Design method for medium-thickness plate components and hot rolling key process parameters

Through artificial neural network model and multi-objective optimization algorithm, the components of medium and thick plates and hot rolling process parameters are optimized, which solves the problems of cost and resource waste caused by production trial and error, and improves production efficiency and product stability.

CN119092020APending Publication Date: 2024-12-06ANSTEEL BEIJING RES INST CO LTD +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411222979.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-03
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The design of medium and thick plate composition and hot rolling process parameters now mainly relies on production experience, resulting in frequent trial and error in production, affecting the production rhythm and supply cycle, increasing costs and causing waste of resources.

Method used

The artificial neural network model is used to establish a mapping relationship between components and mechanical properties, and combined with multi-objective optimization algorithms, the components and process parameters are predicted and optimized to meet multiple mechanical performance indicators, shorten the production plan formulation cycle and improve production efficiency.

Benefits of technology

It achieves rapid completion of component and process design, improves the stability of product mechanical properties, reduces the cost caused by production trial and error, and reduces resource waste.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119092020A_ABST
    Figure CN119092020A_ABST
Patent Text Reader

Abstract

The invention relates to a design method for medium-thickness plate components and hot rolling key process parameters, which comprises the following steps: determining mechanical property indexes of a plate blank, designing value ranges of the plate blank components and the hot rolling key process parameters, establishing a mapping model of a relationship among the medium-thickness plate components, a production process and mechanical properties, and obtaining errors with expected mechanical property indexes. And determining constraint conditions and key parameters of a multi-objective optimization problem by taking an error between each mechanical property predicted value and an expected mechanical property index as an optimization objective, solving the constraint conditions and the key parameters to obtain optimal solutions of optimized components and key process parameters, calculating the predicted values of the mechanical properties corresponding to the optimized components and the key process parameters, and calculating the optimal solution of the optimized components and key process parameters. The error value with the mechanical property index and the evaluation index of the optimization problem are obtained, and the rationality of the optimization scheme is judged according to the evaluation index; component and process design can be rapidly completed, the production efficiency and the stability of the mechanical property of the product are improved, and resource waste is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of metal manufacturing, and in particular to a design method for the composition of a medium and thick plate and key hot rolling process parameters. Background Art

[0002] Before actual production, the composition and hot rolling process parameters of medium and thick plates are often formulated based on production experience, and then adjusted in combination with the mechanical properties requirements of the order. After trial production, the mechanical properties are tested to see if they meet the order requirements. If they meet the performance requirements, they can be shipped. If they do not meet the performance requirements, they will be downgraded and remedied by heat treatment processes. If the performance is seriously unsatisfactory, they will be scrapped. Due to the complex influencing factors at the production site, the production process is determined by trial and error, which affects the normal production rhythm and supply cycle, and at the same time increases production costs and causes waste of resources.

[0003] Based on the production data accumulated in the early stage, a mapping relationship model between the production process and the mechanical properties of the product is established. The mechanical properties of the product can be predicted according to the process design value, so as to obtain the gap between the mechanical properties required by the order, and the production composition and process parameters are optimized with multiple objectives to meet multiple mechanical performance indicators. The present invention has certain reference significance for shortening the cycle of medium and thick plate production planning, improving production efficiency, reducing the production cost of enterprises due to trial and error in production, and reducing resource waste. Summary of the invention

[0004] The present invention provides a method for designing the composition of medium and thick plates and key process parameters for hot rolling, which can quickly complete the design of the composition and process, improve production efficiency and the stability of product mechanical properties, and at the same time reduce the production cost of enterprises due to trial and error in production and reduce waste of resources.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for designing the composition of medium and thick plates and key process parameters for hot rolling comprises the following steps:

[0007] S1. Determine the specifications and expected mechanical properties of the slab according to the user's order requirements;

[0008] S2. Design the value range of slab composition and key hot rolling process parameters;

[0009] S3. Based on the previous production data, a mapping model of the relationship between the composition, production process and mechanical properties of medium and thick plates is established;

[0010] S4. Preset the composition of the batch of steel billets and the initial values ​​of key process parameters, calculate the predicted values ​​of various mechanical properties according to the mapping model, and obtain the error between the predicted mechanical properties and the expected mechanical properties;

[0011] S5. Taking the error between the predicted value of each mechanical property and the expected mechanical property index as the optimization target, determine the constraint conditions of the multi-objective optimization problem according to the actual production;

[0012] S6. Determine the key parameters of the multi-objective optimization algorithm, and solve the optimization problem to obtain the optimal solution of the optimized composition and key process parameters;

[0013] S7. Calculate the predicted values ​​of mechanical properties corresponding to the optimized components and key process parameters, obtain the error values ​​between the mechanical performance indicators and the evaluation indicators of the optimization problem, and judge the rationality of the optimization plan based on this.

[0014] Furthermore, the expected mechanical performance indicators in step S1 include tensile strength, yield strength, elongation, and impact strength.

[0015] Furthermore, the key process parameters of hot rolling in step S2 include the start rolling temperature, the number of rough rolling passes, the number of finishing rolling passes, the finishing rolling deformation rate, the finishing rolling starting core temperature and the finishing rolling core temperature drop value.

[0016] Furthermore, the mapping model in step S3 adopts an artificial neural network model, and through the production big data accumulated in the early stage and the results of mechanical properties inspection, an intelligent prediction model with input as billet composition and key process parameters and output as various mechanical properties is established.

[0017] Furthermore, the constraint conditions in step S5 include the difference between yield strength and tensile strength, the deformation rate of the roughing pass, the deformation rate of the finishing pass, the deformation rate of a single roughing pass, and the deformation rate of a single finishing pass.

[0018] Furthermore, the key parameters of the multi-objective optimization algorithm in step S6 include the number of iterations and the population size.

[0019] Furthermore, the evaluation indicators of the optimization problem in step S7 include calculation time, evaluation times, space measurement indicator spacing and hypervolume indicator HV.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1) Based on machine learning algorithms such as artificial neural networks, a model of the mapping relationship between composition, process parameters and mechanical properties such as yield strength, tensile strength, elongation, and impact strength is established; then a multi-objective optimization algorithm is used to optimize the error between the predicted value of the mechanical properties and the set target value of the mechanical properties, thereby obtaining the composition and process design that can meet the order requirements;

[0022] 2) Modeling based on actual production data can take into account the coupling effects between the complex environment of the production site, equipment status, and process parameters;

[0023] 3) Combined with multi-objective optimization algorithms, the production ingredients and process parameters are optimized to meet multiple mechanical performance indicators. While quickly completing the composition and process design, the R&D cycle is shortened, the production costs caused by trial and error in production are reduced, and the waste of resources is reduced;

[0024] 4) It can be extended to the production practice in the fields of hot-rolled plates and strips, cold-rolled steel plates, titanium alloy plates, aluminum alloy plates, etc., and has a broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic flow chart of the method of the present invention.

[0026] Figure 2 It is a schematic diagram of a curve showing a change of the spatial metric index along with the genetic generation number described in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The specific implementation of the present invention will be further described below in conjunction with the accompanying drawings:

[0028] See Figure 1 , is a schematic diagram of the method flow of the present invention. A method for designing the composition of a medium and thick plate and key hot rolling process parameters of the present invention comprises the following steps:

[0029] S1. Determine the specifications, tensile strength, yield strength, elongation, impact strength and other mechanical properties of the slab according to the user's order requirements.

[0030] S2. Design the upper and lower limits of slab composition and key hot rolling parameters based on production experience. Key hot rolling process parameters include start rolling temperature, number of rough rolling passes, finishing deformation rate, finishing starting core temperature, number of finishing passes and finishing core temperature drop value.

[0031] S3. Based on the previous production data, a mapping model of the relationship between the composition, production process and mechanical properties of medium and thick plates is established. The mapping model adopts an artificial neural network model. Through the production big data accumulated in the early stage and the results of mechanical properties inspection, an intelligent prediction model with input as billet composition and key process parameters and output as various mechanical properties is established.

[0032] S4. Preset the composition of the batch of steel billets and the initial values ​​of key process parameters, calculate the predicted values ​​of various mechanical properties according to the mapping model, and obtain the error between the predicted mechanical properties and the expected mechanical properties.

[0033] S5. Take the error between the predicted value and the target value of each mechanical property as the optimization target, and determine the constraints of the optimization problem in combination with actual production. The constraints include the difference between yield strength and tensile strength, deformation rate of roughing pass, deformation rate of finishing pass, deformation rate of single roughing pass, and deformation rate of single finishing pass.

[0034] S6. Determine the number of iterations and population size of key parameters of the multi-objective optimization algorithm, and solve the optimization problem to obtain the optimal solution of the optimized composition and key process parameters, which specifically includes the following processes:

[0035] S6.1. Determine the number of iterations and population size of the multi-objective optimization algorithm;

[0036] S6.2, according to the prediction model described in step S3, calculate the mechanical properties corresponding to the initial population, obtain the error between the predicted performance and the performance target value, and use this as the fitness judgment standard;

[0037] S6.3. According to the constraints set in step S5, the optimal solution set is continuously optimized until the upper limit of the number of iterations is reached.

[0038] S7. Calculate the predicted values ​​of mechanical properties corresponding to the optimized components and key process parameters, obtain the error value between the mechanical property index and the evaluation index of the optimization problem, and judge the rationality of the optimization scheme based on this, which specifically includes the following steps:

[0039] S7.1. Output the error between the predicted performance of the optimal solution set after optimization and the performance target value, as well as the corresponding material element composition and process parameter values;

[0040] S7.2. Output the evaluation indicators of the optimization algorithm, including time, number of evaluations, and drawing a tracking graph of the evolution process.

[0041] The following examples are implemented on the premise of the technical solution of the present invention, and provide detailed implementation methods and specific operation processes, but the protection scope of the present invention is not limited to the following examples. The methods used in the following examples are conventional methods unless otherwise specified.

[0042] [Example]

[0043] The present invention provides a method for designing the composition of a medium and thick plate and key process parameters for hot rolling, comprising the following steps:

[0044] S1. According to the user's order requirements, the specifications of the slab are determined as follows: finished product thickness 25mm, width ratio 1.5, yield strength ≥355MPa, tensile strength ≥480MPa, elongation ≥20%, and impact energy ≥63J.

[0045] S2. Design the upper and lower limits of slab composition and key hot rolling process parameters based on production experience, as shown in Table 1 and Table 2;

[0046] Table 1 Ingredient range

[0047] C / wt% Si / wt% Mn / wt% Cr / wt% Nb / wt% V / wt% Ti / wt% Ni / wt% Upper limit 0.18 0.8 1.5 0.04 0.03 0.03 0.03 0.03 Lower limit 0.03 0.2 0.7 0 0 0 0 0

[0048] Table 2 Key process parameter ranges for hot rolling

[0049]

[0050] S3. Based on the previous production data, a mapping model of the relationship between the composition, production process and mechanical properties of medium and thick plates is established. The mapping model adopts an artificial neural network model. Through the previous accumulated production big data and the results of mechanical properties inspection, an intelligent prediction model with input as billet composition and key process parameters and output as various mechanical properties is established;

[0051] S4. Preset the composition of the batch of steel billets and the initial values ​​of key process parameters, calculate the predicted values ​​of various mechanical properties according to the mapping model, and obtain the error between the predicted mechanical properties and the expected mechanical properties, as shown in Table 3-5;

[0052] Table 3 Composition before optimization

[0053]

[0054]

[0055] Table 4 Process parameters before optimization

[0056]

[0057] Table 5 Mechanical properties error

[0058] Absolute error of yield strength Absolute error of tensile strength Total elongation absolute error Impact energy 1 16.51 19.01 1.41 220.13 2 0.87 1.70 2.92 169.53 3 13.53 57.09 0.002 188.01

[0059] S5. Take the error between the predicted value of each mechanical property and the expected mechanical property index as the optimization target, and determine the constraint conditions of the multi-objective optimization problem according to the actual production. The constraint conditions include the difference between yield strength and tensile strength, the upper limit of single deformation rate of rough rolling pass, the lower limit of single deformation rate of rough rolling pass, the upper limit of single deformation rate of finishing pass, and the lower limit of single deformation rate of finishing pass.

[0060] S6. Set the number of iterations of the multi-objective optimization algorithm to 50 and the population size to 500, and solve the optimization problem.

[0061] S7, outputting the optimal solution set of optimized components and key process parameters, the error value between the predicted mechanical properties and the mechanical property index, and the evaluation index of the optimization problem. Here, three of the solutions are taken as examples, as shown in Tables 6-8;

[0062] Table 6 Optimal solution of ingredients

[0063] C / wt% Si / wt% Mn / wt% Cr / wt% Nb / wt% V / wt% Ti / wt% Ni / wt% 1 0.11 0.31 1.39 0.03 0.01 0 0 0.01 2 0.12 0.37 0.96 0.04 0.03 0.03 0 0.02 3 0.107 0.23 1.36 0.04 0.03 0.02 0 0.01

[0064] Table 7 Optimal process solution

[0065]

[0066] Table 8 Errors between predicted values ​​and target values

[0067]

[0068]

[0069] The calculation took 0.993 seconds, the number of evaluations was 22750, the hypervolume index HV value was 0.999, and the spatial metric spacing changed with the genetic generation. Figure 2 shown.

Claims

1. A method for designing the composition of medium and thick plates and key process parameters for hot rolling, characterized in that: The steps include: S1. Determine the specifications and expected mechanical properties of the slab according to the user's order requirements; S2. Design the value range of slab composition and key hot rolling process parameters; S3. Based on the previous production data, a mapping model of the relationship between the composition, production process and mechanical properties of medium and thick plates is established; S4. Preset the composition of the batch of steel billets and the initial values ​​of key process parameters, calculate the predicted values ​​of various mechanical properties according to the mapping model, and obtain the error between the predicted mechanical properties and the expected mechanical properties; S5. Taking the error between the predicted value of each mechanical property and the expected mechanical property index as the optimization target, determine the constraint conditions of the multi-objective optimization problem according to the actual production; S6. Determine the key parameters of the multi-objective optimization algorithm, and solve the optimization problem to obtain the optimal solution of the optimized composition and key process parameters; S7. Calculate the predicted values ​​of mechanical properties corresponding to the optimized components and key process parameters, obtain the error values ​​between the mechanical performance indicators and the evaluation indicators of the optimization problem, and judge the rationality of the optimization plan based on this.

2. A method for designing the composition of medium and thick plates and key process parameters for hot rolling according to claim 1, characterized in that: The expected mechanical performance indicators in step S1 include tensile strength, yield strength, elongation, and impact strength.

3. The method for designing the composition of medium and thick plates and key process parameters for hot rolling according to claim 1, characterized in that: The key process parameters of hot rolling in step S2 include the start rolling temperature, the number of rough rolling passes, the number of finishing rolling passes, the finishing rolling deformation rate, the finishing rolling starting core temperature and the finishing rolling core temperature drop value.

4. The method for designing the composition of medium and thick plates and key process parameters for hot rolling according to claim 1, characterized in that: The mapping model in step S3 adopts an artificial neural network model, and establishes an intelligent prediction model with inputs of billet composition and key process parameters and outputs various mechanical properties through the production big data accumulated in the early stage and the results of mechanical property inspection.

5. The method for designing the composition of medium and thick plates and key process parameters for hot rolling according to claim 1, characterized in that: The constraint conditions in step S5 include the difference between yield strength and tensile strength, the deformation rate of the roughing pass, the deformation rate of the finishing pass, the deformation rate of a single roughing pass, and the deformation rate of a single finishing pass.

6. The method for designing the composition of medium and thick plates and key process parameters for hot rolling according to claim 1, characterized in that: The key parameters of the multi-objective optimization algorithm in step S6 include the number of iterations and the population size.

7. The method for designing the composition of medium and thick plates and key process parameters for hot rolling according to claim 1, characterized in that: The evaluation indicators of the optimization problem in step S7 include calculation time, evaluation times, space measurement indicator spacing and hypervolume indicator HV.

Citation Information

Patent Citations

  • Hot-rolled H profile steel mechanical property forecasting method based on multivariate linear regression analysis

    CN104573278A

  • Control method for steel plate performance optimization

    CN106991242A

  • Hot-rolled strip steel production process dynamic optimization method based on MOEA / D intelligent optimization algorithm

    CN116779062A

  • Medium plate rolling process feedforward adjusting method, device and equipment and medium

    CN116934066A