Roller wear error checking method and system based on edge computing

Abstract

The application provides a kind of based on edge computing's roll wear error checking method and system, from how to realize the automation angle of roll wear error statistical checking process, comply with the time line and process line of data generation, redesign roll wear error statistical checking process, clear the role of secondary, roll grinding department in roll wear error checking process, introduce edge computing into roll wear error checking process, realize the full-automatic operation of roll wear error checking, solve the long-standing problem of roll wear data flow in hot rolling strip production. It is suitable for model parameter optimization when the hit rate of hot rolling strip shape crown control needs to be improved, especially when there is a large error between the calculation results of the hot rolling secondary wear model and the measured results. Reduce the error between the calculation results of the secondary wear model and the measured results, and minimize the impact of the calculation error of the wear model on the secondary model strip shape setting.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control, in particular to a roll wear error checking method and system based on edge computing. BACKGROUND

[0002] Plate strip production is mainly divided into hot rolling and cold rolling two stages, due to the difference of many factors such as incoming material, etc., causing the hot rolling stage secondary plate shape setting model to be mainly semi-open loop setting, and the cold rolling stage to be mainly closed loop setting. In the process of secondary model plate shape setting in the hot rolling stage, two key influencing factors of roll thermal expansion and roll wear are calculated by using semi-theoretical and semi-empirical model. In the actual use process, due to the misunderstanding of the design concept and model checking method of the model, many hot rolling plate strip production plants do not perform any model calibration work after the preliminary checking is completed by the secondary model delivery company when the model is delivered. Most hot rolling plate strip production plants do not perform model checking even in the entire life cycle, which causes the decrease of product crown hit rate and product competitiveness, and causes huge economic losses to the enterprise. Figure 1 The existing roll wear error analysis flowchart.

[0003] After in-depth research on the roll wear model of hot rolling plate strip production plant and detailed analysis on the roll wear model parameter checking process of hot rolling plate strip production plant, it can be determined that the reasons why the hot rolling plate strip production plant does not check the wear model mainly include the following points:

[0004] (1) There is serious hysteresis between the roll grinding process and the roll wear process, and the automation degree is not high. The wear data of many hot rolling plants still needs to be stored on paper, and data entry cannot be automated, and data automatic transmission is even more difficult to achieve.

[0005] (2) A large amount of manual operation is required in the hot rolling plate strip roll wear checking process, such as obtaining the grinding curve, searching for the rolling performance, obtaining the corresponding rolling plan single secondary calculation result, etc. It is complex and tedious, and involves multiple departments such as secondary and roll grinding department, and the multiple departments are in a flat state, which cannot be well communicated, thereby causing the data chain to be broken.

[0006] (3) The model parameter correction coefficient of roll wear in the hot rolling plate strip secondary model is generally responsible by the secondary model department, which is irrelevant to the hot rolling plate strip roll wear workshop, thereby causing unclear responsibility and splitting the parameter correction process of the roll wear model.

[0007] In the traditional roll wear error statistical checking process, manual participation in judgment or data tracking is required in each link, the overall efficiency is low, and automation and intelligentization cannot be realized. SUMMARY

[0008] In order to solve the technical problems put forward in the background art, the present application provides a kind of based on edge computing's roll wear error checking method and system, from how to realize the automation angle of roll wear error statistical checking process, conform to the time line and process line of data generation, redesign the roll wear error statistical checking process, clear the role of secondary, roll wear error checking process in mill, introduce edge computing into roll wear error checking process, realize the full-automatic operation of roll wear error checking, solve the long-standing problem of roll wear data flow in hot strip production.

[0009] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0010] A kind of based on edge computing's roll wear error checking method, comprising the following steps:

[0011] Step 1: root plate shape setting model, i.e. SSU model, transmits the wear calculation value data of each rolling production to the edge computing system of roll wear checking;

[0012] Step 2: the edge computing system transmits the wear calculation value of different rolling production to the corresponding roll data in roll database;

[0013] Step 3: when roll grinding, the data of this grinding is transmitted to the edge computing system by operator;

[0014] Step 4: the edge computing system or artificial judges the accuracy of the data recorded during roll grinding, determines whether the reorganized data is credible and usable and carries out data cleaning, and eliminates invalid and erroneous data;

[0015] Step 5: the edge computing system automatically calculates the actual roll wear value according to the internal algorithm and rules;

[0016] Step 6: the edge computing system automatically calculates the secondary model correction coefficient according to the actual roll wear value and the SSU grinding calculation value of rolling plan sheet;

[0017] Step 7: the edge computing system is used to correct the secondary model correction coefficient regularly, and the automatic checking work of roll wear error is completed.

[0018] Further, the method adjusts the original roll wear error analysis process of finding required data in reverse with result as index to the operation process of time line and process line of data generation.

[0019] Further, in step 4, the edge computing system judges whether the curve is available through image recognition according to the wear curve of artificial grinding, carries out data cleaning, and eliminates invalid and erroneous data.

[0020] The application also provides a system for the roll wear error checking method based on edge calculation, which comprises a rolling production secondary control system, a secondary model is arranged in the rolling production secondary control system, an edge server is further arranged, the edge server is in communication connection with the rolling production secondary control system, an edge calculation system is arranged in the edge server, and a software model of the roll wear error checking method is arranged in the edge calculation system.

[0021] Compared with the prior art, the application has the following beneficial effects:

[0022] 1) The application is used for checking and optimizing the parameters of the existing hot-rolled plate strip secondary model roll wear calculation model, and is particularly suitable for model parameter optimization when the hot-rolled plate shape crown control hit rate needs to be improved, and is particularly used when there is a large error between the calculation results of the hot-rolled secondary wear model and the measured results. The use of the technology can reduce the error between the calculation results of the secondary wear model and the measured results, and complete the periodic automatic calibration work, thereby minimizing the influence of the wear model calculation error on the secondary model plate shape setting;

[0023] 2) The application adjusts the original roll wear error analysis process of finding required data in reverse according to the results to an operation process of the time line and the process line of data generation, so that the automatic transmission of data is facilitated;

[0024] 3) The application redesigns the roll wear error statistical checking process from the perspective of how to realize the automation of the roll wear error statistical checking process, thereby minimizing the manual operation in the whole process and improving the closed loop of the roll wear;

[0025] 4) The application does not need to modify the existing SSU model, only needs to introduce the calculation value of the SSU into the edge calculation system of the roll wear, has high portability and wide adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 The existing roll wear error analysis flowchart mentioned in the background art;

[0027] Figure 2 The roll wear error analysis flowchart based on edge calculation of the application. DETAILED DESCRIPTION

[0028] The specific embodiment of the application provided in the following is described in detail in combination with the drawings.

[0029] As shown in the drawings, Figure 2 a roll wear error checking method based on edge calculation comprises the following steps:

[0030] Step 1: The SSU model transmits the wear calculation value data of each rolling production to the edge computing system for checking the roll wear;

[0031] Step 2: The edge computing system transmits the wear calculation value of different rolling productions to the corresponding roll data in the roll database;

[0032] Step 3: When the roll is ground, the data of this grinding is transmitted to the edge computing system by the operator;

[0033] Step 4: The edge computing system or the human judges the accuracy of the data recorded during the roll grinding, determines whether the reorganized data is reliable and usable, and performs data cleaning to eliminate invalid and erroneous data;

[0034] Step 5: The edge computing system automatically calculates the actual roll wear value according to the internal algorithm and rules;

[0035] Step 6: The edge computing system automatically calculates the secondary model correction coefficient according to the actual roll wear value and the SSU grinding calculation value of the rolling plan;

[0036] Step 7: The edge computing system regularly corrects the secondary model correction coefficient to complete the automatic checking of the roll wear error.

[0037] Further, the method adjusts the original roll wear error analysis process of finding the required data in reverse according to the result to an operation process of the timeline and process line of the data generation.

[0038] Further, in step 4, the edge computing system judges whether the wear curve of manual grinding is available through image recognition, performs data cleaning, and eliminates invalid and erroneous data.

[0039] The application also provides a system for the roll wear error checking method based on edge computing, which comprises a rolling production secondary control system, the rolling production secondary control system is provided with a secondary model, further comprises an edge server, the edge server is in communication connection with the rolling production secondary control system, and the edge server is internally provided with an edge computing system, and the edge computing system is provided with a software model of the roll wear error checking method. Specific embodiments:

[0041] Next, the roll wear error checking method and device based on edge computing are exemplified by optimizing parameters in a certain hot rolling production line.

[0042] I. Add a new edge computing server;

[0043] II. Add new software, develop a response to the roll wear error checking analysis model based on edge computing server;

[0044] III. After the completion of the single rolling plan of the secondary model, the roll wear data calculated by the SSU, the roll number, the roll diameter, the roll rolling kilometers, the roll rolling block number and other data are transmitted to the edge computing server;

[0045] IV. The edge computing system transmits the wear calculation values of different rolling productions to the corresponding roll data in the roll database;

[0046] V. After the roll cooling in the roll mill, the roll grinding is carried out on the grinding machine, and the measured value of the roll wear, the grinding curve and other data are transmitted to the edge computing server through the roll mill system;

[0047] VI. The edge computing server judges whether the curve is available through manual or image recognition according to the wear curve of manual grinding, carries out data cleaning, and eliminates useless data;

[0048] VII. The edge computing system automatically calculates the actual roll wear value according to the built-in algorithm and rules;

[0049] VIII. The edge computing system automatically calculates the correction coefficient of the secondary model according to the built-in algorithm, the error calculation value of the measured value and the calculated value;

[0050] IX. The secondary model correction coefficient is corrected regularly by the edge computing system, and the automatic checking work of the roll wear error is completed. Here, it needs to be judged whether the on-site development permission is complete.

[0051] The results show that the present application is used for checking and optimizing the existing hot rolling plate strip secondary model roll wear calculation model setting parameters, especially suitable for model parameter optimization work when the hot rolling plate shape crown control hit rate needs to be improved, especially when there is a large error between the hot rolling secondary wear model calculation result and the measured result. The use of this technology can reduce the error between the secondary wear model calculation result and the measured result, and complete the regular automatic calibration work, which can minimize the influence of the wear model calculation error on the secondary model plate shape setting.

[0052] Those skilled in the art will appreciate that embodiments of the application can be readily used as software, hardware, or a combination of software and hardware. In one embodiment, the application can be implemented in software and / or firmware. In addition, the software implementation can be implemented by one or more computer programs.

[0053] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing each of the flowchart blocks or the functions noted in the blocks

[0054] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing each of the flowchart blocks or the functions noted in the blocks

[0055] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing each of the flowchart blocks or the functions noted in the blocks

[0056] While the preferred embodiments of the application have been described, additional variations and modifications can be employed. Therefore, the terms and expressions

[0057] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for checking roll wear error based on edge computing, characterized in that, Includes the following steps: Step 1: Based on the plate shape setting model, i.e., the SSU model, transmit the wear calculation value data of each rolling production to the edge calculation system for roll wear verification; Step 2: The edge computing system transmits the wear calculation values ​​for different rolling processes to the corresponding roll data in the roll database; Step 3: During roll grinding, the operator transmits the grinding data to the edge computing system; Step 4: The edge computing system or manual judgment is used to determine the accuracy of the data entered during the roll grinding process, to determine whether the data is reliable and usable, and to clean the data, removing invalid and erroneous data. Step 5: The edge computing system automatically calculates the actual wear value of the rolls based on its built-in algorithms and rules; Step 6: The edge computing system automatically calculates the secondary model correction coefficients based on the actual wear values ​​of the rolls in the rolling plan and the SSU grinding calculation values ​​in the rolling plan; Step 7: Use the edge computing system to periodically correct the secondary model correction coefficients to complete the automatic verification of roll wear error.

2. The method for checking roll wear error based on edge calculation according to claim 1, characterized in that, The method described here adjusts the original process of reverse searching for the required data by using the results as an index in the analysis of roll wear error to an operation process based on the timeline and process line of data generation.

3. The method for checking roll wear error based on edge calculation according to claim 1, characterized in that, In step 4, the edge computing system uses image recognition to determine whether the wear curve of manual grinding is usable, performs data cleaning, and removes invalid and erroneous data.

4. A system for a roll wear error verification method based on edge computing as described in claim 1, comprising a two-level rolling production control system, wherein the two-level rolling production control system includes a two-level model, characterized in that, It also includes an edge server, which is communicatively connected to the secondary control system of rolling production. The edge server has an edge computing system inside, which contains a software model of the roll wear error verification method.

Images