in real time, including iron loss data, appearance grade data, edge loss data, and steel coil specifications. The iron loss level determination module is used to obtain the iron loss level results of each length region of the steel coil according to the preset iron loss level determination rules. The appearance grade grading module is used to obtain the appearance grade grading of each area of ​​the steel coil according to the preset appearance grade grading rules. The trademark determination module is used to obtain the trademark corresponding to each region according to the preset trademark determination rules; The comprehensive quality grade determination module is used to determine the comprehensive quality grade from two dimensions, iron loss and appearance grade, according to the preset comprehensive quality grade determination rules, and to obtain the comprehensive quality grade and location information of each area of ​​the steel coil. The overall quality calculation module for the steel coil is used to calculate the overall positive rate of the entire steel coil according to the formula. The heat map generation module is used to generate a heat map using the comprehensive quality level as a parameter.

[0047] Specifically, the iron loss level determination module, appearance level grading module, grade determination module, and comprehensive quality level determination module all support custom determination rules, facilitating manual review to verify the rationality of the determination rules. If any unreasonable situations are found, the determination threshold can be modified to validate the rule determination model, thereby improving the accuracy of the comprehensive quality determination of the steel coil. The system allows users to select and view any area on the heatmap using a mouse, and instantly displays the corresponding grade, appearance level, comprehensive quality rating, and iron loss value for that area.

Claims

1. A method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based judgment model, characterized in that: The method includes the following steps: S1. Receive steel coil data from the production line in real time. The steel coil data includes iron loss data, appearance grade data, edge loss data, and steel coil specifications. The iron loss data refers to the iron loss values ​​at W locations continuously detected along the length of the steel coil. The appearance grade data refers to the appearance grade obtained by continuously detecting all areas after dividing the length of the steel coil into N segments, and each segment into several layers in the width direction. The edge loss data includes the upper edge loss width and the lower edge loss width. The steel coil specifications include the steel coil type, steel coil number, and steel coil thickness. S2. Based on the specifications of the steel coil, predefine the judgment rules and build a set of rule judgment models. The rule judgment models include iron loss level judgment model, appearance level classification model and comprehensive quality level judgment model. S3. Iron loss level determination: First, the received iron loss value is processed to obtain the iron loss value of each length region of the steel coil. Then, the processed iron loss value is used as a parameter to input into the iron loss level determination model to obtain the iron loss level determination result of each length region. S4. Appearance grade classification: Since the appearance grade inspection proceeds from the tail to the head of the steel coil, the appearance grade data must first be reversed before being input into the appearance grade classification model to obtain the appearance grade classification results for each area. S5. Comprehensive quality grade determination: The iron loss grade and appearance grade are simultaneously input into the comprehensive quality grade determination model as parameters to obtain the comprehensive quality grade determination results and location information of each region of the steel coil. S6. Establish a two-dimensional coordinate system with the lower left corner of the steel coil head as the origin, use the comprehensive quality level as the heat map parameter, design a color configuration scheme, construct a two-dimensional matrix, and generate a heat map. S7. Map the processed iron loss value, appearance grade, and overall quality grade to the corresponding positions on the heat map, and display them on the heat map through mouse operation, so as to intuitively analyze and understand the distribution and correlation of each indicator.

2. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 1, characterized in that: In step S1, the appearance grade is divided into five levels: GG01, GG02, GG03, GG04, and GG05 by comprehensively evaluating the surface insulation resistance, unevenness, color difference, appearance, and coating quality of the steel coil.

3. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 2, characterized in that: The rule determination model in step S2 also includes a grade determination model, where the grade is an identifier for steel coils and is named according to the range of iron loss.

4. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 3, characterized in that: The specific method for defining the determination rule in step S2 is as follows: S201. Define the iron loss level determination rule: The iron loss value is set as P, the lower threshold is set as P1, and the upper threshold is set as P2. If P≤P1, the iron loss level is "excellent"; if P1<P≤P2, the iron loss level is "medium"; if P>P2, the iron loss level is "poor". P1 and P2 are set according to the steel coil specifications and enterprise standards. S202. Define the appearance grade classification rules: appearance grades GG01 and GG02 are "genuine products", and appearance grades GG03 to GG05 are "concessions". S203. Define grade determination rules: Set multiple thresholds, compare the iron loss value with the thresholds in sequence, determine the iron loss range that falls into, and obtain the corresponding grade. S204. Define the comprehensive quality grade judgment rules: Define the judgment rules from two dimensions: iron loss grade and appearance grade. When the iron loss grade is excellent and the appearance grade is genuine, the comprehensive quality grade is "comprehensive genuine". All others are "comprehensive concession".

5. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 4, characterized in that: In step S202, the appearance grade classification can be refined, and the determination conditions are: appearance grade GG01 is "Genuine 1", GG02 is "Genuine 2", GG03 is "Concession 1", GG04 is "Concession 2", and GG05 is "Concession 3". Correspondingly, in step S204, the determination conditions are: when the iron loss grade is excellent and the appearance grade is "Genuine 1", the comprehensive quality grade is "Genuine 1-1"; when the iron loss grade is excellent and the appearance grade is "Genuine 2", the comprehensive quality grade is "Genuine 1-2". The comprehensive quality grade is defined sequentially according to the iron loss grade and the appearance grade classification.

6. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 4 or 5, characterized in that: The specific process for determining the overall quality level in step S5 is as follows: S501, Parameter Definition and Initialization Define the total length of the steel coil as X and the total width as Y; For the iron loss interval, let X i-1 X i Let represent the left and right boundary lengths of the i-th iron loss interval, respectively, where i is a positive integer and satisfies 1≤i≤W-1; For the length range of appearance grade, let L j-1 L j Let represent the left and right boundary lengths of the j-th appearance level length interval, respectively, where j is a positive integer and satisfies 1≤j≤N-1; For the appearance grade width range, let Y m-1 Y m Let m and m represent the lower and upper boundaries of the m-th appearance level width interval, respectively, where m is a positive integer, 1≤m≤M-1; During initialization, X0=L0=0 and X1<L1 are set, and it is determined that the first iron loss interval completely falls within the first appearance level length interval. S502, Loop Decision and Iterative Processing Next, set m=0 and, based on the pre-built rule-based judgment model, determine the comprehensive quality level of the first appearance grade width range under the first iron loss interval; then, increment m sequentially to complete the comprehensive quality level judgment of all layered areas under the first iron loss interval; after completing the judgment, exit the current loop and increment i sequentially to i+1, if X i-1 ≥L j-1 And X i ≤L j If the iron loss range falls entirely within the length range of the corresponding appearance grade, repeat the above process; if X i-1 <L j <X i To determine that the current iron loss interval spans two different appearance grade length intervals, the current iron loss interval needs to be subdivided into two sub-intervals: (X i-1 L j ) and (L j X i Then, perform a comprehensive quality level determination on each of the two sub-intervals, and exit the current loop after completion. Repeat the above loops and iterations for i, j, and m in turn until all areas of the steel coil have been determined; S503. Result Output and Recording Finally, the overall quality grade results and corresponding location information for all regions of the steel coil are recorded to generate a heat map. Simultaneously, the location information for all positive product regions is output separately, including X... i-1 X i L j Y m-1 Y m It is used to calculate the overall quality rate of steel coils.

7. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 6, characterized in that: The overall quality rate is an important quantitative indicator for measuring the overall quality of steel coils. The formula for calculating the overall quality rate is as follows: Overall Validity Rate = Total Validity Area / (Total Steel Coil Area - Total Edge Damage Area), where the total validity area refers to the sum of the areas of all valid areas of the steel coil, the total steel coil area refers to the area of ​​the entire steel coil, and the total edge damage area refers to the sum of the areas of all edge damage areas. Based on the location information corresponding to all valid areas output in step S503, the location of the overall valid area is determined one by one. When the valid area involves upper and lower edge damage areas, the area of ​​the valid area needs to be reduced accordingly.

8. The method for generating a heat map of the comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 6, characterized in that: The color configuration scheme in step S6 is as follows: S601. Differentiated color configurations are implemented for each overall quality level, and corresponding color values ​​are assigned. Among them, "overall genuine product" and "overall concession" are distinguished by contrasting colors, and their sub-levels are distinguished by colors of the same color family but different brightness. S602. Configure the color of the upper and lower edge damage of the steel coil: Based on the received edge damage data, calculate the location of all edge damage areas and reset the edge damage areas to "gray".

9. A heat map generation system for comprehensive quality grade of steel coils based on a rule-based judgment model, characterized in that: The system includes the following modules: Data receiving module: used to receive iron loss data, appearance grade data, edge loss data, and specification data in real time; Iron loss level determination module: used to determine the iron loss level of steel coils in different length regions; Appearance grade classification module: used to classify the appearance grade of different areas of the steel coil; Grade determination module: used to determine the grade of steel coils in different length regions; Comprehensive Quality Rating Module: Used to determine the comprehensive quality rating of each area of ​​the steel coil; Heatmap generation module: Used to generate heatmaps in real time using visualization technology with the overall quality level as a parameter; Overall quality calculation module for steel coils: This module calculates the overall quality rate of the entire steel coil based on a formula.

10. The heat map generation system for comprehensive quality grade of steel coils based on a rule-based decision model as described in claim 9, characterized in that: The iron loss level determination module, appearance level classification module, grade determination module, and comprehensive quality level determination module all support custom determination rules. The system allows users to select and view any area on the heat map using mouse operations, and instantly displays the grade, appearance level, comprehensive quality rating, and iron loss value corresponding to that area.

Images