Broken belt prediction and management and control method adaptive to control of eighteen-high rolling mill
By performing state detection and optimization processing on the eighteen-roll mill equipment, identifying strip break data is generated, critical parameters of strip breaking control are analyzed and iteratively updated, the problem of strip breaking of the equipment in the high-speed production state is solved, and production stability is improved Product quality.
Patent Information
- Application Number
- CN202410968299.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-18
AI Technical Summary
The eighteen-roller mill is difficult to effectively predict and prevent the problem of strip breaking in high-speed production. The existing technology has a single detection variable, low detection accuracy, and fails to cover all control characteristics of the equipment, and ignores the influence law of historical data.
By performing equipment status detection and repair optimization processing on the 18-roll mill equipment, identifying strip break data is generated, critical parameters of strip break control are analyzed and iteratively updated, and abnormal fluctuations in the production process are monitored and controlled in real time to ensure that the equipment is working optimally Improve production efficiency and product quality.
It significantly improves the stability of the production process and the consistency of product quality, reduces production interruptions and maintenance costs, enhances the dynamic adaptability of the equipment, and improves the accuracy and prediction accuracy of the interruption belt detection.
Smart Images

Figure CN118904933A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data management, and in particular to a strip break prediction and control method suitable for eighteen-roll mill control. Background Art
[0002] As a transitional machine between the conventional four-six-roller machine and the twenty-roller multi-roller machine, the eighteen-roller mill combines the former's easy roll replacement and maintenance characteristics with the latter's small roll diameter and strong thinning ability, which is conducive to the rolling processing of stainless steel, electrical steel, high-strength steel and other plate and strip products, and has become the main machine for the stable production of the above steel grades. Although the above technical advantages have reduced the production cost of this model and improved the production quality of the corresponding products, it is still difficult to deal with the problem of belt breaking under high-speed production conditions, that is, the relevant technical means cannot provide solutions from the perspective of belt breaking risk prevention. The diagnosis and analysis of belt breaking during the production of the eighteen-roller mill still needs to use the conventional belt breaking investigation method and treatment plan, and a targeted belt breaking treatment method suitable for the eighteen-roller mill has not yet been formed. However, the traditional belt breaking prediction and control method of the roller mill control has a single detection variable, which is difficult to cover all products put into production by this type of machine; and the detection variable is difficult to cover the control characteristics of the equipment, which makes the detection accuracy of the belt breaking detection low; and the detection logic is only for the real-time monitoring data, but ignores the influence of historical data, and fails to consider the periodic changes of the equipment, resulting in poor belt breaking detection results. Summary of the invention
[0003] Based on this, the present invention provides a method for predicting and controlling strip breakage that is suitable for the control of an eighteen-roll rolling mill to solve at least one of the above-mentioned technical problems.
[0004] To achieve the above purpose, a belt break prediction and control method adapted for eighteen-high rolling mill control comprises the following steps:
[0005] Step S1: performing equipment status detection and repair optimization processing on the 18-high rolling mill equipment to obtain the optimized 18-high rolling mill equipment;
[0006] Step S2: executing strip steel production operation based on the optimized 18-high rolling mill equipment; analyzing the strip break data according to the strip steel production operation to generate strip break data;
[0007] Step S3: analyzing the critical parameters of the strip breaking control based on the marked strip breaking data, and generating the critical parameters of the strip breaking control; executing the real-time operation of strip steel production based on the optimized 18-high rolling mill equipment; performing abnormal fluctuation control processing on the real-time operation of strip steel production based on the critical parameters of the strip breaking control, so as to obtain the real-time data of strip steel production;
[0008] Step S4: performing iterative update operation of critical parameters for strip break control according to real-time data of strip production.
[0009] The present invention performs equipment status detection and repair optimization processing on the 18-roller rolling mill equipment. Through regular status detection and repair, it is ensured that each component of the rolling mill is in the best working state, and the overall performance of the equipment is improved, including the centering accuracy of the roller shaft, the flatness and finish of the roller surface, and the key performance indicators such as the response speed. The optimized equipment is more reliable, and the production interruption caused by equipment failure is reduced, which ensures the stable operation of the production line and improves the production efficiency. Using the optimized 18-roller rolling mill equipment to perform strip production operations can greatly improve the production efficiency and the quality of strip products. Since the state of the rolling mill equipment has been optimized, it is ensured that the thickness and flatness of the strip are more accurately controlled during the production process, which directly affects the quality of the final product. In addition, by analyzing the strip production operation and generating the strip breaking data, the risk factors that cause the strip breaking in the production process can be effectively identified, and data support is provided for the subsequent strip breaking prevention and control, which not only reduces the risk of production interruption, but also reduces the maintenance cost and improves the overall operation efficiency of the production line. Based on the identification of broken strip data, the critical parameters of broken strip control are deeply analyzed and the corresponding parameters are generated. This supports the real-time monitoring and control of abnormal fluctuation data in the production process, and can accurately identify the conditions that may cause broken strips in the production process, and then prevent the occurrence of broken strips by adjusting the production parameters in real time. In addition, the optimized 18-roller rolling mill equipment is used to perform real-time operations of strip production, and abnormal fluctuation control is carried out based on the critical parameters of broken strip control, which significantly improves the stability of the production process and ensures the consistency of product quality. By performing iterative update operations on the critical parameters of broken strip control for real-time data of strip production, the dynamic adaptability of the 18-roller rolling mill equipment in producing strip steel is enhanced, so that the production control system can adapt to changes in the production process more flexibly. This iterative update mechanism ensures that the production parameters are always optimized based on the latest data, thereby continuously improving production efficiency and product quality.
[0010] Preferably, the equipment status detection and repair optimization processing includes spatial accuracy detection and repair processing, actuator detection and repair processing, and sensor detection and optimization processing at all levels, and step S1 includes the following steps:
[0011] Step S11: Performing spatial accuracy detection and repair processing on the 18-high rolling mill equipment to obtain the 18-high rolling mill equipment with spatial accuracy repair;
[0012] Step S12: performing an actuator detection and repair process on the 18-high rolling mill equipment with spatial accuracy repair to obtain an 18-high rolling mill equipment with an effective actuator;
[0013] Step S13: Perform sensor detection optimization processing at various levels on the 18-high rolling mill equipment with effective actuators to obtain the optimized 18-high rolling mill equipment.
[0014] The present invention ensures the correct alignment and coordinated work of each component of the equipment by accurately detecting and repairing the spatial accuracy of the 18-roller rolling mill equipment as necessary, thereby significantly improving the processing accuracy and finished product quality of the rolling mill. The improvement of spatial accuracy is particularly critical for the production of strip steel with high precision requirements, which can ensure the dimensional stability and surface flatness of the product and reduce material waste in the production process. After the spatial accuracy is repaired, the detection and repair of the actuator of the 18-roller rolling mill equipment further ensures the operating efficiency and reliability of the rolling mill. The good state of the actuator is the key to ensuring continuous production and reducing downtime for maintenance. By ensuring the effectiveness of the actuator, the stability and response speed of the production line are improved, which has a direct impact on improving production efficiency and reducing operating costs. Sensors are an indispensable component of modern rolling mill equipment. They are responsible for real-time monitoring of equipment status and key parameters in the production process. By detecting and optimizing these sensors, the accuracy and reliability of data acquisition are greatly improved. The optimized sensor system ensures fine control of the production process, which is crucial for improving product quality, reducing production abnormalities and improving the level of automation.
[0015] Preferably, step S11 includes the following steps:
[0016] Based on the preset wear state detection sequence data, the equipment wear state data of the 18-high rolling mill is collected to generate the equipment wear state data;
[0017] Use gyroscope equipment to detect the rotation angle deviation state of the roll axis of the 18-high rolling mill equipment and generate roll axis state data;
[0018] The laser tracker is used to detect the position deviation state of the transmission roller axis of the 18-high rolling mill equipment, and the transmission roller axis position deviation data is generated;
[0019] Perform equipment spatial accuracy repair analysis and processing based on equipment wear status data, roller axis status data, and transmission roller axis position offset data to generate equipment spatial accuracy repair data;
[0020] Based on the equipment spatial accuracy repair data, the eighteen-roller rolling mill equipment is spatially accurate repaired to obtain the eighteen-roller rolling mill equipment with spatial accuracy repair.
[0021] The present invention collects wear state data of the equipment based on preset wear state detection sequence data, accurately identifies the wear degree and wear parts of the equipment, helps to timely discover and repair wear problems, thereby improving the overall accuracy of the equipment and extending its service life. Gyroscope equipment and laser tracker are used to detect the rotation angle offset and position offset state of key components of the rolling mill, such as the roller system axis and the transmission roller axis, to ensure the accurate correction of these components. These precision correction measures directly affect the stability of the rolling process and the quality of the final product, and are crucial to optimizing the operating performance of the equipment. Through the equipment spatial accuracy repair analysis and processing based on various status data, the equipment structure is accurately adjusted and optimized to ensure the optimal position and state of each component in the production process, reduce material waste caused by equipment accuracy problems, and improve the operating efficiency and product output speed of the production line. By systematically detecting and repairing the spatial accuracy problems of the equipment, the uniform thickness and surface flatness of the material in the production process are ensured to meet high standards of product quality requirements.
[0022] Preferably, step S12 comprises the following steps:
[0023] The main roll system drive test processing is carried out according to the 18-high rolling mill equipment repaired with spatial accuracy to obtain the main roll system drive test data;
[0024] The main roll system drive test data is analyzed for the main roll system drive test compliance data by using a preset main roll system drive threshold, including: when the main roll system drive test data is not less than the main roll system drive threshold, the 18-roller rolling mill equipment corresponding to the main roll system drive test data and whose spatial accuracy is repaired is marked as a main roll system effective drive, so as to obtain the 18-roller rolling mill equipment with effective main roll system drive; when the main roll system drive test data is less than the main roll system drive threshold, the 18-roller rolling mill equipment corresponding to the main roll system drive test data and whose spatial accuracy is repaired is marked as a main roll system invalid drive, so as to obtain the 18-roller rolling mill equipment with invalid main roll system drive;
[0025] Feedback the 18-high rolling mill equipment with invalid main roll drive to the terminal to perform maintenance work on the press-down device model;
[0026] Performing a support roller system drive test process on an 18-high rolling mill device that effectively drives the main roller system to obtain support roller system drive test data;
[0027] The support roller system drive test data is analyzed for support roller system drive compliance data using a preset support roller system drive threshold, including: when the support roller system drive test data is not less than the support roller system drive threshold, the 18-roller rolling mill equipment whose main roller system corresponding to the support roller system drive test data is effectively driven is marked as a support roller system effective drive, so as to obtain the 18-roller rolling mill equipment with effective actuator; when the support roller system drive test data is less than the support roller system drive threshold, the 18-roller rolling mill equipment whose main roller system corresponding to the support roller system drive test data is effectively driven is marked as a support roller system invalid drive, so as to obtain the 18-roller rolling mill equipment with invalid support roller system drive;
[0028] Feedback the invalid drive of the main roll system of the eighteen-high mill equipment to the terminal to perform maintenance work on the support roll system.
[0029] The present invention tests and analyzes the driving capabilities of the main roll system and the support roll system to ensure that the driving performance of these key components reaches a preset threshold, thereby ensuring the reliability and stability of the rolling mill in the production process, timely discovering potential problems of the drive system, preventing failures in the production process, and reducing unplanned downtime. The main roll system and the support roll system of the rolling mill are the key to ensuring the flatness and thickness accuracy of the strip steel. Ensuring the effective driving of these roll systems can maintain the uniformity and consistency of the material during the production process, which directly affects the quality of the final product. Comparing the drive test data with the preset threshold, and marking and feedback of the actuators that do not meet the standards, it is helpful to quickly and accurately identify the components that need to be repaired or adjusted, and can guide maintenance personnel to perform targeted maintenance operations to improve maintenance efficiency and effectiveness. Through regular drive testing and timely repair and maintenance, the wear and damage of the equipment are effectively reduced, and the service life of the rolling mill and related actuators is extended. This preventive maintenance reduces long-term maintenance costs and improves the overall return on investment of the equipment. The 18-high rolling mill equipment with effective actuators ensures the continuity and efficiency of the production process, and achieves higher production efficiency and lower operating costs by reducing production interruptions caused by actuator problems.
[0030] Preferably, step S13 comprises the following steps:
[0031] Collect sensor accuracy data at all levels for the 18-high rolling mill equipment with effective actuators and generate sensor accuracy data at all levels;
[0032] Perform sensor accuracy repair data analysis at all levels based on sensor accuracy data at all levels, and generate sensor accuracy repair data at all levels;
[0033] According to the precision repair data of sensors at all levels, the precision repair and optimization processing of sensors at all levels of the 18-roller rolling mill equipment with effective actuators is carried out to obtain the optimized 18-roller rolling mill equipment.
[0034] The present invention collects and repairs precision data of sensors at all levels, ensuring that the sensor's measurement data is more accurate and improving the reliability of data monitoring, which is crucial for the precise control and operation of the rolling mill, because the sensor data directly affects the rolling mill's control system. The precision-repaired sensors can provide more accurate real-time data, allowing the rolling mill's control system to operate based on more reliable information, reducing production problems caused by sensor failure or reduced precision.
[0035] Preferably, step S2 comprises the following steps:
[0036] Step S21: executing a strip steel production operation based on the optimized 18-high rolling mill equipment, and collecting strip steel production data according to the strip steel production operation to generate strip steel production data;
[0037] Step S22: extracting and processing the broken strip data according to the strip production data to generate the broken strip data;
[0038] Step S23: monitoring equipment status change parameters according to the strip steel production operation and generating equipment change parameters;
[0039] Step S24: using the equipment change parameters to perform change information identification processing on the broken tape data to generate marked broken tape data.
[0040] The present invention ensures the comprehensiveness and accuracy of the data by collecting strip production data during the production process. It not only includes the basic information of strip production, but also covers various indicators that are crucial to quality control, and is the basis for subsequent analysis and optimization. Extracting broken strip data from production data can accurately identify the causes and patterns of broken strips, which is crucial for formulating corresponding preventive measures. Equipment status change parameter monitoring provides real-time information about the operating status of the equipment, which helps to promptly discover potential equipment failure problems, helps to optimize maintenance plans and arrange repair work in advance to avoid production interruptions. It is very effective. Combining equipment change parameters with broken strip data and performing identification processing on the change information further deepens the identification of the cause of broken strips, which not only helps to identify the association between broken strips and specific equipment status changes, but also promotes the formulation of targeted improvement measures.
[0041] Preferably, the analysis of critical parameters of belt break control based on the belt break identification data in step S3 includes the following steps:
[0042] Step S301: collecting the tape-breaking index data for the tape-breaking data according to the preset tape-breaking index type to generate the tape-breaking index data;
[0043] Step S302: performing data normalization processing on the belt-breaking index data to generate normalized belt-breaking index data;
[0044] Step S303: performing belt-break event correlation scoring processing on the normalized belt-break index data according to a preset isolation forest algorithm to generate belt-break event correlation scoring data;
[0045] Step S304: performing dimension reduction screening processing on the normalized belt-break index data based on the belt-break event correlation score data to generate screened belt-break index data;
[0046] Step S305: performing belt break control critical parameter analysis based on the screened belt break index data to generate belt break control critical parameters.
[0047] The present invention collects detailed belt-breaking index data for the belt-breaking data, and accurately identifies various potential factors that cause belt-breaking. Data normalization ensures that data from different sources and different magnitudes are compared and analyzed under the same standard, which improves the efficiency of data processing and the accuracy of analysis. The isolated forest algorithm is applied to score the correlation of belt-breaking events, and data nodes that are highly correlated with belt-breaking events are effectively identified. The application of this machine learning method makes the prediction of belt-breaking risks more accurate and efficient. Through dimensionality reduction and screening, only indicator data with higher influence are retained, which simplifies the monitoring and management of belt-breaking risks, not only improves the processing speed, but also improves the prediction accuracy of the monitoring and management of belt-breaking risks. According to the screened belt-breaking index data, in-depth analysis is carried out, and the generated critical parameters for belt-breaking control provide an accurate basis for real-time monitoring and early warning in the production process. These parameters help production managers to formulate preventive measures more effectively and reduce the occurrence of belt-breaking events.
[0048] Preferably, step S305 includes the following steps:
[0049] Perform indicator sensitivity assessment on the screened fault zone indicator data according to the preset random forest algorithm to generate indicator sensitivity assessment data;
[0050] According to the index sensitivity evaluation data, oversensitive index selection processing is performed on the screened broken belt index data to generate oversensitive broken belt index data;
[0051] Performing numerical distribution analysis on the oversensitive belt-breaking index data to generate oversensitive belt-breaking index numerical distribution data;
[0052] Based on the normal distribution algorithm, the critical parameters of the over-sensitive belt-breaking index are analyzed for the numerical distribution data of the over-sensitive belt-breaking index to generate the critical parameters of the belt-breaking control.
[0053] The present invention uses a random forest algorithm to perform a sensitivity assessment on the selected strip-breaking indicators, which can more accurately identify which indicators have a significant impact on strip-breaking events, and concentrate resources and attention on those strip-breaking indicators that are most likely to cause production problems. Select oversensitive indicators and perform numerical distribution analysis on them, and deeply understand how the data subset of oversensitive indicators affects the production process of strip steel, so as to formulate more effective preventive measures and improve the pertinence and efficiency of measures to prevent strip breaking in the production process of strip steel. Based on the normal distribution algorithm, the numerical distribution of oversensitive indicators is analyzed for critical parameters of strip-breaking control, which reflects how operations should be adjusted in the production process to avoid strip breaking, which helps to adjust production strategies in real time, reduce the risk of strip breaking in production, and better predict and manage potential risks in the production process, thereby avoiding unexpected downtime and production losses.
[0054] Preferably, the abnormal fluctuation control processing of the real-time operation of strip production based on the critical parameters of strip break control in step S3 includes the following steps:
[0055] According to the critical parameters of strip break control, the real-time operation of strip production is subjected to abnormal fluctuation control, including: when the over-sensitive strip break index data of the real-time operation of strip production is not within the range of the critical parameters of strip break control, the real-time operation of strip production is stopped; when the over-sensitive strip break index data of the real-time operation of strip production is within the range of the critical parameters of strip break control, the real-time operation of strip production is continued to obtain the real-time data of strip production.
[0056] The present invention can timely identify conditions that may lead to production accidents by real-time monitoring of oversensitive belt-breaking index data and comparing it with the critical parameters of belt-breaking control. Once it is found that these indicators exceed the safe range, the production operation is stopped immediately, thereby preventing potential production accidents and reducing resource waste. It ensures that the oversensitive belt-breaking index data is always within the safe range of the critical parameters of belt-breaking control, which helps to maintain the stability and continuity of the production process. By preventing production interruptions caused by abnormal fluctuations, unnecessary losses caused by shutdown repairs and waste generation are reduced, material utilization and production efficiency are improved, and production costs are reduced. Continuously monitor and adjust the production process in a timely manner to ensure that production activities are carried out under optimal conditions, and continuously optimize the critical parameters of belt-breaking control based on iterative updates of subsequent steps to form a positive feedback loop, continuously improve the intelligence and automation level of the production process, and the real-time data of strip production obtained ensures the consistency and high standards of product quality.
[0057] Preferably, the iterative updating operation of the critical parameters for belt break control in step S4 includes the following steps:
[0058] Real-time strip breakage data is monitored and processed for the real-time data of strip production, including: when the real-time data of strip production does not monitor the real-time strip breakage data, the critical parameters of strip breakage control are fed back to the terminal to execute the establishment operation of the strip breakage prediction mechanism; when the real-time data of strip production monitors the real-time strip breakage data, the real-time segment strip data is used to iteratively update the strip breakage data to generate updated strip breakage data, and the updated strip breakage data is transmitted to step S22 to execute the iterative update of the critical parameters of strip breakage control.
[0059] When the real-time data does not monitor the belt breakage, the present invention maintains the current critical parameters for belt breakage control to ensure the stability of the production process; when the real-time belt breakage data is monitored, the belt breakage data and the critical parameters for control are updated in time, and the adjusted critical parameters for control take into account the influence of historical data. This dynamic adjustment mechanism enables the belt breakage prediction mechanism to adapt to changes in the production process. For units that are put into production for a long period of time, factors such as spatial accuracy, equipment wear status, incoming material size and performance will change periodically with the rolling process, which will also affect the changes in the abnormal fluctuation rules of cold-rolled belt breakage parameters in different production cycles. By continuously collecting and analyzing real-time data in the production process and continuously performing iterative updates, deficiencies and potential improvement points in the production process are identified, providing data support for continuous improvement of the production process to adapt to changes in production conditions and ensure high quality and stability of the production process.
[0060] The beneficial effect of the present application lies in that the present invention utilizes the actual production data of the cold rolling production line (specifically, the eighteen-roll mill cold rolling production line), and extracts the historical broken strip data under different steel grades, different sizes and specifications, and different rolling conditions from a large amount of data, and uses this as a broken strip sensitive data set; then, with the help of isolation forest algorithm and random forest algorithm, the sensitivity relationship between the cold rolling process data under different conditions and the broken strip phenomenon is analyzed; then, focusing on different steel grades, different sizes and specifications, and different rolling conditions, the changes in the broken strip sensitive data are observed in a targeted manner; once abnormal fluctuations in the sensitivity data in actual production are found, the main drive motor can be mobilized to implement a shutdown operation, thereby avoiding broken strips. Through multi-dimensional detection variables, it is possible to cover all products of this type of machine and accurately determine whether various products have the risk of strip breakage; and the detection variables of the main roll system and the side support roll system completely cover the control characteristics of the 18-high rolling mill equipment, making the detection accuracy of strip breakage detection higher; and the detection logic is not only for real-time monitoring data, but also for the influencing rules of historical data, including considering the spatial accuracy of the equipment, equipment wear status, incoming material size and performance and other factors will undergo periodic changes with the rolling process, so that the strip breakage detection results are excellent in different production cycles. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] Figure 1It is a schematic diagram of the steps of a method for predicting and controlling a broken strip adapted for the control of an 18-high rolling mill according to the present invention;
[0062] Figure 2 for Figure 1 Detailed implementation steps of step S3 interruption band control critical parameter analysis flow chart;
[0063] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0064] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.
[0065] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.
[0066] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.
[0067] To achieve this, please refer to Figure 1 to Figure 2 The present invention provides a method for predicting and controlling a broken strip suitable for the control of an 18-high rolling mill, comprising the following steps:
[0068] Step S1: performing equipment status detection and repair optimization processing on the 18-high rolling mill equipment to obtain the optimized 18-high rolling mill equipment;
[0069] Step S2: executing strip steel production operation based on the optimized 18-high rolling mill equipment; analyzing the strip break data according to the strip steel production operation to generate strip break data;
[0070] Step S3: analyzing the critical parameters of the strip breaking control based on the marked strip breaking data, and generating the critical parameters of the strip breaking control; executing the real-time operation of strip steel production based on the optimized 18-high rolling mill equipment; performing abnormal fluctuation control processing on the real-time operation of strip steel production based on the critical parameters of the strip breaking control, so as to obtain the real-time data of strip steel production;
[0071] Step S4: performing iterative update operation of critical parameters for strip break control according to real-time data of strip production.
[0072] The present invention performs equipment status detection and repair optimization processing on the 18-roller rolling mill equipment. Through regular status detection and repair, it is ensured that each component of the rolling mill is in the best working state, and the overall performance of the equipment is improved, including the centering accuracy of the roller shaft, the flatness and finish of the roller surface, and the key performance indicators such as the response speed. The optimized equipment is more reliable, and the production interruption caused by equipment failure is reduced, which ensures the stable operation of the production line and improves the production efficiency. Using the optimized 18-roller rolling mill equipment to perform strip production operations can greatly improve the production efficiency and the quality of strip products. Since the state of the rolling mill equipment has been optimized, it is ensured that the thickness and flatness of the strip are more accurately controlled during the production process, which directly affects the quality of the final product. In addition, by analyzing the strip production operation and generating the strip breaking data, the risk factors that cause the strip breaking in the production process can be effectively identified, and data support is provided for the subsequent strip breaking prevention and control, which not only reduces the risk of production interruption, but also reduces the maintenance cost and improves the overall operation efficiency of the production line. Based on the identification of broken strip data, the critical parameters of broken strip control are deeply analyzed and the corresponding parameters are generated. This supports the real-time monitoring and control of abnormal fluctuation data in the production process, and can accurately identify the conditions that may cause broken strips in the production process, and then prevent the occurrence of broken strips by adjusting the production parameters in real time. In addition, the optimized 18-roller rolling mill equipment is used to perform real-time operations of strip production, and abnormal fluctuation control is carried out based on the critical parameters of broken strip control, which significantly improves the stability of the production process and ensures the consistency of product quality. By performing iterative update operations on the critical parameters of broken strip control for real-time data of strip production, the dynamic adaptability of the 18-roller rolling mill equipment in producing strip steel is enhanced, so that the production control system can adapt to changes in the production process more flexibly. This iterative update mechanism ensures that the production parameters are always optimized based on the latest data, thereby continuously improving production efficiency and product quality.
[0073] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a method for predicting and controlling a broken strip in an 18-high rolling mill control according to the present invention. In this embodiment, the method for predicting and controlling a broken strip adapted for controlling an 18-high rolling mill comprises the following steps:
[0074] Step S1: performing equipment status detection and repair optimization processing on the 18-high rolling mill equipment to obtain the optimized 18-high rolling mill equipment;
[0075] In the embodiment of the present invention, the spatial accuracy detection is first performed on the eighteen-roll mill equipment, including: using a laser tracker, a total station and other devices to detect the wear state of the assembly position in the eighteen-roll mill frame, and recording the wear amount of each roller liner on the transmission side and the operating side in the order of "upper support roller-upper intermediate roller-lower intermediate roller-lower support roller"; using a gyroscope to detect the state of the roller axis (operation side offset, transmission side offset), including the main roller and side support roller system; using a laser tracker to detect the axis state of each transmission roller on the left and right process platforms (operation side offset, transmission side offset) to obtain spatial accuracy monitoring data. According to the spatial accuracy monitoring data, the eighteen-roll mill equipment with abnormal spatial accuracy detection is repaired for abnormal spatial accuracy to obtain the eighteen-roll mill equipment after spatial accuracy repair. After confirming that the spatial accuracy of the 18-rolling mill equipment is correct, the main roll system actuator is tested, including driving the down-pressing device to check whether the main roll system is driven from the negative stroke position to the positive stroke position, and comparing whether the design capacity of the main roll system of the rolling mill is consistent with the actual measurement. If it is not consistent, the down-pressing device is repaired; similar stroke volume detection is also performed on the side support roll system to ensure that its stroke volume meets the design standard. If it does not meet the design standard, the side support roll system is repaired. Accuracy detection is performed on sensors at all levels of the rolling mill, including rolling force sensors, tension meters on the left and right process platforms, various channels of plate-shaped rollers, and thickness gauges, to ensure the accuracy of data collected during the rolling process. For sensor equipment that does not meet the data accuracy, sensor maintenance optimization is performed to maintain the high quality and high stability of the production process, and finally the optimized 18-roller rolling mill equipment is obtained.
[0076] Step S2: executing strip steel production operation based on the optimized 18-high rolling mill equipment; analyzing the strip break data according to the strip steel production operation to generate strip break data;
[0077] In the embodiment of the present invention, an optimized 18-high rolling mill is used to perform a strip production operation of 500 strips, including preparing production materials, setting rolling mill parameters, and starting the production process. Relevant production data, such as rolling force, tension, rolling speed, temperature and other parameters, are collected in real time during the production operation. From the strip production operation, production data specific to each strip is collected, including but not limited to rolling parameters, material properties, production speed and time. According to the collected strip production data, strip break data extraction processing is performed. This process involves analyzing the signs in the production data that may indicate the risk of strip break, such as abnormal rolling force fluctuations, tension abnormalities, etc., identifying and recording any strip break events that actually occur during the production process, including the specific time, location, relevant production parameters and causes of the strip break. During the strip production process, the variable parameters related to the equipment status, such as the degree of mill wear, the position offset of each roller, etc., are continuously monitored, and the correlation between these equipment status variable parameters and the strip break event is analyzed to identify the equipment factors that may affect the risk of strip break. The extracted broken belt data are further processed for change information identification using the monitored equipment change parameters to identify those broken belt data that are closely related to equipment status changes and generate identified broken belt data. These data should clearly indicate which production parameter changes are significantly associated with broken belt events under specific equipment status changes.
[0078] Step S3: analyzing the critical parameters of the strip breaking control based on the marked strip breaking data, and generating the critical parameters of the strip breaking control; executing the real-time operation of strip steel production based on the optimized 18-high rolling mill equipment; performing abnormal fluctuation control processing on the real-time operation of strip steel production based on the critical parameters of the strip breaking control, so as to obtain the real-time data of strip steel production;
[0079] In an embodiment of the present invention, according to the preset belt-breaking indicator type (such as rolling force, tension, abnormal fluctuation of plate shape, etc.), relevant indicator data are extracted from the marked belt-breaking data to establish a belt-breaking indicator data set. The collected belt-breaking indicator data are normalized to eliminate the influence of different dimensions and magnitudes, and ensure the accuracy and effectiveness of subsequent analysis. The preset isolation forest algorithm is applied to perform correlation scoring on the normalized belt-breaking indicator data to identify and evaluate the correlation strength between each indicator and the belt-breaking event. Based on the belt-breaking event correlation scoring data, the normalized belt-breaking indicator data is subjected to dimensionality reduction screening to eliminate indicators with weak correlation and retain the most critical indicators for belt-breaking risk prediction. According to the screened belt-breaking indicator data, a belt-breaking control critical parameter analysis is performed, which involves further statistical analysis of the screened indicators, such as using random forest to further screen oversensitive belt-breaking indicator data, which reflects the equipment parameter indicators that are more relevant to the belt-breaking, and through normal distribution analysis, the normal operating range and control critical value of each key indicator are determined. Using the defined critical parameters for strip break control, the real-time strip production operation performed by the 18-high rolling mill is monitored, and the oversensitive strip break index data is monitored in real time and compared with the preset critical parameters for strip break control. When the monitored oversensitive strip break index data is not within the safe range of the critical parameters for strip break control, the real-time strip production operation is stopped immediately to avoid potential strip break risks. When the oversensitive strip break index data is within the safe range, the real-time strip production operation continues, and the production data is collected in real time for continuous monitoring and subsequent data analysis.
[0080] Step S4: performing iterative update operation of critical parameters for strip break control according to real-time data of strip production.
[0081] In an embodiment of the present invention, the real-time operation of strip steel production is continuously monitored, and strip steel production data, especially oversensitive strip break index data, are collected and analyzed in real time. The oversensitive strip break index data collected in real time are compared with the currently set critical parameters for strip break control to determine whether there is a risk of strip break. If no real-time strip break data is detected in real-time monitoring, that is, the oversensitive strip break index data are all within the safe range of the critical parameters for strip break control, then the current strip break prediction mechanism is considered to be effective, and the existing critical parameters for strip break control are maintained; if real-time strip break data is monitored, that is, there is oversensitive strip break index data that exceeds the range of the critical parameters for control, indicating that the current strip break prediction mechanism may need to be adjusted, and the next step is entered. Using the detected real-time strip break data, the strip break data is iteratively updated, including analyzing the specific conditions for the occurrence of strip break, identifying possible new risk factors, and updating the relevant strip break index data. Based on the updated strip break data, the critical parameters for strip break control are re-evaluated and set, which involves returning to step S3 to re-analyze the system to ensure that the new critical parameters can more accurately reflect the actual risks in the production process. The updated critical parameters for strip breakage control are fed back to the monitoring system of the production line to guide the abnormal fluctuation control and processing of subsequent real-time operations of strip production. Real-time production data is continuously monitored to verify the effectiveness of the new critical parameters for strip breakage control, and further iterative updates are performed as needed.
[0082] Preferably, the equipment status detection and repair optimization processing includes spatial accuracy detection and repair processing, actuator detection and repair processing, and sensor detection and optimization processing at all levels, and step S1 includes the following steps:
[0083] Step S11: Performing spatial accuracy detection and repair processing on the 18-high rolling mill equipment to obtain the 18-high rolling mill equipment with spatial accuracy repair;
[0084] Step S12: performing an actuator detection and repair process on the 18-high rolling mill equipment with spatial accuracy repair to obtain an 18-high rolling mill equipment with an effective actuator;
[0085] Step S13: Perform sensor detection optimization processing at various levels on the 18-high rolling mill equipment with effective actuators to obtain the optimized 18-high rolling mill equipment.
[0086] The present invention ensures the correct alignment and coordinated work of each component of the equipment by accurately detecting and repairing the spatial accuracy of the 18-roller rolling mill equipment as necessary, thereby significantly improving the processing accuracy and finished product quality of the rolling mill. The improvement of spatial accuracy is particularly critical for the production of strip steel with high precision requirements, which can ensure the dimensional stability and surface flatness of the product and reduce material waste in the production process. After the spatial accuracy is repaired, the detection and repair of the actuator of the 18-roller rolling mill equipment further ensures the operating efficiency and reliability of the rolling mill. The good state of the actuator is the key to ensuring continuous production and reducing downtime for maintenance. By ensuring the effectiveness of the actuator, the stability and response speed of the production line are improved, which has a direct impact on improving production efficiency and reducing operating costs. Sensors are an indispensable component of modern rolling mill equipment. They are responsible for real-time monitoring of equipment status and key parameters in the production process. By detecting and optimizing these sensors, the accuracy and reliability of data acquisition are greatly improved. The optimized sensor system ensures fine control of the production process, which is crucial for improving product quality, reducing production abnormalities and improving the level of automation.
[0087] In the embodiment of the present invention, a laser tracker, a total station and other devices are used to detect the wear state of the assembly position in the 18-roller rolling mill equipment, and the "upper left lining wear of the upper support roller-upper right lining wear of the upper support roller-upper left lining wear of the upper intermediate roller-upper right lining wear of the upper intermediate roller-lower left lining wear of the lower intermediate roller-lower right lining wear of the lower intermediate roller" on the transmission side and the operation side are recorded in sequence in the order of "upper support roller-upper intermediate roller-lower intermediate roller-lower support roller"; a gyroscope is used to detect the axis state of the roller system after assembly (operation side offset, transmission side offset) The rollers involved include the upper support roller, the upper middle roller, the lower middle roller, the lower support roller, the upper left support roller system, the upper right support roller system, the lower left support roller system, and the lower right support roller system; the laser tracker is used to detect the axis state (operation side offset, transmission side offset) of each transmission roller of the left and right process platforms. The transmission rollers involved include the left reel, the left guide roller, the left plate-shaped roller, the right plate-shaped roller, the right guide roller, the right reel, etc. The abnormal nodes detected in the spatial accuracy detection of the 18-roller rolling mill equipment are repaired to obtain the 18-roller rolling mill equipment with spatial accuracy repair. On the basis of the 18-roller rolling mill equipment with spatial accuracy repair, the driving pressing device drives the main roller system from the negative stroke position to the positive stroke position, and compares whether the design capacity of the main roller system of the rolling mill is consistent with the actual measurement. When the negative stroke of the main roller system is driven to the positive stroke position and does not reach the preset 80% stroke distance, the drive pressing device needs to be repaired; when the negative stroke of the main roller system is driven to the positive stroke position and reaches the preset 80% stroke distance, the drive pressing device does not need to be repaired. After the side support roller system stroke inspection, the stroke of the upper left, lower left, upper right, and lower right side support roller systems reached 94%, 91%, 90%, and 86% of the design stroke respectively, and the side support roller system does not need to be repaired. If the side support roller system stroke reaches less than 80% of the design stroke, the corresponding side support roller system needs to be repaired.
[0088] Preferably, step S11 includes the following steps:
[0089] Based on the preset wear state detection sequence data, the equipment wear state data of the 18-high rolling mill is collected to generate the equipment wear state data;
[0090] Use gyroscope equipment to detect the rotation angle deviation state of the roll axis of the 18-high rolling mill equipment and generate roll axis state data;
[0091] The laser tracker is used to detect the position deviation state of the transmission roller axis of the 18-high rolling mill equipment, and the transmission roller axis position deviation data is generated;
[0092] Perform equipment spatial accuracy repair analysis and processing based on equipment wear status data, roller axis status data, and transmission roller axis position offset data to generate equipment spatial accuracy repair data;
[0093] Based on the equipment spatial accuracy repair data, the eighteen-roller rolling mill equipment is spatially accurate repaired to obtain the eighteen-roller rolling mill equipment with spatial accuracy repair.
[0094] The present invention collects wear state data of the equipment based on preset wear state detection sequence data, accurately identifies the wear degree and wear parts of the equipment, helps to timely discover and repair wear problems, thereby improving the overall accuracy of the equipment and extending its service life. Gyroscope equipment and laser tracker are used to detect the rotation angle offset and position offset state of key components of the rolling mill, such as the roller system axis and the transmission roller axis, to ensure the accurate correction of these components. These precision correction measures directly affect the stability of the rolling process and the quality of the final product, and are crucial to optimizing the operating performance of the equipment. Through the equipment spatial accuracy repair analysis and processing based on various status data, the equipment structure is accurately adjusted and optimized to ensure the optimal position and state of each component in the production process, reduce material waste caused by equipment accuracy problems, and improve the operating efficiency and product output speed of the production line. By systematically detecting and repairing the spatial accuracy problems of the equipment, the uniform thickness and surface flatness of the material in the production process are ensured to meet high standards of product quality requirements.
[0095] In an embodiment of the present invention, a laser tracker, a total station and other devices are used to detect the wear status of the assembly position in the eighteen-roll rolling mill equipment, and the wear status detection sequence data is pre-designed as follows: "upper support roller-upper intermediate roller-lower intermediate roller-lower support roller" in the order of "upper left lining wear of upper support roller-upper right lining wear of upper support roller-upper left lining wear of upper intermediate roller-upper right lining wear of upper intermediate roller-lower left lining wear of lower intermediate roller-lower right lining wear of lower intermediate roller" on the transmission side and the operating side are recorded in turn, and suitable detection tools and equipment (such as micrometers, depth gauges, etc.) are used to measure and record the degree of wear, and the degree of wear is transmitted to the inside of the system. The gyroscope is used to detect the axis state of the roller system after assembly (operation side offset, transmission side offset). The rollers involved include the upper support roller, the upper middle roller, the lower middle roller, the lower support roller, the upper left support roller system, the upper right support roller system, the lower left support roller system, and the lower right support roller system. The focus is on detecting the rotation angle and offset state of the roller system axis, analyzing the centering and parallelism of the roller system axis, ensuring that the offset of each axis relative to the ideal position is minimized, and the offset data of the roller system axis rotation angle is recorded in the system. The laser tracker is used to detect the axis state (operation side offset, transmission side offset) of each transmission roller on the left and right process platforms. The transmission rollers involved include the left reel, the left guide roller, the left plate roller, the right plate roller, the right guide roller, the right reel, etc. The focus is on detecting the horizontal and vertical position offset of the transmission roller, as well as the angle offset relative to the rolling direction, and these offset parameters are recorded in the system. The collected equipment wear status data, roller axis status data and transmission roller axis position offset data are comprehensively analyzed to identify the key factors affecting the equipment spatial accuracy, and the parameter data of the equipment spatial accuracy are repaired through the key factors affecting the equipment spatial accuracy. According to the analysis and processing results, the spatial accuracy repair of the 18-high rolling mill equipment is performed. This may include re-adjusting the roller system position, replacing severely worn parts, adjusting the transmission roller position, etc. After completing the spatial accuracy repair, necessary verification and testing work is carried out to confirm that the repair effect meets expectations and ensure that the spatial accuracy of the equipment is restored to the optimal state. The verification process may include repeating the detection work of steps S11 to S13, comparing the data differences before and after the repair, and ensuring that all key indicators meet or exceed the design standards.
[0096] Preferably, step S12 comprises the following steps:
[0097] The main roll system drive test processing is carried out according to the 18-high rolling mill equipment repaired with spatial accuracy to obtain the main roll system drive test data;
[0098] The main roll system drive test data is analyzed for the main roll system drive test compliance data by using a preset main roll system drive threshold, including: when the main roll system drive test data is not less than the main roll system drive threshold, the 18-roller rolling mill equipment corresponding to the main roll system drive test data and whose spatial accuracy is repaired is marked as a main roll system effective drive, so as to obtain the 18-roller rolling mill equipment with effective main roll system drive; when the main roll system drive test data is less than the main roll system drive threshold, the 18-roller rolling mill equipment corresponding to the main roll system drive test data and whose spatial accuracy is repaired is marked as a main roll system invalid drive, so as to obtain the 18-roller rolling mill equipment with invalid main roll system drive;
[0099] Feedback the 18-high rolling mill equipment with invalid main roll drive to the terminal to perform maintenance work on the press-down device model;
[0100] Performing a support roller system drive test process on an 18-high rolling mill device that effectively drives the main roller system to obtain support roller system drive test data;
[0101] The support roller system drive test data is analyzed for support roller system drive compliance data using a preset support roller system drive threshold, including: when the support roller system drive test data is not less than the support roller system drive threshold, the 18-roller rolling mill equipment whose main roller system corresponding to the support roller system drive test data is effectively driven is marked as a support roller system effective drive, so as to obtain the 18-roller rolling mill equipment with effective actuator; when the support roller system drive test data is less than the support roller system drive threshold, the 18-roller rolling mill equipment whose main roller system corresponding to the support roller system drive test data is effectively driven is marked as a support roller system invalid drive, so as to obtain the 18-roller rolling mill equipment with invalid support roller system drive;
[0102] Feedback the invalid drive of the main roll system of the eighteen-high mill equipment to the terminal to perform maintenance work on the support roll system.
[0103] The present invention tests and analyzes the driving capabilities of the main roll system and the support roll system to ensure that the driving performance of these key components reaches a preset threshold, thereby ensuring the reliability and stability of the rolling mill in the production process, timely discovering potential problems of the drive system, preventing failures in the production process, and reducing unplanned downtime. The main roll system and the support roll system of the rolling mill are the key to ensuring the flatness and thickness accuracy of the strip steel. Ensuring the effective driving of these roll systems can maintain the uniformity and consistency of the material during the production process, which directly affects the quality of the final product. Comparing the drive test data with the preset threshold, and marking and feedback of the actuators that do not meet the standards, it is helpful to quickly and accurately identify the components that need to be repaired or adjusted, and can guide maintenance personnel to perform targeted maintenance operations to improve maintenance efficiency and effectiveness. Through regular drive testing and timely repair and maintenance, the wear and damage of the equipment are effectively reduced, and the service life of the rolling mill and related actuators is extended. This preventive maintenance reduces long-term maintenance costs and improves the overall return on investment of the equipment. The 18-high rolling mill equipment with effective actuators ensures the continuity and efficiency of the production process, and achieves higher production efficiency and lower operating costs by reducing production interruptions caused by actuator problems.
[0104] In the embodiment of the present invention, the main roll system of the eighteen-high rolling mill equipment after the spatial accuracy repair is driven and tested to evaluate whether its driving performance meets the design requirements. The test includes but is not limited to measuring the speed, torque and power output of the main roll system to ensure that it meets the production requirements. The main roll system is driven from the negative stroke position to the positive stroke position by driving the pressing device, and the design capacity of the main roll system of the rolling mill is measured to see whether it is consistent with the actual measurement. The collected main roll system drive test data is analyzed by using the preset main roll system drive threshold (such as 80% of the pre-designed main roll system drive target stroke). If the test data is not less than the main roll system drive threshold, an effective drive mark is performed to confirm that it is an eighteen-high rolling mill equipment with effective main roll system drive. If the test data is less than the main roll system drive threshold, an invalid drive mark is performed to identify it as an invalid main roll system drive device that needs maintenance. For example, after detection, if the actual stroke of the pressing drive device or the main drive component reaches 83% of the design, there is no need to perform maintenance operations on the pressing device or the main drive component. All devices marked as invalid main roll system drives are fed back to the terminal to perform maintenance operations on the pressing device or the main drive component. The driving test of the support roller system is carried out on the 18-high rolling mill equipment marked as the effective drive of the main roller system to ensure that the support roller system can also meet the performance requirements. Similar to the main roller system, the key driving parameters such as the rotation speed and torque of the support roller system are tested. The test data is also analyzed using the preset support roller system drive threshold (such as 80% of the pre-designed support roller system drive target stroke). When the test data is not less than the threshold, it is marked as the effective drive of the support roller system; when it is less than the threshold, it is marked as the invalid drive of the support roller system, and subsequent maintenance is carried out. For example, after the side support roller system stroke inspection, the stroke of the upper left, lower left, upper right, and lower right side support roller systems reached 94%, 91%, 90%, and 86% of the design, respectively, and no support roller system maintenance is required. Equipment marked as invalid drive of the support roller system is also fed back to the maintenance team for targeted maintenance. The equipment after maintenance should be driven again to verify the repair effect and ensure that all drive components meet or exceed the preset performance standards.
[0105] Preferably, step S13 comprises the following steps:
[0106] Collect sensor accuracy data at all levels for the 18-high rolling mill equipment with effective actuators and generate sensor accuracy data at all levels;
[0107] Perform sensor accuracy repair data analysis at all levels based on sensor accuracy data at all levels, and generate sensor accuracy repair data at all levels;
[0108] According to the precision repair data of sensors at all levels, the precision repair and optimization processing of sensors at all levels of the 18-roller rolling mill equipment with effective actuators is carried out to obtain the optimized 18-roller rolling mill equipment.
[0109] The present invention collects and repairs precision data of sensors at all levels, ensuring that the sensor's measurement data is more accurate and improving the reliability of data monitoring, which is crucial for the precise control and operation of the rolling mill, because the sensor data directly affects the rolling mill's control system. The precision-repaired sensors can provide more accurate real-time data, allowing the rolling mill's control system to operate based on more reliable information, reducing production problems caused by sensor failure or reduced precision.
[0110] In the embodiment of the present invention, a comprehensive sensor accuracy test is performed on the 18-high rolling mill equipment with confirmed effective actuators, including but not limited to rolling force sensors, tension sensors, displacement sensors, speed sensors and temperature sensors. When collecting data, the model, installation location, current reading, theoretical reading (if available) and any known calibration data of each sensor are recorded. The collected sensor accuracy data is analyzed, and the actual reading of the sensor is compared with its theoretical reading or calibration data to determine its degree of deviation and accuracy status, identify the sensor whose accuracy does not meet the preset standard, and determine the sensor that needs to be calibrated or replaced. For sensors that do not meet the accuracy standards, corresponding repair or optimization processing is performed according to their type and the nature of the problem. Including but not limited to: calibration: recalibrate the sensor to ensure its accurate reading; adjustment: fine-tune the installation position or angle of the sensor to improve its performance; replacement: replace the sensor that cannot restore its accuracy through calibration or adjustment. After repairing or replacing the sensor, re-test the accuracy to verify the effect of the repair measures, ensure that all sensors have reached the required accuracy standards, and update the equipment maintenance records, including the repair, calibration and replacement information of the sensor, for future tracking and management. After the above steps, it was confirmed that the optimized 18-high rolling mill equipment was obtained, and its sensor system could accurately reflect the equipment operating status and production parameters. A comprehensive equipment function test was carried out to ensure that the overall performance of the equipment and sensor system met the production requirements.
[0111] Preferably, step S2 comprises the following steps:
[0112] Step S21: executing a strip steel production operation based on the optimized 18-high rolling mill equipment, and collecting strip steel production data according to the strip steel production operation to generate strip steel production data;
[0113] Step S22: extracting and processing the broken strip data according to the strip production data to generate the broken strip data;
[0114] Step S23: monitoring equipment status change parameters according to the strip steel production operation and generating equipment change parameters;
[0115] Step S24: using the equipment change parameters to perform change information identification processing on the broken tape data to generate marked broken tape data.
[0116] The present invention ensures the comprehensiveness and accuracy of the data by collecting strip production data during the production process. It not only includes the basic information of strip production, but also covers various indicators that are crucial to quality control, and is the basis for subsequent analysis and optimization. Extracting broken strip data from production data can accurately identify the causes and patterns of broken strips, which is crucial for formulating corresponding preventive measures. Equipment status change parameter monitoring provides real-time information about the operating status of the equipment, which helps to promptly discover potential equipment failure problems, helps to optimize maintenance plans and arrange repair work in advance to avoid production interruptions. It is very effective. Combining equipment change parameters with broken strip data and performing identification processing on the change information further deepens the identification of the cause of broken strips, which not only helps to identify the association between broken strips and specific equipment status changes, but also promotes the formulation of targeted improvement measures.
[0117] In the embodiment of the present invention, since the state detection of the 18-roll rolling mill equipment itself is completed, the production data of 500 rolls of strip steel is continuously detected by the optimized 18-roll rolling mill equipment; the production information of the broken strip rolls in the 500 rolls of strip steel is extracted, including rolling force, tension, broken strip pass, outlet plate shape value of the broken strip pass, and forward slip value of the broken strip pass. According to statistics, there are 137 broken strips in the 500 rolls of strip steel, and its parameters are shown in the production information of the broken strip rolls of 137 rolls of strip steel in Table 1 below. The collected strip production data is analyzed, the key data points and abnormal indicators related to the broken strip event are extracted, and specific broken strip data are generated for further analysis and preventive measures. In the strip production process, the changes in the equipment state are monitored in real time, including parameters such as equipment wear, sensor deviation, and precision changes of the actuator, and all monitored equipment state change parameters are recorded, especially those key parameter changes that may affect production stability and product quality. Using the equipment status change parameters monitored in step S23, the broken belt data generated in step S22 is deeply analyzed to identify the correlation and influence relationship between these change parameters and the broken belt data, and the broken belt data with clear correlation is marked to indicate its relationship with the specific equipment status change parameters, so as to formulate targeted prevention and adjustment measures.
[0118] Table 1 Production information of broken coils of 137 coils
[0119]
[0120] Preferably, the analysis of critical parameters of belt break control based on the belt break identification data in step S3 includes the following steps:
[0121] Step S301: collecting the tape-breaking index data for the tape-breaking data according to the preset tape-breaking index type to generate the tape-breaking index data;
[0122] Step S302: performing data normalization processing on the belt-breaking index data to generate normalized belt-breaking index data;
[0123] Step S303: performing belt-break event correlation scoring processing on the normalized belt-break index data according to a preset isolation forest algorithm to generate belt-break event correlation scoring data;
[0124] Step S304: performing dimension reduction screening processing on the normalized belt-break index data based on the belt-break event correlation score data to generate screened belt-break index data;
[0125] Step S305: performing belt break control critical parameter analysis based on the screened belt break index data to generate belt break control critical parameters.
[0126] The present invention collects detailed belt-breaking index data for the belt-breaking data, and accurately identifies various potential factors that cause belt-breaking. Data normalization ensures that data from different sources and different magnitudes are compared and analyzed under the same standard, which improves the efficiency of data processing and the accuracy of analysis. The isolated forest algorithm is applied to score the correlation of belt-breaking events, and data nodes that are highly correlated with belt-breaking events are effectively identified. The application of this machine learning method makes the prediction of belt-breaking risks more accurate and efficient. Through dimensionality reduction and screening, only indicator data with higher influence are retained, which simplifies the monitoring and management of belt-breaking risks, not only improves the processing speed, but also improves the prediction accuracy of the monitoring and management of belt-breaking risks. According to the screened belt-breaking index data, in-depth analysis is carried out, and the generated critical parameters for belt-breaking control provide an accurate basis for real-time monitoring and early warning in the production process. These parameters help production managers to formulate preventive measures more effectively and reduce the occurrence of belt-breaking events.
[0127] As an example of the present invention, refer to Figure 2 As shown, Figure 1 The detailed implementation steps of step S3 of the interruption band control critical parameter analysis are shown in the flowchart. In this example, step S3 of the interruption band control critical parameter analysis includes:
[0128] Step S301: collecting the tape-breaking index data for the tape-breaking data according to the preset tape-breaking index type to generate the tape-breaking index data;
[0129] In an embodiment of the present invention, according to pre-designed belt-breaking index types, including rolling force, tension, forward slip value of belt-breaking pass and the like, belt-breaking index data are collected for the identified belt-breaking data, and corresponding index data are extracted from the identified belt-breaking data. These data will be used for subsequent analysis and modeling to generate belt-breaking index data.
[0130] Step S302: performing data normalization processing on the belt-breaking index data to generate normalized belt-breaking index data;
[0131] In the embodiment of the present invention, the collected belt-breaking index data are normalized to eliminate the influence of the dimension and unify the scale of the data, and the influence of each index on the belt-breaking is limited to the same range to generate normalized belt-breaking index data.
[0132] Step S303: performing belt-break event correlation scoring processing on the normalized belt-break index data according to a preset isolation forest algorithm to generate belt-break event correlation scoring data;
[0133] In an embodiment of the present invention, the key parameters of the isolation forest algorithm are determined, including the number of trees in the forest, the sample size of each tree, and any other parameters specific to the algorithm implementation. The selection of parameters can be based on the size and dimension of the data set, as well as previous experience. Using the pre-selected algorithm parameters, the isolation forest algorithm is applied to the normalized belt-breaking indicator data, and features and feature values are randomly selected to randomly construct a decision tree. In this process, each data point is isolated on the leaf node of the tree, and the path length to the leaf node can be used as an estimate of the anomaly score. The higher the degree of isolation of the data point (i.e., the shorter the path), the greater the possibility of being considered as an anomaly. This feature makes the isolation forest algorithm particularly suitable for identifying abnormal indicators that are highly correlated with belt-breaking events. The anomaly score calculated for each data point (i.e., the data instance of each belt-breaking indicator) is the belt-breaking event correlation score of the data point. A high score indicates that the data point has a strong correlation with the belt-breaking event, and vice versa, the correlation is weak. The scores of all data points are integrated into a belt-breaking event correlation score data set for subsequent analysis and decision-making.
[0134] Step S304: performing dimension reduction screening processing on the normalized belt-break index data based on the belt-break event correlation score data to generate screened belt-break index data;
[0135] In an embodiment of the present invention, based on the correlation score of the belt-breaking event, the normalized belt-breaking index data is reduced in dimension and screened, and only those indicators that have a high contribution to the prediction of the belt-breaking risk are retained. For example, the isolation forest will gradually isolate the indicators with a poor correlation with the belt-breaking time in the normalized belt-breaking index data, and then retain the normalized belt-breaking index data with a high correlation with the belt-breaking event to generate screened belt-breaking index data. For example, the screened belt-breaking index data obtained after the weft reduction screening include the rolling force of the belt-breaking pass, the tension of the belt-breaking pass, the plate shape value (extreme difference) of the belt-breaking outlet, and the forward slip value of the belt-breaking pass.
[0136] Step S305: performing belt break control critical parameter analysis based on the screened belt break index data to generate belt break control critical parameters.
[0137] In the embodiment of the present invention, the random forest algorithm is used to analyze the evaluation index of the screening belt-breaking index data, obtain the oversensitive index, and combine the normal distribution law of the oversensitive index value to obtain the belt-breaking frequency numerical parameter as the critical parameter of the belt-breaking control of the index, and then complete the construction of the belt-breaking prediction system and put it into production. Once the detection value of the oversensitive index monitored in production exceeds the critical parameter of control, the production line will be shut down. The critical parameters of belt-breaking control corresponding to the 1, 2, 3, 4, and 5 gears of the 18-roller rolling mill equipment after calculation and optimization include: the rolling force / Ton of the belt-breaking pass is: 872, 932, 955, 1107, 1300; the tension / Ton of the belt-breaking pass is: 21, 23, 25, 31, 33; the plate shape value (extreme difference) of the broken belt outlet is: 54I, 40I, 27I, 19I, 14I; the forward slip value of the broken belt pass is: 1.78, 1.65, 1.32, 1.29, 1.16.
[0138] Preferably, step S305 includes the following steps:
[0139] Perform indicator sensitivity assessment on the screened fault zone indicator data according to the preset random forest algorithm to generate indicator sensitivity assessment data;
[0140] According to the index sensitivity evaluation data, oversensitive index selection processing is performed on the screened broken belt index data to generate oversensitive broken belt index data;
[0141] Performing numerical distribution analysis on the oversensitive belt-breaking index data to generate oversensitive belt-breaking index numerical distribution data;
[0142] Based on the normal distribution algorithm, the critical parameters of the over-sensitive belt-breaking index are analyzed for the numerical distribution data of the over-sensitive belt-breaking index to generate the critical parameters of the belt-breaking control.
[0143] The present invention uses a random forest algorithm to perform a sensitivity assessment on the selected strip-breaking indicators, which can more accurately identify which indicators have a significant impact on strip-breaking events, and concentrate resources and attention on those strip-breaking indicators that are most likely to cause production problems. Select oversensitive indicators and perform numerical distribution analysis on them, and deeply understand how the data subset of oversensitive indicators affects the production process of strip steel, so as to formulate more effective preventive measures and improve the pertinence and efficiency of measures to prevent strip breaking in the production process of strip steel. Based on the normal distribution algorithm, the numerical distribution of oversensitive indicators is analyzed for critical parameters of strip-breaking control, which reflects how operations should be adjusted in the production process to avoid strip breaking, which helps to adjust production strategies in real time, reduce the risk of strip breaking in production, and better predict and manage potential risks in the production process, thereby avoiding unexpected downtime and production losses.
[0144] In the embodiment of the present invention, the random forest algorithm is applied to the screened belt-break index data. The random forest can process high-dimensional data and give sensitivity scores for each index and the belt-break event, which is suitable for evaluating the sensitivity of the index. The random forest model is initialized, and the performance of the model is optimized by pre-designing the number of trees and other model parameters, such as the maximum depth, the minimum number of sample splits, etc. The normalized belt-break index data is used to train the random forest model, and the importance score of each index is calculated as a quantitative index of its sensitivity. According to the importance score of the index, the data with higher scores in the index are selected as over-sensitive belt-break indicators. These indicators are considered to have high sensitivity and predictive value for predicting belt-break events, and the selection criteria of over-sensitive indicators are determined, for example, indicators with importance scores in the top percentage are selected. The selected over-sensitive belt-break index data are numerically analyzed for distribution to understand their distribution characteristics, and descriptive statistical analysis is used, including calculating the mean, standard deviation, skewness and kurtosis, etc., to analyze whether the distribution of the index data is close to the normal distribution, or whether there are other significant distribution characteristics, such as bimodal distribution, long-tail distribution, etc. The normal distribution algorithm is used to analyze the critical thresholds of the numerical distribution data of the oversensitive belt-breaking index, and these critical thresholds are used as the critical parameters for the control of the corresponding indexes at each gear. For example, the critical parameters for belt-breaking control corresponding to the 1, 2, 3, 4, and 5 gears of the 18-high rolling mill after calculation and optimization include: the rolling force / Ton of the belt-breaking pass is: 872, 932, 955, 1107, 1300; the tension / Ton of the belt-breaking pass is: 21, 23, 25, 31, 33; the strip-breaking exit shape value (extreme difference) is: 54I, 40I, 27I, 19I, 14I; the forward slip value of the belt-breaking pass is: 1.78, 1.65, 1.32, 1.29, 1.16.
[0145] Preferably, the abnormal fluctuation control processing of the real-time operation of strip production based on the critical parameters of strip break control in step S3 includes the following steps:
[0146] According to the critical parameters of strip break control, the real-time operation of strip production is subjected to abnormal fluctuation control, including: when the over-sensitive strip break index data of the real-time operation of strip production is not within the range of the critical parameters of strip break control, the real-time operation of strip production is stopped; when the over-sensitive strip break index data of the real-time operation of strip production is within the range of the critical parameters of strip break control, the real-time operation of strip production is continued to obtain the real-time data of strip production.
[0147] The present invention can timely identify conditions that may lead to production accidents by real-time monitoring of oversensitive belt-breaking index data and comparing it with the critical parameters of belt-breaking control. Once it is found that these indicators exceed the safe range, the production operation is stopped immediately, thereby preventing potential production accidents and reducing resource waste. It ensures that the oversensitive belt-breaking index data is always within the safe range of the critical parameters of belt-breaking control, which helps to maintain the stability and continuity of the production process. By preventing production interruptions caused by abnormal fluctuations, unnecessary losses caused by shutdown repairs and waste generation are reduced, material utilization and production efficiency are improved, and production costs are reduced. Continuously monitor and adjust the production process in a timely manner to ensure that production activities are carried out under optimal conditions, and continuously optimize the critical parameters of belt-breaking control based on iterative updates of subsequent steps to form a positive feedback loop, continuously improve the intelligence and automation level of the production process, and the real-time data of strip production obtained ensures the consistency and high standards of product quality.
[0148] In an embodiment of the present invention, the real-time operation of strip production is subjected to abnormal fluctuation control processing according to the critical parameters of strip break control. By real-time reading of key operating parameters in the strip production process, especially those that have been identified as oversensitive strip break index data, the critical parameters of strip break control of the real-time operation of strip production are compared with the oversensitive strip break index data, and it is determined whether the current value of the sensitive strip break index data of the real-time operation of strip production is within the corresponding critical parameter range. If any oversensitive strip break index data of the real-time operation of strip production is not within the range of its critical parameters of strip break control, a signal to stop the real-time operation of strip production is immediately sent to the production control system to prevent potential strip break risks. The signal should trigger the safety shutdown procedure of the production line, and notify the maintenance team to conduct inspections and necessary interventions. When all oversensitive strip break index data of the real-time operation of strip production are within the safe range of its critical parameters of strip break control, the monitoring system allows the real-time operation of strip production to continue, ensuring the continuity and efficiency of the production process, and finally collecting the data recorded by strip production in real time to obtain real-time data of strip production.
[0149] Preferably, the iterative updating operation of the critical parameters for belt break control in step S4 includes the following steps:
[0150] Real-time strip breakage data is monitored and processed for the real-time data of strip production, including: when the real-time data of strip production does not monitor the real-time strip breakage data, the critical parameters of strip breakage control are fed back to the terminal to execute the establishment operation of the strip breakage prediction mechanism; when the real-time data of strip production monitors the real-time strip breakage data, the real-time segment strip data is used to iteratively update the strip breakage data to generate updated strip breakage data, and the updated strip breakage data is transmitted to step S22 to execute the iterative update of the critical parameters of strip breakage control.
[0151] When the real-time data does not monitor the belt breakage, the present invention maintains the current critical parameters for belt breakage control to ensure the stability of the production process; when the real-time belt breakage data is monitored, the belt breakage data and the critical parameters for control are updated in time, and the adjusted critical parameters for control take into account the influence of historical data. This dynamic adjustment mechanism enables the belt breakage prediction mechanism to adapt to changes in the production process. For units that are put into production for a long period of time, factors such as spatial accuracy, equipment wear status, incoming material size and performance will change periodically with the rolling process, which will also affect the changes in the abnormal fluctuation rules of cold-rolled belt breakage parameters in different production cycles. By continuously collecting and analyzing real-time data in the production process and continuously performing iterative updates, deficiencies and potential improvement points in the production process are identified, providing data support for continuous improvement of the production process to adapt to changes in production conditions and ensure high quality and stability of the production process.
[0152] In the embodiment of the present invention, the real-time data of strip steel production is continuously collected and analyzed, and real-time monitoring is performed to detect whether there is real-time strip break data, that is, whether there is data indicating a possible strip break event during the production process. If there is no strip break event in the monitored real-time data (that is, all monitored oversensitive strip break index data are within the safe range of their strip break control critical parameters), the existing strip break control critical parameters are continued to be used for monitoring. If the real-time data shows a potential strip break event (and the monitored oversensitive strip break index data does not exceed the range of its strip break control critical parameters), the oversensitive strip break index data needs to be iteratively updated. The existing belt-breaking data set is updated using the monitored real-time belt-breaking data, including recording new belt-breaking events, actual values of relevant indicators, and specific production conditions when belt-breaking occurs. The updated belt-breaking data is fed back to step S22 to update the belt-breaking indicators of step S22, and the oversensitive belt-breaking indicator data in the sub-step of step S3 is re-analyzed based on the updated belt-breaking data to identify the corresponding belt-breaking control critical parameters to obtain updated oversensitive belt-breaking indicator data and corresponding updated belt-breaking control critical parameters. The updated oversensitive belt-breaking indicator data and corresponding updated belt-breaking control critical parameters are used to re-monitor and process the real-time belt-breaking data of strip production using the updated oversensitive belt-breaking indicator data and corresponding updated belt-breaking control critical parameters, and the real-time belt-breaking data is continuously updated until no belt-breaking data occurs in the latest strip production operation, then the iterative update is stopped, so as to establish a belt-breaking prediction mechanism. The belt-breaking prediction mechanism is used to set the equipment parameters of the 18-roller rolling mill equipment according to the latest belt-breaking control critical parameters to prevent belt-breaking.
[0153] The beneficial effect of the present application lies in that the present invention utilizes the actual production data of the cold rolling production line (specifically, the eighteen-roll mill cold rolling production line), and extracts the historical broken strip data under different steel grades, different sizes and specifications, and different rolling conditions from a large amount of data, and uses this as a broken strip sensitive data set; then, with the help of isolation forest algorithm and random forest algorithm, the sensitivity relationship between the cold rolling process data under different conditions and the broken strip phenomenon is analyzed; then, focusing on different steel grades, different sizes and specifications, and different rolling conditions, the changes in the broken strip sensitive data are observed in a targeted manner; once abnormal fluctuations in the sensitivity data in actual production are found, the main drive motor can be mobilized to implement a shutdown operation, thereby avoiding broken strips. Through multi-dimensional detection variables, it is possible to cover all products of this type of machine and accurately determine whether various products have the risk of strip breakage; and the detection variables of the main roll system and the side support roll system completely cover the control characteristics of the 18-high rolling mill equipment, making the detection accuracy of strip breakage detection higher; and the detection logic is not only for real-time monitoring data, but also for the influencing rules of historical data, including considering the spatial accuracy of the equipment, equipment wear status, incoming material size and performance and other factors will undergo periodic changes with the rolling process, so that the strip breakage detection results are excellent in different production cycles.
[0154] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.
[0155] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.
Claims
1. A method for predicting and controlling belt breakage suitable for eighteen-high rolling mill control. It is characterized in that The following steps are involved: Step S1: performing equipment status detection and repair optimization processing on the 18-high rolling mill equipment to obtain the optimized 18-high rolling mill equipment; Step S2: performing strip steel production operations based on the optimized 18-high rolling mill equipment; Analyze the strip breaking data according to the strip production operation and generate the strip breaking data; Step S3: analyzing the critical parameters of belt break control based on the marked belt break data to generate critical parameters of belt break control; Based on the optimized 18-high rolling mill equipment, the real-time operation of strip steel production is carried out; based on the critical parameters of strip break control, the real-time operation of strip steel production is controlled and processed to obtain the real-time data of strip steel production; Step S4: performing iterative update operation of critical parameters for strip break control according to real-time data of strip production.
2. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 1, It is characterized in that The equipment status detection and repair optimization processing includes spatial accuracy detection and repair processing, actuator detection and repair processing, and sensor detection and optimization processing at all levels. Step S1 includes the following steps: Step S11: Performing spatial accuracy detection and repair processing on the 18-high rolling mill equipment to obtain the 18-high rolling mill equipment with spatial accuracy repair; Step S12: performing an actuator detection and repair process on the 18-high rolling mill equipment with spatial accuracy repair to obtain an 18-high rolling mill equipment with an effective actuator; Step S13: Perform sensor detection optimization processing at various levels on the 18-high rolling mill equipment with effective actuators to obtain the optimized 18-high rolling mill equipment.
3. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 2, It is characterized in that Step S11 includes the following steps: Based on the preset wear state detection sequence data, the equipment wear state data of the 18-high rolling mill is collected to generate the equipment wear state data; Use gyroscope equipment to detect the rotation angle deviation state of the roll axis of the 18-high rolling mill equipment and generate roll axis state data; The laser tracker is used to detect the position deviation state of the transmission roller axis of the 18-high rolling mill equipment, and the transmission roller axis position deviation data is generated; Perform equipment spatial accuracy repair analysis and processing based on equipment wear status data, roller axis status data, and transmission roller axis position offset data to generate equipment spatial accuracy repair data; Based on the equipment spatial accuracy repair data, the eighteen-roller rolling mill equipment is spatially accurate repaired to obtain the eighteen-roller rolling mill equipment with spatial accuracy repair.
4. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 2, It is characterized in that Step S12 includes the following steps: The main roll system drive test processing is carried out according to the 18-high rolling mill equipment repaired with spatial accuracy to obtain the main roll system drive test data; The main roll system drive test data is analyzed for the main roll system drive test compliance data by using a preset main roll system drive threshold, including: when the main roll system drive test data is not less than the main roll system drive threshold, the 18-roller rolling mill equipment corresponding to the main roll system drive test data and whose spatial accuracy is repaired is marked as a main roll system effective drive, so as to obtain the 18-roller rolling mill equipment with effective main roll system drive; when the main roll system drive test data is less than the main roll system drive threshold, the 18-roller rolling mill equipment corresponding to the main roll system drive test data and whose spatial accuracy is repaired is marked as a main roll system invalid drive, so as to obtain the 18-roller rolling mill equipment with invalid main roll system drive; Feedback the 18-high rolling mill equipment with invalid main roll drive to the terminal to perform maintenance work on the press-down device model; Performing a support roller system drive test process on an 18-high rolling mill device that effectively drives the main roller system to obtain support roller system drive test data; The support roller system drive test data is analyzed for support roller system drive compliance data using a preset support roller system drive threshold, including: when the support roller system drive test data is not less than the support roller system drive threshold, the 18-roller rolling mill equipment whose main roller system corresponding to the support roller system drive test data is effectively driven is marked as a support roller system effective drive, so as to obtain the 18-roller rolling mill equipment with effective actuator; when the support roller system drive test data is less than the support roller system drive threshold, the 18-roller rolling mill equipment whose main roller system corresponding to the support roller system drive test data is effectively driven is marked as a support roller system invalid drive, so as to obtain the 18-roller rolling mill equipment with invalid support roller system drive; Feedback the invalid drive of the main roll system of the eighteen-high mill equipment to the terminal to perform maintenance work on the support roll system.
5. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 2, It is characterized in that Step S13 includes the following steps: Collect sensor accuracy data at all levels for the 18-high rolling mill equipment with effective actuators and generate sensor accuracy data at all levels; Perform sensor accuracy repair data analysis at all levels based on sensor accuracy data at all levels, and generate sensor accuracy repair data at all levels; According to the precision repair data of sensors at all levels, the precision repair and optimization processing of sensors at all levels of the 18-roller rolling mill equipment with effective actuators is carried out to obtain the optimized 18-roller rolling mill equipment.
6. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 1, It is characterized in that Step S2 includes the following steps: Step S21: executing a strip steel production operation based on the optimized 18-high rolling mill equipment, and collecting strip steel production data according to the strip steel production operation to generate strip steel production data; Step S22: extracting and processing the broken strip data according to the strip production data to generate the broken strip data; Step S23: monitoring equipment status change parameters according to the strip steel production operation and generating equipment change parameters; Step S24: using the equipment change parameters to perform change information identification processing on the broken tape data to generate marked broken tape data.
7. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 6, It is characterized in that The analysis of critical parameters for belt break control based on the belt break identification data in step S3 includes the following steps: Step S301: collecting the tape-breaking index data for the tape-breaking data according to the preset tape-breaking index type to generate the tape-breaking index data; Step S302: performing data normalization processing on the belt-breaking index data to generate normalized belt-breaking index data; Step S303: performing belt-break event correlation scoring processing on the normalized belt-break index data according to a preset isolation forest algorithm to generate belt-break event correlation scoring data; Step S304: performing dimension reduction screening processing on the normalized belt-break index data based on the belt-break event correlation score data to generate screened belt-break index data; Step S305: performing belt break control critical parameter analysis based on the screened belt break index data to generate belt break control critical parameters.
8. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 7, It is characterized in that Step S305 includes the following steps: Perform indicator sensitivity assessment on the screened fault zone indicator data according to the preset random forest algorithm to generate indicator sensitivity assessment data; According to the index sensitivity evaluation data, oversensitive index selection processing is performed on the screened broken belt index data to generate oversensitive broken belt index data; Performing numerical distribution analysis on the oversensitive belt-breaking index data to generate oversensitive belt-breaking index numerical distribution data; Based on the normal distribution algorithm, the critical parameters of belt break control of over-sensitive belt break indicators are analyzed on the numerical distribution data of over-sensitive belt break indicators to generate critical parameters of belt break control.
9. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 8, It is characterized in that The abnormal fluctuation control processing of the real-time operation of strip steel production based on the critical parameters of strip break control in step S3 includes the following steps: According to the critical parameters of strip break control, the real-time operation of strip production is subjected to abnormal fluctuation control, including: when the over-sensitive strip break index data of the real-time operation of strip production is not within the range of the critical parameters of strip break control, the real-time operation of strip production is stopped; when the over-sensitive strip break index data of the real-time operation of strip production is within the range of the critical parameters of strip break control, the real-time operation of strip production is continued to obtain the real-time data of strip production.
10. The method for predicting and controlling strip breakage adapted for eighteen-high rolling mill control according to claim 9, It is characterized in that The iterative update operation of the critical parameters of the belt break control in step S4 includes the following steps: Real-time strip breakage data is monitored and processed for the real-time data of strip production, including: when the real-time data of strip production does not monitor the real-time strip breakage data, the critical parameters of strip breakage control are fed back to the terminal to execute the establishment operation of the strip breakage prediction mechanism; when the real-time data of strip production monitors the real-time strip breakage data, the real-time segment strip data is used to iteratively update the strip breakage data to generate updated strip breakage data, and the updated strip breakage data is transmitted to step S22 to execute the iterative update of the critical parameters of strip breakage control.
Citation Information
Patent Citations
Belt breakage protection method of Sendzimir mill
CN112371729A
Comprehensive diagnosis method for cold rolling strip breakage of brittle material
CN114247760A
Strip breakage early warning method and device in ultra-thin strip rolling process based on strip shape evolution
CN115722536A
Broken belt detection method and device for belt conveyor
CN117142043A
Prediction and control method for corrugation and belt breakage of cold continuous rolling unit
CN118045871A
