Hot-rolled product steel grade design method based on performance forecasting model

Abstract

The invention discloses a hot-rolled product steel grade design method based on a performance forecasting model, which comprises the following steps of: screening similar steel grades as reference steel grades according to target steel grades required by new users and historical production data, and optimizing components and processes of the reference steel grades by using the performance forecasting model. And components and process design values meeting the requirements of the target steel grade are obtained and used for supporting a product engineer to carry out new steel grade design work. According to the method, a stable and reliable performance forecasting model is established, a component, process and specification design scheme meeting the requirements of a new user is provided, the research and development period is effectively shortened, the research and development cost is effectively reduced, and therefore the technical competitiveness of an enterprise is improved.

Description

Technical Field

[0001] The present invention relates to the rolling process control technology of hot-rolled products in iron and steel metallurgy, and more specifically, to a steel grade design method for hot-rolled products based on a performance prediction model. Background Art

[0002] Hot-rolled products are one of the key strategic products of major iron and steel enterprises. Under the severe market situation, quality and cost are inevitably the core of competition in the same industry. Moreover, there are numerous personalized requirements from users of hot-rolled products. How to quickly respond when users put forward new requirements can not only reduce the R & D efficiency of new products, but also reflect the core technical competitiveness of the enterprise.

[0003] In all aspects of improving quality and reducing costs, making good use of production data and quickly finding a suitable optimization direction through analysis and modeling. Therefore, the research on performance prediction models and their applications has always been the focus of research by iron and steel enterprises and research institutions at home and abroad.

[0004] In existing patent applications, for example, Chinese Invention Patent Application No. 201110036223.X discloses a dynamic control method for the mechanical properties of hot-rolled strip steel based on a performance prediction model. This method uses a prediction model for the mechanical properties of hot-rolled strip steel to dynamically adjust the rolling process parameters of the strip steel, realizing the dynamic control of the mechanical properties of the strip steel and improving the qualification rate of hot-rolled products and the control accuracy of mechanical properties. Another example is Chinese Invention Patent Application No. 201210046441.6, which discloses a heating furnace energy-saving control method based on a strip steel mechanical property prediction model. This method uses a prediction model to reduce the furnace outlet temperature and reduce the gas consumption of the heating furnace, optimally controlling the process temperature of the strip steel when it leaves the furnace, and achieving a reduction in the gas consumption of the heating furnace / energy consumption per unit product. Another example is Chinese Invention Patent Application No.

[0005] CN 201410415017.3 discloses a finishing mill energy-saving control method based on mechanical property prediction and rolling energy consumption model. This method uses a prediction model to optimize the finishing mill rolling temperature and reduce the finishing mill rolling energy consumption. For example, the above patent focuses more on using a performance prediction model to achieve product quality management, monitoring process parameters using the model, and giving timely product quality risk warnings, so as to prevent losses and improve product quality. Another example is that Chinese invention patent application No. 200510011865.9 discloses a quality design method with extremely few new steel grade data samples. This method extracts no less than 50 production data close to the production process of the new steel grade from historical samples, combines them with the new steel grade test data samples to form a final sample, then establishes a new steel grade model through a weighted optimization method, and finally conducts quality design after establishing the new steel grade model according to the modeling objective function, effectively solving the problem that there are extremely few new steel grade test data samples and it is impossible to establish an effective model, and improving the accuracy of the new steel grade model and the reliability of the new steel grade design. This patent assists in the design of new products by combining historical production data. Its main function is to verify the reliability of the new steel grade design process through historical data and improve the test efficiency of new product design, but it does not intervene in the original design of the composition and process of the new steel grade in the user demand stage, thus improving the overall efficiency of the new steel grade design of new products.

[0006] From the above patent technologies, it can be seen that through data analysis and modeling, the development efficiency of new products can be effectively improved. Summary of the Invention

[0007] Aiming at the defects existing in the prior art, the purpose of the present invention is to provide a hot-rolled product steel grade design method based on a performance prediction model. By establishing a stable and reliable performance prediction model, it provides composition, process, and specification design schemes that meet the needs of new users, effectively shortening the R & D cycle and reducing R & D costs, thereby improving the technical competitiveness of enterprises.

[0008] To achieve the above purpose, the present invention adopts the following technical solutions:

[0009] A hot-rolled product steel grade design method based on a performance prediction model:

[0010] According to the target steel grade required by the new user, similar steel grades are selected as reference steel grades by combining historical production data, and the composition and process of the reference steel grades are optimized using the performance prediction model to obtain the composition and process design values that meet the requirements of the target steel grade, so as to support product engineers to carry out the design work of new steel grades.

[0011] Preferably, the hot-rolled product steel grade design method specifically includes the following steps:

[0012] S1. Collect historical production data of hot-rolled products to form a historical production information database of hot-rolled products;

[0013] S2. Based on the target steel grade required by the new user, combined with the historical production information database of the hot-rolled products, screen similar steel grades that meet the target steel grade as the reference steel grade;

[0014] S3. Obtain the element composition design information and hot-rolling process design information of the reference steel grade;

[0015] S4. According to the target steel grade, select the optimization parameters of composition, process, and specification, and design the optimization calculation step lengths for different parameters;

[0016] S5. Divide the optimization combinations according to the minimum values, maximum values, and calculation step lengths of the optimization parameters of composition, process, and specification, and use the performance prediction model to predict the divided optimization combinations. Combine with the target steel grade to calculate the performance compliance rate, and screen the optimization combinations that meet the requirements;

[0017] S6. If there is no optimization combination result that meets the requirements, return to step S2, re-select the reference steel grade, and conduct the optimization calculation for the new reference steel grade again; if there is an optimization combination result that meets the requirements, provide the screened optimization combination result as the design scheme of the composition, process, and specification of the new target steel grade to the product design engineer.

[0018] Preferably, in step S1, the historical production data of the hot-rolled products includes the composition design, specification size requirements, performance requirements, and contract design data of the hot-rolling process of the hot-rolled products; and

[0019] the production actual performance data of the steelmaking composition, hot-rolling temperature, and outlet thickness; and

[0020] the performance actual performance data of the yield strength, tensile strength, and fracture elongation.

[0021] Preferably, step S2 specifically includes:

[0022] Based on the target steel grade required by the new user, according to the parameter range [X 1 [a 1 , b 1 , X 2 [a 2 , b 2 ,,, X n [a n , b n , combined with the composition design data in the historical production information database of the hot-rolled products, screen the design target values (X 1 _AIM, X 2 _AIM,,, X n _AIM) or the maximum values (X 1_MAX,X 2 _MAX,,,X n _MAX), and the steel grades that meet the parameter range of the target steel grade are similar steel grades:

[0023]

[0024] Among them, X 1 ,X 2 ,,,X n represents the component parameter of elements C, SI, Mn, P, S, AL, B, N, V, Nb, Ti, Cu, Ni, Mo, Cr, H, W, Zr, O, As, Sn, Ca;

[0025] a 1 ,a 2 ,,,a n represents the minimum value of the X 1 ,X 2 ,,,X n component parameter range;

[0026] b 1 ,b 2 ,,,b n represents the maximum value of the X 1 ,X 2 ,,,X n component parameter range.

[0027] Preferably, in the step S2, for the selected similar steel grades, the mean value and standard deviation of each performance index of each similar steel grade need to be statistically analyzed, and the compliance rate of the historical performance of each steel grade is calculated in combination with the performance requirement range of the target steel grade, and the similar steel grade with the highest compliance rate is preferentially selected as the reference steel grade.

[0028] Preferably, in the step S3, the element composition design information includes the maximum value, minimum value and target value of the element composition parameters of C, SI, Mn, P, S, AL, B, N, V, Nb, Ti, Cu, Ni, Mo, Cr, H, W, Zr, O, As, Sn, Ca;

[0029] The hot rolling process design information includes the maximum value, minimum value and target value of the process parameters of the tapping temperature, rough rolling temperature, finish rolling temperature, coiling temperature.

[0030] Preferably, the step S4 specifically includes:

[0031] The specified composition requirements of the target steel grade (X 1 ,X 2 ,,,X n)All of them are used as optimization parameters. Then, combined with the correlation laws between components and processes, the remaining components, hot rolling processes, and specifications are selected as optimization parameters, and different calculation steps are given for the component, process, and specification parameters.

[0032] Preferably, the step S5 specifically includes:

[0033] The maximum and minimum values of the specified components of the target steel grade are comprehensively given by taking the intersection of the specified component range of the target steel grade and the component range of the reference steel grade. The maximum and minimum values of the optimization parameters of the remaining components, processes, and specifications are given according to the design values of the reference steel grade;

[0034] According to the maximum and minimum values and calculation steps of the optimization parameters of the components, hot rolling processes, and thickness specifications, M groups of optimization combinations composed of different component, hot rolling process, and thickness specification data are divided, and the mechanical properties of the M groups of optimization combinations are predicted by using the performance prediction model.

[0035] Preferably, the mechanical properties include yield strength, tensile strength, elongation, and impact energy.

[0036] Preferably, for the index prediction results of each of the mechanical properties [(P_YS, P_YS_STD), (P_TS, P_TS_STD), (P_EL, P_EL_STD), (P_EA, P_EA_STD),...], combined with the performance requirements of the target steel grade [[YS_min, YS_max], [TS_min, TS_max], [EL_min, EL_max], [EA_min, EA_max],...], calculate the compliance rate Qr_P_Y of the index prediction results of each of the mechanical properties = min[(Y_max - P_Y) / (3 * P_Y_STD), (P_Y - Y_min) / (3 * P_Y_STD)];

[0037] Take the minimum value of the compliance rates in the index prediction results of each of the mechanical properties as the comprehensive prediction performance compliance rate Qr_Com_P of the optimization combination = min[Qr_P_YS, Qr_P_TS, Qr_P_EL, Qr_P_EA,...], and a compliance rate threshold Qr_lim is given to screen the optimization combinations with a comprehensive compliance rate greater than the threshold;

[0038] In the formula, YS represents yield strength, TS represents tensile strength, EL represents fracture elongation, and EA represents impact energy.

[0039] A method for designing steel grades of hot-rolled products based on a performance prediction model provided by the present invention combines historical production data, uses the performance prediction model to predict the mechanical properties under different compositions, processes, and specifications, provides a suitable new steel grade design plan that meets the user's requirements for product engineers, and can greatly shorten the R & D cycle and R & D costs of the new steel grade process design. Brief Description of the Drawings

[0040] Figure 1 is a schematic flow chart of the method for designing steel grades of hot-rolled products of the present invention;

[0041] Figure 2 is a schematic flow chart of an embodiment of the method for designing steel grades of hot-rolled products of the present invention. Detailed Embodiments

[0042] In order to better understand the above technical solutions of the present invention, the technical solutions of the present invention will be further described below in conjunction with the drawings and embodiments.

[0043] A method for designing steel grades of hot-rolled products provided by the present invention:

[0044] According to the target steel grade required by the new user, similar steel grades are selected as the reference steel grades by combining historical production data, and the performance prediction model is used to optimize the composition and process of the reference steel grades to obtain the composition and process design values that meet the requirements of the target steel grade, so as to support the complex work of product engineers in designing new steel grades and improve efficiency.

[0045] Combined with Figure 1 As shown, the method for designing steel grades of hot-rolled products of the present invention specifically includes the following steps:

[0046] S1. Collect historical production data of hot-rolled products to form a historical production information database of hot-rolled products;

[0047] The historical production data of hot-rolled products includes contract design data such as the composition design, specification size requirements, performance requirements, and hot-rolling process of hot-rolled products; and

[0048] production performance data such as steelmaking composition, hot-rolling temperature, and outlet thickness; and

[0049] performance performance data such as yield strength, tensile strength, and fracture elongation.

[0050] S2. According to the target steel grade required by the new user, combine the historical production information database of hot-rolled products to select similar steel grades that meet the target steel grade as the reference steel grades;

[0051] According to the target steel grade required by the new user, according to the parameter range of the specified composition [X 1 [a 1 , b 1,X 2 [a 2 ,b 2 ,,,X n [a n ,b n , combined with the composition design data in the historical production information database of hot-rolled products, screen the design target values (X 1 _AIM,X 2 _AIM,,,X n _AIM) or the maximum values (X 1 _MAX,X 2 _MAX,,,X n _MAX) of the corresponding components in the historical steel grades. The steel grades that meet the parameter range of the target steel grade are similar steel grades:

[0052]

[0053] Among them, X 1 ,X 2 ,,,X n represents the composition parameter values of elements such as C, SI, Mn, P, S, AL, B, N, V, Nb, Ti, Cu, Ni, Mo, Cr, H, W, Zr, O, As, Sn, and Ca;

[0054] a 1 ,a 2 ,,,a n represents the minimum value of the X 1 ,X 2 ,,,X n composition parameter range;

[0055] b 1 ,b 2 ,,,b n represents the maximum value of the X 1 ,X 2 ,,,X n composition parameter range.

[0056] For the selected similar steel grades, based on the historical production data of each similar steel grade, calculate the mean values [YS_mean, TS_mean, EL_mean, EA_mean,...] and standard deviations [YS_std, TS_std, EL_std, EA_std,...] of mechanical properties such as yield strength YS, tensile strength TS, elongation at fracture EL, and impact energy EA. Combining with the performance requirement ranges [[YS_min, YS_max], [TS_min, TS_max], [EL_min, EL_max], [EA_min, EA_max],...] of the new target steel grade, calculate the performance compliance rate Qr_Y of each mechanical property index Y = Min[(Y_max - Y_mean) / (3*Y_std), (Y_mean - Y_min) / (3*Y_std)], and take the lowest performance compliance rate of each mechanical property index as the comprehensive performance compliance rate Qr_Com = min[Qr_YS, Qr_TS, Qr_EL, Qr_EA,...] of this similar steel grade.

[0057] Sort according to the comprehensive performance compliance rate Qr_Com of similar steel grades, and preferentially select the similar steel grade with the highest compliance rate as the reference steel grade.

[0058] S3. Obtain the element composition design information and hot rolling process design information of the reference steel grade;

[0059] The element composition design information includes the maximum values, minimum values, and target values of the element composition parameters of C, SI, Mn, P, S, AL, B, N, V, Nb, Ti, Cu, Ni, Mo, Cr, H, W, Zr, O, As, Sn, and Ca;

[0060] The hot rolling process design information includes the maximum values, minimum values, and target values of the process parameters of tapping temperature, rough rolling temperature, finish rolling temperature, and coiling temperature.

[0061] S4. According to the target steel grade, select the optimization parameters of composition, process, and specification, and design the optimization calculation step sizes of different parameters;

[0062] Take the specified composition requirements (X 1 , X 2 ,,, X n ) of the target steel grade as all optimization parameters, and then, combined with the correlation laws between composition and process, select the remaining composition, hot rolling process, and specification as optimization parameters, and give different calculation step sizes for composition, process, and specification parameters.

[0063] S5. Divide the optimization combinations according to the minimum values, maximum values and calculation steps of the optimization parameters of composition, process and specifications, use the performance prediction model to predict the divided optimization combinations, calculate the performance compliance rate in combination with the target steel grade, and screen out the optimization combinations that meet the requirements;

[0064] The maximum and minimum values of the specified components of the target steel grade are comprehensively given by taking the intersection of the component ranges specified by the target steel grade and the component ranges of the reference steel grade. The maximum and minimum values of the remaining composition, process and specification optimization parameters are given according to the design values of the reference steel grade;

[0065] According to the maximum values, minimum values and calculation steps of the composition, hot rolling process and thickness specification optimization parameters, divide M groups of optimization combinations composed of different composition, hot rolling process and thickness specification data, and use the performance prediction model to predict the mechanical properties of the M groups of optimization combinations.

[0066] The mechanical properties include yield strength, tensile strength, elongation and impact energy.

[0067] For the predicted results of the indexes of each mechanical property [(P_YS, P_YS_STD), (P_TS, P_TS_STD), (P_EL, P_EL_STD), (P_EA, P_EA_STD),...], combined with the performance requirements of the target steel grade [[YS_min, YS_max], [TS_min, TS_max], [EL_min, EL_max], [EA_min, EA_max],...], calculate the compliance rate Qr_P_Y of the predicted results of the indexes of each mechanical property = min[(Y_max - P_Y) / (3 * P_Y_STD), (P_Y - Y_min) / (3 * P_Y_STD)];

[0068] Take the minimum value of the compliance rate in the predicted results of the indexes of each mechanical property as the comprehensive prediction performance compliance rate Qr_Com_P of the optimization combination = min[Qr_P_YS, Qr_P_TS, Qr_P_EL, Qr_P_EA,...], and give the compliance rate threshold Qr_lim, and screen out the optimization combinations with a comprehensive compliance rate greater than the threshold.

[0069] S6. If there is no optimization combination result that meets the requirements, return to step S2, reselect the reference steel grade, and carry out the optimization calculation of the new reference steel grade again; if there is an optimization combination result that meets the requirements, provide the screened optimization combination result as the composition, process and specification design scheme of the new target steel grade to the product design engineer.

[0070] Example

[0071] Combined with Figure 2 As shown, this example involves a certain hot rolling production line of a certain steel plant.

[0072] The performance prediction models for selected hot-rolled products are as follows:

[0073] YS = f1(C, SI, MN, P, S, ALT, B, N, V, NB, TI, CU, NI, MO, CR, H, W, ZR, O, AS, SN, CA,,, DT, RT, FT, CT,,, THICK,,,);

[0074] TS = f2(C, SI, MN, P, S, ALT, B, N, V, NB, TI, CU, NI, MO, CR, H, W, ZR, O, AS, SN, CA,,, DT, RT, FT, CT,,, THICK,,,);

[0075] EL = f3(C, SI, MN, P, S, ALT, B, N, V, NB, TI, CU, NI, MO, CR, H, W, ZR, O, AS, SN, CA,,, DT, RT, FT, CT,,, THICK,,,);

[0076] EA = f4(C, SI, MN, P, S, ALT, B, N, V, NB, TI, CU, NI, MO, CR, H, W, ZR, O, AS, SN, CA,,, DT, RT, FT, CT,,, THICK,,,);

[0077] Among them, RE, RM, EL, and EA represent yield strength, tensile strength, elongation at break, and impact energy respectively;

[0078] Among them, C, SI, MN, P, S, ALT, ALS, B, N, V, NB, TI, CU, NI, MO, CR, H, W, ZR, O, AS, SN, CA, etc. represent carbon element, silicon element, manganese element, phosphorus element, sulfur element, aluminum element, boron element, nitrogen element, vanadium element, niobium element, titanium element, copper element, nickel element, molybdenum element, chromium element, hydrogen element, tungsten element, zirconium element, oxygen element, arsenic element, tin element, calcium element respectively;

[0079] Among them, DT, RT, FT, CT, and THICK represent tapping temperature, rough rolling temperature, finish rolling temperature, coiling temperature, and hot-rolled exit thickness respectively;

[0080] Collect data such as hot-rolled product design information, production performance, and performance performance to form a historical production information database for hot-rolled products. The details of the data items in the database are shown in Table 1.

[0081] Table 1 Details of data items in the historical production information database for hot-rolled products

[0082] Serial Number Parameter English Name Parameter Chinese Name … 1 ST_NO Steel Grade … 2 H_LINE Hot Rolling Production Line … … … … … 13 C_AIM Designed Target Value of Carbon Element … 14 C_MIN Designed Minimum Value of Carbon Element … 15 C_MAX Designed Maximum Value of Carbon Element … … … … … 75 DT Actual Tap Temperature … 76 RT Actual Rough Rolling Temperature … … … … … 103 YS Measured Value of Yield Strength … 104 TS Measured Value of Tensile Strength … … … … …

[0083] The optimized calculation step lengths of each component, process, and specification set according to the experience of technical personnel are shown in Tables 2 and 3.

[0084] Table 2 Main Component and Process Step Length Setting Table

[0085] Parameter Accuracy Step Size Number of Calculations C 4 0.005 5 Si 3 0.05 5 Mn 3 0.1 5 P 4 0.05 5 S 4 0.001 5 N 4 0.001 5 H 6 0.0001 5 A1 4 0.01 5 O 4 0.001 5 V 4 0.01 5 Nb 4 0.01 5 Ti 4 0.01 5 Cu 3 0.1 5 Ca 4 0.001 5 Mo 3 0.005 5 Cr 3 0.05 5 Ni 3 0.05 5 B 4 0.0005 5 W 4 0.001 5 Zr 4 0.001 5 As 4 0.001 5 Sn 4 0.001 5 DT 1 10 5 RT 1 10 5 FT 1 10 5 CT 1 10 5 …

[0086] Table 3 Main Specification Dimension Step Length Setting Table

[0087]

[0088] The component requirements and performance requirements of a new user requirement XQ1 of a certain user are shown in Tables 4 and 5.

[0089] Table 4 Component Requirements of User Requirement XQ1

[0090]

[0091]

[0092] Table 5 Performance Requirements of User Requirement XQ1

[0093]

[0094] Combined with the component design data in the historical production design information database of hot-rolled products, steel grades whose design target values or maximum values of corresponding components in the historical steel grades meet the component range of the target steel grade are screened. The screened similar steel grades are shown in Table 6.

[0095] Table 6 Screening Results of Similar Steel Grades for XQ1

[0096] Steel Grade C_Aim C_Min C_Max Si_Aim Si_Min Si_Max Mn_Aim Mn_Min Mn_Max ST1 0.08000 0.06500 0.09000 0.25000 0.19500 0.30400 1.48000 1.40000 1.55000 ST2 0.08300 0.07000 0.09500 0.20000 0.15000 0.25000 1.45000 1.40000 1.50000 ST3 0.07500 0.06000 0.09000 0.22000 0.17500 0.27500 1.38000 1.30000 1.45000 … … … … … … … … … … Steel Grade S_Aim S_Min S_Max P_Aim P_Min P_Max N_Aim N_Min N_Max ST1 0.00000 0.00000 0.00400 0.00000 0.00000 0.01500 0.00000 0.00000 0.00600 ST2 0.00000 0.00000 0.00600 0.00000 0.00000 0.01500 0.00000 0.00000 0.00500 ST3 0.00000 0.00000 0.00500 0.00000 0.00000 0.01500 0.00000 0.00000 0.00500 … … … … … … … … … … Steel Grade V_Aim V_Min V_Max Nb_Aim Nb_Min Nb_Max Ti_Aim Ti_Min Ti_Max ST1 0.05000 0.04500 0.05500 0.04500 0.04000 0.050010 0.01400 0.00800 0.02000 ST2 0.04000 0.03500 0.04500 0.02500 0.02000 0.03000 0.01500 0.01000 0.02000 ST3 0.04000 0.03500 0.04500 0.04000 0.03500 0.04500 0.02000 0.01500 0.02500 … … … … … … … … … … Steel Grade Tal_Aim Tal_Min Tal_Max … … … … … … ST1 0.02500 0.01500 0.04000 … … … … … … ST2 0.02800 0.01500 0.04000 … … … … … … ST3 0.03700 0.02500 0.05000 … … … … … … … … … … … … … … … …

[0097] For the screened similar steel grades, according to the historical production data of each similar steel grade, calculate the mean and standard deviation of each mechanical property such as yield strength YS, tensile strength TS, elongation at fracture EL, and impact energy FA. The performance statistical results of each similar steel grade are shown in Table 7.

[0098] Table 7 Performance Statistical Results of Similar Steel Grades

[0099] Steel Grade FREQ Ts_Mean Ys_Mean el_1_Mean ea_Mean TS_StD ys_Std el_Std ea_Std ST1 310 601.31 522.45 42.62 159.91 15.64 25.20 4.49 88.12 ST2 196 571.80 482.57 27.73 0.00 17.16 21.03 1.91 0.00 ST3 1393 594.29 517.25 41.24 140.34 15.39 18.95 4.83 67.40 …

[0100] Combined with the performance requirement range of the new target steel grade, calculate the performance compliance rate Qr_Y of each mechanical property index Y, and take the lowest performance compliance rate of each mechanical property index as the comprehensive performance compliance rate Qr_Com of this similar steel grade. The performance compliance rate calculation results of similar steel grades are sorted according to the comprehensive performance compliance rate Qr_Com of similar steel grades and shown in Table 8.

[0101] Calculation Results of Performance Compliance Rates of Similar Steel Grades in Table 8

[0102] Steel Grade Frequency Qr_TS Qr_Ys Qr_EL Qr_EA … QR_Com ST1 310 0.45 0.50 1.83 - 0.45 … ST3 1393 0.31 0.57 1.60 - 0.31 … ST2 196 -0.16 -0.04 1.70 - -0.16 …

[0103] Preferentially select the similar steel grade ST1 with the highest compliance rate as the reference steel grade.

[0104] For the selected reference steel grade, the design information such as the elemental composition design information and hot rolling process design information of this steel grade are shown in Table 9 and Table 10.

[0105] Table 9 Main Composition Design of Steel Grade ST1

[0106] Steel Grade C_Aim C_Min C_Max Si_Aim Si_Min Si_Max Mn_Aim Mn_Min Mn_Max ST1 0.08000 0.06500 0.09000 0.25000 0.19500 0.30400 1.48000 1.40000 1.55000 Steel Grade S_Aim S_Min S_Max P_Aim P_Min P_Max N_Aim N_Min N_Max ST1 0.00000 0.00000 0.00400 0.00000 0.00000 0.01500 0.00000 0.00000 0.00600 Steel Grade V_Aim V_Mim V_Max Nb_Aim Nb_Min Nb_Max Ti_Aim Ti_Min Ti_Max ST1 0.05000 0.04500 0.05500 0.04500 0.04000 0.05000 0.01400 0.00800 0.02000 Steel Grade Tal_Aim Tal_Min Tal_Max … … … … … … ST1 0.02500 0.01500 0.04000 … … … … … …

[0107] Table 10 Main Hot Rolling Process Parameter Design of Steel Grade ST1

[0108] Process Minimum Value Maximum Value Target Value DT 1170 1210 1190 RT 970 990 980 FT 830 850 840 cT 550 570 560

[0109] According to the user requirements of the new steel grade, all the specified components (C, Si, Mn, P, S, Nb, Ti, V, Tal) of the new target steel grade are taken as optimization parameters, and the remaining components are not taken as optimization parameters; when the component parameters change, the corresponding rolling process also needs to be adjusted appropriately. Therefore, hot rolling processes such as the tapping temperature DT, rough rolling temperature RT, finish rolling temperature FT, and coiling temperature CT all need to be taken as optimization parameters; at the same time, different thicknesses will also affect the process design. Therefore, the exit thickness THICK also needs to be taken as an optimization parameter.

[0110] Divide the optimization combinations according to the minimum values, maximum values, and calculation steps of parameters such as composition, process, and specification. Among them, the maximum and minimum values of the components specified for the new steel grade are comprehensively given by taking the intersection of the composition ranges specified for the new steel grade and the composition range of the reference steel grade. The maximum and minimum values of the remaining component, process, specification, etc. optimization parameters are given according to the design values of the reference steel grade. According to the maximum values, minimum values, and calculation steps of the optimization parameters such as composition, hot rolling process, and thickness specification, divide into M groups of optimization combinations composed of different composition, hot rolling process, and thickness specification data. The maximum values, minimum values, steps, and optimization combinations of each optimization parameter are shown in Table 11 below.

[0111] Table 11 Division Results of Optimization Combinations of Composition, Process, Specification, etc. of Steel Grade ST1

[0112] Optimization Parameter Minimum Value Maximum Value Step Size Optimization Array C 0.06500 0.09000 0.005 [0.065，0.070，，，0.09] Si 0.19500 0.30400 0.05 [0.195，0.245，，，，0.304] Mn 1.40000 1.55000 0.1 [1，4，1.5，1.55] P 0.00000 0.01500 0.05 [0，0.05，0.1，0.15] S 0.00000 0.00400 0.001 [0.000，0.001，，，0.004] Nb 0.04000 0.05000 0.01 [0.04.0.05] Ti 0.01 0.02000 0.01 [0.01，0.02] V 0.04500 0.05500 0.01 [0.045，0.055] Tal 0.01500 0.04000 0.01 [0.015，0.025，，，0.04] DT 1170 1210 10 [1170，1180，，，1210] RT 970 990 10 [970，980，990] FT 830 850 10 [830，840，850] CT 550 570 10 [550，560，570] Thick 4 21 (1，2，5) [4，5，，，10，12，，，20，21] … … …

[0113] Use the performance prediction model to predict the mechanical properties of the optimized combinations divided in Table 11. The prediction results include the predicted values and standard deviations of mechanical properties such as yield strength, tensile strength, elongation, impact energy, etc. Combine the performance requirements of the new target steel grade, calculate the compliance rate Qr_P_Y of the prediction results of each mechanical property index, and take the minimum value of the compliance rates among the indexes as the comprehensive prediction performance compliance rate Qr_Com_P of this optimized combination.

[0114] The optimization calculation results of steel grade ST1 are shown in Table 12.

[0115] Table 12 Optimization calculation results of steel grade ST1

[0116] Serial Number Thick C Si Mn P S Nb Ti V Tal 1 4 0.065 0.195 1.4 0 0 0.04 0.01 0.045 0.015 … … … … … … … … … … 72 12 0.075 0.245 1.5 0.1 0.002 0.04 0.01 0.055 0.035 … … … … … … … … … … … 131 20 0.09 0.304 1.55 0.1 0.001 0.05 0.01 0.045 0.025 … … … … … … … … … … … Serial Number DT RT FT CT … Qr_TS Qr_YS Qr_El Qr_EA Qr_Com 1 1170 970 830 550 0.965 0.840 1.225 - 0.840 … … … … … … … … … … … 72 1180 980 850 550 0.841 0.491 1.011 - 0.491 … … … … … … … … … … … 131 1210 990 830 560 0.223 0.377 0.574 - 0.223 … … … … … … … … … … …

[0117] Given a suitable compliance rate threshold Qr_lim = 0.8, screen the optimized combinations with a comprehensive compliance rate greater than the threshold, and sort the results of the screened optimized combinations according to the comprehensive performance compliance rate, and provide them to the product design engineer as the composition, process, and specification design scheme of the new target steel grade for steel grade design.

[0118] Table 13 Composition, process, and specification design scheme of the new target steel grade XQ1

[0119] Serial Number Thick C Si Mn P S Nb Ti V Tal 1 4 0.08 0.295 1.55 0.05 0.001 0.05 0.02 0.045 0.025 … … … … … … … … … … … 13 4 0.065 0.195 1.4 0 0 0.04 0.01 0.045 0.015 … … … … … … … … … … … Serial Number DT RT FT CT … Qr_TS Qr_YS Qr_E1 Qr_EA Qr_Com 1 1180 970 830 850 1.411 1.015 1.633 1.015 … … … … … … … … … … … 13 1170 970 830 550 0.965 0.740 1.225 - 0.840 … … … … … … … … … … …

[0120] Those of ordinary skill in the art in this technical field should recognize that the above embodiments are only used to illustrate the present invention, rather than to limit the present invention. As long as it is within the scope of the spirit of the present invention, changes and modifications to the above embodiments will fall within the scope of the claims of the present invention.

Claims

1. A method for designing the steel grade of hot-rolled products based on a performance prediction model, characterized in that: According to the target steel grade required by the new user, similar steel grades are selected as the reference steel grades by combining historical production data, and the performance prediction model is used to optimize the composition and process of the reference steel grades, so as to obtain the composition and process design values that meet the requirements of the target steel grade, which are used to support product engineers in carrying out the design work of new steel grades.

2. The method for designing the steel grade of hot-rolled products based on a performance prediction model according to claim 1, characterized in that, The method for designing the steel grade of hot-rolled products specifically includes the following steps: S1. Collect the historical production data of hot-rolled products to form a historical production information database of hot-rolled products; S2. According to the target steel grade required by the new user, combine the historical production information database of hot-rolled products, and select similar steel grades that meet the target steel grade as the reference steel grades; S3. Obtain the element composition design information and hot-rolling process design information of the reference steel grades; S4. According to the target steel grade, select the optimization parameters of composition, process and specification, and design the optimization calculation step length of different parameters; S5. Divide the optimization combinations according to the minimum value, maximum value and calculation step length of the optimization parameters of composition, process and specification, and use the performance prediction model to predict the divided optimization combinations, calculate the performance compliance rate in combination with the target steel grade, and screen out the optimization combinations that meet the requirements; S6. If there is no optimization combination result that meets the requirements, return to step S2, re-select the reference steel grade, and carry out the optimization calculation of the new reference steel grade again; if there is an optimization combination result that meets the requirements, then provide the screened optimization combination result as the composition, process and specification design scheme of the new target steel grade to the product design engineer.

3. The method for designing the steel grade of hot-rolled products based on a performance prediction model according to claim 2, characterized in that, In the step S1, the historical production data of the hot-rolled products include the contract design data of the composition design, specification size requirements, performance requirements and hot-rolling process of the hot-rolled products; and The production actual performance data of the steelmaking composition, hot-rolling temperature and outlet thickness; and The performance actual performance data of the yield strength, tensile strength and fracture elongation.

4. The method for designing the steel grade of hot-rolled products based on a performance prediction model according to claim 2, characterized in that, The step S2 specifically includes: According to the target steel grade required by the new user, based on the parameter range of the specified composition [X 1 [a 1 ,b 1 ,X 2 [a 2 ,b 2 ,,,X n [a n ,b n , combined with the composition design data in the historical production information database of the hot-rolled products, screen the design target values (X 1 _AIM,X 2 _AIM,,,X n _AIM) or the maximum values (X 1 _MAX,X 2 _MAX,,,X n _MAX) of the corresponding compositions in the historical steel grades. The steel grades that meet the parameter range of the target steel grade are the similar steel grades: Among them, X 1 , X 2 ,,, X n represent the elemental composition parameters of C, Si, Mn, P, S, Al, B, N, V, Nb, Ti, Cu, Ni, Mo, Cr, H, W, Zr, O, As, Sn, Ca; a 1 ,a 2 ,,,a n represent X 1 ,X 2 ,,,X n the minimum value of the component parameter range; b 1 ,b 2 ,,,b n represents X 1 ,X 2 ,,,X n The maximum value of the component parameter range.

5. The method for designing the steel grade of hot-rolled products based on a performance prediction model according to claim 4, characterized in that, In the step S2, for the selected similar steel grades, the mean value and standard deviation of each performance index of each similar steel grade also need to be statistically analyzed, and the compliance rate of the historical performance of each steel grade is calculated in combination with the performance requirement range of the target steel grade, and the similar steel grade with the highest compliance rate is preferentially selected as the reference steel grade.

6. The method for designing the steel grade of hot-rolled products based on a performance prediction model according to claim 5, characterized in that, In the step S3, the element composition design information includes the maximum values, minimum values, and target values of the element composition parameters of C, SI, Mn, P, S, AL, B, N, V, Nb, Ti, Cu, Ni, Mo, Cr, H, W, Zr, O, As, Sn, and Ca; The hot rolling process design information includes the maximum values, minimum values, and target values of the process parameters of the tapping temperature, rough rolling temperature, finish rolling temperature, and coiling temperature.

7. The hot rolling product steel type design method based on the performance prediction model according to claim 6, characterized in that, The step S4 specifically includes: Specify the composition requirements (X 1 , X 2 ,,, X n ) specified for the target steel grade as optimization parameters, and then, in combination with the correlation rules between composition and process, select the remaining composition, hot rolling process, and specifications as optimization parameters, and specify different calculation step sizes for the composition, process, and specification parameters.

8. The hot rolling product steel type design method based on the performance prediction model according to claim 7, characterized in that, The step S5 specifically includes: The maximum value and minimum value of the specified composition of the target steel type are comprehensively given by taking the intersection of the specified composition range of the target steel type and the composition range of the reference steel type, and the maximum value and minimum value of the optimization parameters of the remaining components, processes, and specifications are given according to the design values of the reference steel type; According to the maximum value, minimum value, and calculation step of the optimization parameters of the composition, hot rolling process, and thickness specification, M groups of optimization combinations composed of different composition, hot rolling process, and thickness specification data are divided, and the mechanical properties of the M groups of optimization combinations are predicted by using the performance prediction model.

9. The hot rolling product steel type design method based on the performance prediction model according to claim 8, characterized in that, The mechanical properties include yield strength, tensile strength, elongation, and impact energy.

10. The hot rolling product steel type design method based on the performance prediction model according to claim 9, characterized in that: For the index prediction results of each of the mechanical properties [(P_YS, P_YS_STD), (P_TS, P_TS_STD), (P_EL, P_EL_STD), (P_EA, P_EA_STD),,,], combined with the performance requirements of the target steel type [[YS_min, YS_max], [TS_min, TS_max], [EL_min, EL_max], [EA_min, EA_max],,,], calculate the compliance rate Qr_P_Y of the index prediction results of each of the mechanical properties = min[(Y_max - P_Y) / (3*P_Y_STD), (P_Y - Y_min) / (3*P_Y_STD)]; Take the minimum value of the compliance rates in the index prediction results of each of the mechanical properties as the comprehensive prediction performance compliance rate Qr_Com_P of the optimization combination = min[Qr_P_YS, Qr_P_TS, Qr_P_EL, Qr_P_EA,...], and give a compliance rate threshold Qr_lim, and screen the optimization combinations with a comprehensive compliance rate greater than the threshold; In the formula, YS represents the yield strength, TS represents the tensile strength, EL represents the fracture elongation, and EA represents the impact energy.

Images