Non-oriented silicon steel SACL energy-saving method based on magnetic property prediction model

Abstract

The application discloses a non-oriented silicon steel SACL energy-saving method based on a magnetic performance prediction model, and is characterized in that a magnetic performance prediction model of non-oriented silicon steel is established, before a SACL continuous annealing plan is issued, steelmaking components, steelmaking process actual values, hot rolling process actual values, normalizing process actual values, SACL process parameter set values, SACL process parameter threshold values, performance contract requirement values, internal control requirement values and specification parameters are extracted, the magnetic performance prediction model is substituted, the magnetic performance indexes of the non-oriented silicon steel are predicted, annealing temperature adjustment of a SACL annealing furnace of the non-oriented silicon steel is carried out, and energy saving and emission reduction of the SACL annealing furnace are realized. In the heating and soaking temperature range allowed by the metallurgical mechanism, the annealing furnace heating or soaking section temperature is optimized and adjusted based on the magnetic performance prediction model, the magnetic performance prediction value, the contract performance requirement and the factory performance internal control requirement, and the purpose of temperature reduction and energy saving is realized.

Description

Technical Field

[0001] This invention relates to temperature control technology in the production process of non-oriented silicon steel (SACL) products in the iron and steel metallurgy industry, and more specifically, to an energy-saving method for non-oriented silicon steel (SACL) based on a magnetic property prediction model. Background Technology

[0002] Cold-rolled non-oriented silicon steel is a soft magnetic material and the most widely used alloy material among magnetic materials, playing a vital role in the power, electronics, and military industries. With the increasing global awareness of energy conservation and emission reduction, thinner profiles, lower iron loss, higher magnetic induction, and higher mechanical properties have become the mainstream development trend, indicating huge market potential.

[0003] Annealing is a heat treatment technique primarily used to improve the physical and chemical properties of metals and certain other materials. It involves heating the material to a specific temperature, holding it for a period of time, and then cooling it at an appropriate rate. The main purpose of annealing is:

[0004] (1) Relieving internal stress. Internal stress may be generated during metal processing, such as casting, forging, rolling, and welding. Annealing can reduce hardness and increase the ductility and toughness of the material, thereby helping to eliminate these internal stresses.

[0005] (2) Improve the microstructure of materials. Annealing can change the crystal structure of materials, promote the recovery and reformation of ordered crystal structures, and thus improve the mechanical properties of materials.

[0006] (3) Prepare for subsequent heat treatment processes. Annealing is a necessary step in preparing the microstructure for subsequent heat treatment processes, such as quenching, tempering, or aging.

[0007] For non-oriented silicon steel products, the SACL (Silicon Steel Continuous Annealing and Coating Unit) annealing temperature has a crucial impact on the final performance of the product.

[0008] During operation, the annealing furnace uses heat generated by burning coal gas or natural gas in radiant tubes in the heating section to radiate heat the strip steel, while the soaking section is heated by resistance wires on both sides. This process promotes a continuous circulation of heat within the furnace, ensuring temperature uniformity within the furnace cavity. Therefore, the higher the annealing temperature setting, the higher the energy consumption to achieve stable furnace temperature.

[0009] Typically, the target annealing temperature is entered into the computer by process engineers before the SACL rolling schedule is issued. Therefore, setting a reasonable target annealing temperature not only ensures the final performance of the product but also reduces energy consumption, production costs, and carbon emissions.

[0010] With the rapid development of computer technology, big data analytics, with its strong information processing capabilities and advantages such as real-time and predictive analysis, is widely used in various industries, including finance and industrial production. Data models built by mining potential information from big data can, to some extent, replace traditional mechanistic models in guiding production. Therefore, attempting to optimize SACL annealing temperature using data models to reduce temperature, thereby lowering energy consumption and costs, is a feasible approach.

[0011] In the field of energy-saving control during the annealing of non-oriented silicon steel, research is currently limited. For example, CN114139350A discloses a method for predicting iron loss in the continuous annealing process of non-oriented silicon steel. By inputting the composition and process parameters into the regression equation of the established prediction model, the predicted iron loss value is obtained. Adjustments are made for cases that do not meet the target range, and this model can improve product quality. CN115261600B discloses an artificial intelligence automatic control method for annealing furnace tension. Through three steps—an inter-roller tension control model, an in-furnace equal tension distribution model, and a genetic algorithm tension control logic—the method accurately outputs the target value of inter-roller tension, controls the tension distribution in the furnace using an equal tension model, and finally uses a genetic algorithm for intelligent iterative tension control, improving control accuracy and enhancing the magnetic properties of the actual product. CN116536486A discloses an energy-saving bell-type annealing furnace system and its usage method. This annealing system includes at least two bell-type annealing furnaces, which utilize internal fans to circulate and exchange gas and transfer heat between the furnace platforms, achieving secondary heat utilization. It is simple to operate, energy-saving, and environmentally friendly. For example, CN106591555A discloses an annealing process for cold-rolled non-oriented cold-rolled silicon steel sheets. The process involves two annealing treatments to heat-treat the silicon steel sheets. The temperatures and furnace atmospheres used in the two annealing treatments are different, which can greatly improve the heat treatment effect of the silicon steel sheets.

[0012] Overall, there is currently little research on energy-saving control of annealing for non-oriented silicon steel. The main research focuses on specific annealing steps and the design of annealing furnaces, as mentioned in the patented technologies mentioned above. There is also a small amount of research on parameter models in annealing, but very little research on annealing temperature control. Summary of the Invention

[0013] To address the shortcomings of existing technologies, the purpose of this invention is to provide an energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model. Within the heating and soaking temperature range allowed by metallurgical mechanisms, the method optimizes and adjusts the heating or soaking temperature of the annealing furnace based on the magnetic property prediction model, the predicted magnetic property value, contract performance requirements, and the internal control requirements of the plant performance. This reduces the annealing heating or soaking temperature, thereby reducing the energy consumption per unit product of the annealing furnace and achieving the goal of cooling and energy saving.

[0014] To achieve the above objectives, the present invention adopts the following technical solution:

[0015] An energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model;

[0016] Based on the established magnetic property prediction model for non-oriented silicon steel, before the issuance of the SACL continuous annealing plan, the following data are extracted: steelmaking composition, actual steelmaking process values, actual hot rolling process values, actual normalizing process values, SACL process parameter setpoints, SACL process parameter thresholds, performance contract requirements, internal control requirements, and specification parameters. These data are then substituted into the magnetic property prediction model to predict the magnetic property indicators of the non-oriented silicon steel. This allows for the adjustment of the annealing temperature in the SACL annealing furnace for the non-oriented silicon steel, thereby achieving energy conservation and emission reduction in the SACL annealing furnace.

[0017] Preferably, the steelmaking composition includes Si, Al, Mn, P, Ti, S, C, N, Nb, and V;

[0018] The actual values ​​of the steelmaking process include ladle slag thickness and final free oxygen after decarburization.

[0019] The actual values ​​of the hot rolling process include furnace inlet temperature, furnace time, furnace outlet temperature, roughing temperature, finishing temperature, and coiling temperature;

[0020] The actual values ​​of the normalization process include the furnace temperature of the heating section and the furnace temperature of the soaking section;

[0021] The SACL process parameters are set as follows: annealing rate, furnace temperature in the annealing soaking zone, furnace temperature in the annealing heating zone, and annealing furnace tension.

[0022] The SACL process parameter thresholds include the upper and lower limits of the SACL process parameters;

[0023] The performance contract requirements include the minimum magnetic flux density and the maximum iron loss.

[0024] The internal control requirements include upper limits, lower limits, and target values ​​for performance indicators;

[0025] The specifications include the thickness and width of the strip.

[0026] Preferably, the SACL process parameter thresholds are configured individually based on the steel type and grade of the non-oriented silicon steel.

[0027] The performance contract requirement value is the release requirement set by the user and is determined according to different user needs;

[0028] The internal control requirements are based on performance control requirements formed through long-term actual production and must be stricter than the performance contract requirements.

[0029] Preferably, the magnetic performance prediction model predicts the magnetic performance indicators of the non-oriented silicon steel, including steel loss and magnetic induction.

[0030] Preferably, the calculation of the magnetic property prediction model specifically includes:

[0031] Y = f(steelmaking composition, actual steelmaking process value, actual hot rolling process value, actual normalizing process value, SACL process parameter setting value)

[0032] Where Y is the magnetic property index to be predicted.

[0033] Preferably, the temperature control of the SACL annealing furnace for the non-oriented silicon steel specifically includes:

[0034] Based on the performance contract requirements and the internal control requirements, and using the magnetic property prediction model to predict the relationship between the magnetic property indicators of the non-oriented silicon steel and the temperature of the SACL annealing furnace, it is determined whether the annealing temperature can be adjusted. If so, optimization is performed; otherwise, production is carried out according to the original process.

[0035] Preferably, optimizing the annealing temperature specifically includes:

[0036] Based on the predicted values ​​and boundary values, combined with the magnetic property prediction model and the threshold of the SACL annealing temperature, an optimization objective function and constraints are established, and an optimization algorithm is used to solve the problem to minimize the objective function value, thereby obtaining the optimal value of the annealing temperature of the SACL annealing furnace.

[0037] Preferably, establishing the optimization objective function specifically includes:

[0038] Delta_Y1=(P_Y1-Y1_BOUND)^2

[0039] Delta_Y2=(P_Y2-Y2_BOUND)^2

[0040] Delta_Y3=(P_Y3-Y3_BOUND)^2 ...

[0042] F_LOSS=Delta_Y1+m1*Delta_Y2+...

[0043] Wherein, P_Y1 is the predicted value of performance indicator Y1; Y1_BOUND is the boundary value of performance indicator Y1, i.e., the upper or lower limit, the specific upper or lower limit is determined by the indicator and its contract or internal control requirements; Delta_Y1 is used to measure the deviation between the predicted value and the boundary value; the meanings of other performance indicator parameters such as Y2 are the same as those of Y1.

[0044] F_LOSS is the loss function, which is a weighted sum of the deviation values ​​of each performance index, serving as the optimization objective function; where the parameter m is the weighting coefficient for balancing the deviation values ​​of multiple performance indices, which is determined by the specific steel grade and grade.

[0045] Preferably, the optimization constraints include:

[0046] (1) Y1 = f1 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters);

[0047] (2) Y2 = f2 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters);

[0048] (3) Y3 = f3 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters);

[0049] (4) HS_MIN <= HS <= HS_MAX;

[0050] (5) SS_MIN <= SS <= SS_MAX;

[0051] (6) Y1_MIN <= Y1 <= Y1_MAX;

[0052] (7) Y2_MIN <= Y2 <= Y2_MAX;

[0053] (8) Y3_MIN <= Y3 <= Y3_MAX;

[0054] Among them, equations (1), (2), and (3) are magnetic property prediction models for Y1, Y2, and Y3, which are three performance indicators;

[0055] Equations (4) and (5) are the temperature constraints of the SACL annealing furnace, where HS is the heating section temperature and SS is the soaking section temperature; HS_MIN, HS_MAX, SS_MIN, and SS_MAX are the minimum, maximum, minimum, and maximum temperatures of the heating section, respectively, and are all specified by the SACL process parameter thresholds.

[0056] Equations (6), (7), and (8) define the range of magnetic performance indicators, where Y_MIN is the minimum value of the magnetic performance indicator and Y_MAX is the maximum value of the magnetic performance indicator.

[0057] This invention provides an energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model. Within the metallurgical mechanism-permissible heating and soaking temperature range, based on the magnetic property prediction model, predicted magnetic property values, contract performance requirements, and internal plant performance control requirements, the method optimizes and adjusts the heating or soaking temperature of the annealing furnace, thereby reducing the annealing heating or soaking temperature and lowering the energy consumption per unit product of the annealing furnace, achieving the goal of cooling and energy saving. This control method covers a wide range of production lines and steel grades and can be widely applied in the field of temperature control for non-oriented silicon steel SACL annealing furnaces. Attached Figure Description

[0058] Figure 1 This is a schematic flowchart of the energy-saving method for non-oriented silicon steel SACL of the present invention;

[0059] Figure 2 This is a schematic flowchart of Example 1 of the energy-saving method for non-oriented silicon steel SACL of the present invention. Detailed Implementation

[0060] To better understand the above-mentioned technical solutions of the present invention, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0061] Combination Figure 1 As shown, this invention provides an energy-saving method for non-oriented silicon steel SACL based on a magnetic performance prediction model;

[0062] Based on the establishment of a magnetic property prediction model for non-oriented silicon steel, before the issuance of the SACL continuous annealing plan, the following data are extracted: steelmaking composition, actual values ​​of steelmaking process, actual values ​​of hot rolling process, actual values ​​of normalizing process, SACL process parameter setpoints, SACL process parameter thresholds, performance contract requirements, internal control requirements, and specification parameters. These data are then substituted into the magnetic property prediction model to predict the magnetic property indicators of non-oriented silicon steel. This allows for the adjustment of the annealing temperature in the SACL annealing furnace for non-oriented silicon steel, thereby achieving energy conservation and emission reduction in the SACL annealing furnace.

[0063] The steelmaking components include elements such as Si, Al, Mn, P, Ti, S, C, N, Nb, and V.

[0064] Actual values ​​of steelmaking processes include ladle slag thickness, final free oxygen after decarburization, etc.

[0065] The actual values ​​of hot rolling process include furnace entry temperature, furnace time, furnace exit temperature, roughing rolling temperature, finishing rolling temperature, coiling temperature, etc.

[0066] The actual values ​​of the normalizing process include the furnace temperature in the heating section and the furnace temperature in the soaking section;

[0067] SACL process parameter settings include annealing rate, annealing soaking zone furnace temperature, annealing heating zone furnace temperature, and annealing furnace tension.

[0068] SACL process parameter thresholds include the upper and lower limits of SACL process parameters;

[0069] The performance contract requirements include the minimum magnetic flux density and the maximum iron loss.

[0070] Internal control requirements include the upper limit, lower limit, and target value (i.e., historical average) of performance indicators;

[0071] Specifications include the thickness and width of the strip.

[0072] SACL process parameter thresholds are configured individually based on the steel type and grade of non-oriented silicon steel.

[0073] The performance contract requirement value is the release requirement set by the user and is determined according to different user needs;

[0074] The internal control requirements are based on performance control requirements formed in long-term actual production and must be stricter than the performance contract requirements.

[0075] The magnetic property prediction model predicts the magnetic property indicators of non-oriented silicon steel, including steel loss and magnetic induction.

[0076] The calculation of the magnetic property prediction model specifically includes:

[0077] Y = f(steelmaking composition, actual steelmaking process value, actual hot rolling process value, actual normalizing process value, SACL process parameter setting value)

[0078] Where Y is the magnetic property index to be predicted.

[0079] The specific temperature control of the SACL annealing furnace for non-oriented silicon steel includes:

[0080] Combining performance contract requirements and internal control requirements, the relationship between the magnetic properties of non-oriented silicon steel and the temperature of the SACL annealing furnace is predicted based on the magnetic property prediction model. It is then determined whether the annealing temperature can be adjusted. If so, optimization is carried out; otherwise, production is carried out according to the original process.

[0081] Optimizing the annealing temperature specifically includes:

[0082] For steel coils whose annealing temperature can be optimized, process parameters are optimized. To ensure that multiple performance indicators meet contractual performance requirements while minimizing the annealing temperature, an optimization objective function and constraints are established based on predicted and boundary values, combined with a magnetic property prediction model and the threshold of the SACL annealing temperature. Optimization algorithms (such as gradient descent, Newton's method, and simulated annealing) are then used to solve the problem, minimizing the objective function value and thus obtaining the optimal annealing temperature for the SACL annealing furnace.

[0083] Specifically, establishing the optimization objective function includes:

[0084] Delta_Y1=(P_Y1-Y1_BOUND)^2

[0085] Delta_Y2=(P_Y2-Y2_BOUND)^2

[0086] Delta_Y3=(P_Y3-Y3_BOUND)^2 ...

[0088] F_LOSS=Delta_Y1+m1*Delta_Y2+...

[0089] Wherein, P_Y1 is the predicted value of performance indicator Y1; Y1_BOUND is the boundary value of performance indicator Y1, i.e., the upper or lower limit, the specific upper or lower limit is determined by the indicator and its contract or internal control requirements; Delta_Y1 is used to measure the deviation between the predicted value and the boundary value; the meaning of other performance indicator parameters such as Y2 is the same as that of Y1; there can be one or more performance indicators.

[0090] F_LOSS is the loss function, which is a weighted sum of the deviation values ​​of each performance index, serving as the optimization objective function; where the parameter m is the weighting coefficient for balancing the deviation values ​​of multiple performance indices, which is determined by the specific steel grade and grade.

[0091] Establishing optimization constraints specifically includes:

[0092] (1) Y1 = f1 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters);

[0093] (2) Y2 = f2 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters);

[0094] (3)Y3=...

[0095] (4) HS_MIN <= HS <= HS_MAX;

[0096] (5) SS_MIN <= SS <= SS_MAX;

[0097] (6) Y1_MIN <= Y1 <= Y1_MAX;

[0098] (7) Y2_MIN <= Y2 <= Y2_MAX;

[0099] (8)...Y3...

[0100] Equations (1), (2), and (3) are magnetic property prediction models for three performance indicators, Y1, Y2, and Y3; there may be one or more performance indicators.

[0101] Equations (4) and (5) are the temperature constraints of the SACL annealing furnace, where HS is the temperature of the heating section and SS is the temperature of the soaking section; HS_MIN, HS_MAX, SS_MIN, and SS_MAX are the minimum, maximum, minimum, and maximum temperatures of the heating section, respectively, and are all specified by the threshold values ​​of the SACL process parameters.

[0102] Equations (6), (7), and (8) define the range of magnetic performance indicators, where Y_MIN is the minimum value of the magnetic performance indicator and Y_MAX is the maximum value of the magnetic performance indicator.

[0103] Example 1

[0104] Combination Figure 2 As described in Embodiment 1, an energy-saving method for non-oriented silicon steel SACL based on a magnetic performance prediction model is provided, including the following steps:

[0105] 1) Before the SACL annealing plan is issued, retrieve the steelmaking composition and process performance data, hot rolling performance data, normalizing performance data (do not retrieve normalizing process parameters for steel coils that have not undergone normalizing), SACL process parameter setpoints, SACL process parameter thresholds, performance contract requirements, internal control requirements, specifications, etc.

[0106] 2) Substitute the above data into the magnetic property prediction model to predict the magnetic properties of the non-oriented silicon steel that is planned to undergo SACL annealing;

[0107] 3) Calculate the deviation delta_XNi between each performance prediction value P_XNi and the performance target boundary value BJ_XNi, and then calculate the adjustment value of the annealing furnace temperature, delta_Ti=delta_XNi / k_Ti. If delta_Ti>ΔT exists, the annealing furnace temperature can be optimized.

[0108] Where k_Ti is the coefficient corresponding to the annealing temperature in the model; ΔT is the minimum adjustment amount of the annealing temperature, which is determined by the temperature control requirements of the production site.

[0109] 4) If there exists |delta_Ti|>0, meaning the steel coil can be optimized for annealing temperature, then the process parameters are optimized. To ensure that the final multiple performance indicators meet the contract performance requirements and minimize the annealing temperature, based on the performance prediction values ​​and boundary values, combined with the magnetic property prediction model, the threshold of SACL annealing temperature, and the weight relationship of multiple performance indicators, an optimization objective function and constraints are established. An optimization algorithm such as gradient descent is used to solve the problem, minimizing the objective function value, thereby obtaining the adjusted SACL annealing temperature value.

[0110] 5) Substitute the new SACL section temperature into the performance prediction model, and produce rolls with qualified performance according to the new process;

[0111] 6) For rolls whose forecast values ​​are outside the performance requirements or whose performance is unqualified after adjustment, they shall be produced according to the original process.

[0112] Example 2

[0113] This Example 2 involves the entire production process of non-oriented silicon steel products in a steel plant, and further explains the steel coils produced by a certain production line.

[0114] The performance prediction model used in this embodiment 2 is as follows:

[0115] TS=f1(Si,Al,Mn,P,Ti,S,C,N,FRN,FT,CT,NOF,SF,HS,SS,Speed,Thick);

[0116] CG=f2(Si,Al,Mn,P,Ti,S,C,N,FRN,FT,CT,NOF,SF,HS,SS,Speed,Thick);

[0117] TS and CG are the performance indicators of iron loss and magnetic induction, respectively.

[0118] Where Si, Al, Mn, P, Ti, S, C, and N represent the content of silicon, aluminum, manganese, phosphorus, titanium, sulfur, carbon, and nitrogen, respectively.

[0119] Wherein FRN, FT, CT, NOF, SF, HS, SS, Speed, and Thick represent the furnace exit temperature, finishing rolling temperature, coiling temperature, furnace temperature of the normalizing and annealing heating section, furnace temperature of the normalizing and annealing soaking section, furnace temperature of the high-temperature zone of the SACL continuous annealing furnace heating section, furnace temperature of the SACL continuous annealing furnace soaking section, annealing speed, and SACL exit thickness, respectively.

[0120] The actual chemical composition of the selected non-oriented silicon steel coil 1 is shown in the table below:

[0121] Si_ACT AL_ACT Mn_ACT P_ACT 0.928％ 0.4256％ a1％ b1％ Ti_ACT S_ACT C_ACT N_ACT c1％ d1％ e1％ f1％

[0122] Wherein, Si_ACT, Al_ACT, Mn_ACT, P_ACT, Ti_ACT, S_ACT, C_ACT, and N_ACT represent the actual values ​​of silicon, aluminum, manganese, phosphorus, titanium, sulfur, carbon, and nitrogen content, respectively.

[0123] The actual values ​​for the hot rolling process of steel coil 1 are shown in the table below:

[0124] Furnace temperature Final rolling temperature winding temperature frn_r1 ft_r1 ct_r1

[0125] The production of steel coil 1 does not involve normalization, therefore the normalization process parameters are not listed.

[0126] The SACL process design values ​​and process range requirements for steel coil 1 are as follows:

[0127] HS_AIM T_hs_aim1 HS_MIN T_hs_min1 HS_MAX T_hs_max1 SS_AIM T_ss_aim1 SS_MIN T_ss_min1 SS_MAX T_ss_max1

[0128] Among them, HS_AIM, HS_MIN, and HS_MAX are the setpoint, minimum, and maximum furnace temperature of the high-temperature zone in the heating section of the SACL continuous annealing furnace, respectively; SS_AIM, SS_MIN, and SS_MAX are the setpoint, minimum, and maximum furnace temperature of the soaking section of the SACL continuous annealing furnace, respectively.

[0129] The contract performance requirements for steel coil 1 are as follows:

[0130] Iron loss (TS) TS≤5.50 Magnetism (CG) CG≥1.72

[0131] The internal control requirements for steel coil 1 are as follows:

[0132] TS_MIN 4.70 TS_MAX 5.20 CG_MIN 1.720 CG_MAX 1.750

[0133] TS_MIN and TS_MAX are the minimum and maximum values ​​required for iron loss, respectively; CG_MIN and CG_MAX are the minimum and maximum values ​​required for magnetic induction control, respectively.

[0134] The target thickness of the finished steel coil 1 is 0.50 mm, and the target speed of the central section of SACL is v1 m / min.

[0135] Since the internal control requirements are within the scope of the contract, only the internal control requirements need to be considered.

[0136] Before the SACL plan is issued, data such as steelmaking composition, hot rolling process, SACL process target value, and finished product thickness target value are input into the magnetic performance prediction model. The predicted values ​​of indicators such as iron loss and magnetic induction are calculated as follows:

[0137] Iron loss forecast value P_TS Magnetic induction prediction value P_CG 5.012478 1.736237

[0138] Calculate the prediction bias of iron loss and magnetic induction:

[0139] Delta_TS=5.012478-5.20=-0.187522

[0140] Delta_CG=1.736237-1.72=0.016237

[0141] The adjustment amounts for heating temperature and soaking temperature are calculated based on the deviation, respectively:

[0142] Delta_HS_TS=Delta_TS / k_hs_ts=-0.187522 / -0.005086=36.87

[0143] Delta_HS_CG=Delta_CG / k_hs_cg=0.016237 / -0.00008867=-183.11

[0144] Delta_SS_TS=Delta_TS / k_ss_ts=-0.187522 / -0.004065138=46.13

[0145] Delta_SS_CG=Delta_CG / k_ss_cg=0.016237 / -0.00006854=-236.89 In actual production, the SACL temperature control system requires that the SACL temperature be set in increments of 5°C. Therefore, the temperature can only be adjusted when the adjustable range is greater than or equal to 5°C.

[0146] Since the absolute values ​​of the calculated heating temperature and homogenization temperature adjustment amounts Delta_HS_TS, Delta_HS_CG, Delta_SS_TS, and Delta_SS_CG are all greater than 5, it is possible to try adjusting the annealing furnace temperature.

[0147] First, the differences between the predicted iron loss and the maximum iron loss, and the predicted magnetic flux density and the minimum magnetic flux density, are calculated separately. Then, the two performance indicators are summed according to their weights to obtain the target loss function:

[0148] Delta_TS=(P_TS-TS_MAX)^2

[0149] Delta_CG=(P_CG-CG_MIN)^2

[0150] F_LOSS = Delta_TS + m * Delta_CG

[0151] Where m is the coefficient for balancing iron loss and magnetic induction weights, which is determined by the specific steel grade and grade; when the loss function LOSS is minimized, the comprehensive index of iron loss and magnetic induction is closer to the boundary value, and the obtained temperature can not only meet the performance requirements of the contract and internal control, but also be the most energy-efficient temperature scheme.

[0152] Based on the relationship between SACL process parameters and performance indicators, as well as the threshold values ​​of SACL process parameters and the requirements of performance indicators, the constraints are determined as follows:

[0153] (1)TS=f1(Si,Al,Mn,P,Ti,S,C,N,FRN,FT,CT,HS,SS,Speed,Thick);

[0154] (2)CG=f2(Si,Al,Mn,P,Ti,S,C,N,FRN,FT,CT,HS,SS,Speed,Thick);

[0155] (3) HS_MIN <= HS <= HS_MAX;

[0156] (4) SS_MIN <= SS <= SS_MAX;

[0157] (5) TS_MIN <= TS <= TS_MAX;

[0158] (6) CG_MIN <= CG <= CG_MAX.

[0159] Solving this optimization problem using the gradient descent method yields two optimal design values ​​for furnace temperatures. Rounding these values ​​by 5°C, we obtain the following temperature values:

[0160] HS = hs_c1 SS = ss_c1

[0161] Substituting the new SACL process temperature into the performance prediction model, the new predicted performance is calculated as follows:

[0162] Iron loss forecast value P_TS Magnetic induction prediction value P_CG 5.12941938 1.73825245 Iron loss standard deviation TS_STD Magnetic induction standard deviation CG_STD 0.038058 0.003081

[0163] The optimized predicted values ​​of iron loss and magnetic induction both meet the internal control requirements. However, since there may be some deviation between the model prediction value and the actual value of magnetic properties, when using the predicted value of magnetic properties to determine whether the performance is qualified, we do not directly judge whether the predicted performance is qualified based on the upper and lower limits. Instead, we judge by calculating the probability that the predicted value meets the upper and lower limits of the performance requirements, and choose the probability of ±2σ (i.e., 0.9545) to judge whether the predicted performance is qualified.

[0164] The formula for the pass rate is as follows:

[0165] CDF_Y=CDF('NORMAL',(Y_MAX-P_Y) / Y_STD)-CDF('NORMAL',(Y_MIN-P_Y) / Y_STD);

[0166] Where Y represents the performance metric, which can be TS or CG;

[0167] Y_MAX, Y_MIN, P_Y, and Y_STD represent the maximum performance value, minimum performance value, predicted performance value, and predicted performance standard deviation, respectively.

[0168] Substituting the performance requirements and predicted performance values, the pass rates for different performance indicators are calculated:

[0169] CDF_TS=0.9682; CDF_CG=0.9999;

[0170] Both CDF_TS and CDF_CG are greater than 0.9545, meaning that both performance indicators are qualified after optimizing and adjusting the SACL process temperature. Therefore, the SACL process optimization is successful, and the SACL process can be produced according to the new process.

[0171] Adjusted SACL heating section temperature Adjusted SACL soaking temperature hs_c1 ss_c1

[0172] Example 3

[0173] The actual content of the steelmaking chemical composition of the selected non-oriented steel coil 2 is shown in the table below:

[0174] Si_ACT AL_ACT Mn_ACT P_ACT 3.345％ 0.702％ a2％ b2％ Ti_ACT S_ACT C_ACT N_ACT c2％ d2％ e2％ f2％

[0175] The actual values ​​for the hot rolling process of steel coil 2 are shown in the table below:

[0176] Furnace temperature Final rolling temperature winding temperature frn_r2 ft_r2 ct_r2

[0177] The normalizing process performance values ​​for steel coil 2 are shown in the table below:

[0178] Normalizing heating section temperature Normalizing temperature of the homogenization zone nof_r2 sf_r2

[0179] The SACL process design values ​​and process range requirements for steel coil 2 are as follows:

[0180] HS_AIM T_hs_aim2 HS_MIN T_hs_min2 HS_MAX T_hs_max2 SS_AIM T_ss_aim2 SS_MIN T_ss_min2 SS_MAX T_ss_max2

[0181] The contract performance requirements for steel coil 2 are as follows:

[0182]

[0183]

[0184] The internal control requirements for steel coil 2 are as follows:

[0185] TS_MIN 2.05 TS_MAX 2.28 CG_MIN 1.62 CG_MAX 1.66

[0186] The target thickness of the finished steel coil 2 is 0.27 mm, and the target speed of the central section of SACL is v2 m / min.

[0187] Since the internal control requirements are within the scope of the contract, only the internal control requirements need to be considered.

[0188] Before the SACL plan is issued, data such as steelmaking composition, hot rolling process, SACL process target value, and finished product thickness target value are input into the magnetic performance prediction model. The predicted values ​​of indicators such as iron loss and magnetic induction are calculated as follows:

[0189] Iron loss forecast value P_TS Magnetic induction prediction value P_CG 2.149428 1.633435

[0190] Calculate the prediction bias of iron loss and magnetic induction:

[0191] Delta_TS=2.149428-2.28=-0.130572

[0192] Delta_CG=1.633435-1.62=0.013435

[0193] The adjustment amounts for heating temperature and soaking temperature are calculated based on the deviation, respectively:

[0194] Delta_HS_TS=Delta_TS / k_hs_ts=-0.130572 / -0.00170917=76.39

[0195] Delta_HS_CG=Delta_CG / k_hs_cg=0.013435 / -0.000090395=-148.62

[0196] Delta_SS_TS=Delta_TS / k_ss_ts=-0.130572 / -0.0011671=111.87

[0197] Delta_SS_CG=Delta_CG / k_ss_cg=0.013435 / -0.00008972=-149.74 According to the requirements of the on-site SACL temperature control system, the SACL temperature is set in increments of 5℃.

[0198] Since the absolute values ​​of the calculated heating temperature and homogenization temperature adjustment amounts Delta_HS_TS, Delta_HS_CG, Delta_SS_TS, and Delta_SS_CG are all greater than 5, it is possible to try adjusting the annealing furnace temperature.

[0199] First, the differences between the predicted iron loss and the maximum iron loss, and the predicted magnetic flux density and the minimum magnetic flux density, are calculated separately. Then, the two performance indicators are summed according to their weights to obtain the target loss function:

[0200] Delta_TS=(P_TS-TS_MAX)^2

[0201] Delta_CG=(P_CG-CG_MIN)^2

[0202] F_LOSS = Delta_TS + m * Delta_CG

[0203] Where m is a coefficient for balancing iron loss and magnetic induction weight, which is determined by the specific steel grade and grade;

[0204] Based on the relationship between SACL process parameters and performance indicators, as well as the threshold values ​​of SACL process parameters and the requirements of performance indicators, the constraints are determined as follows:

[0205] (1)TS=f1(Si,Al,Mn,P,Ti,S,C,N,FRN,FT,CT,HS,SS,Speed,Thick);

[0206] (2)CG=f2(Si,Al,Mn,P,Ti,S,C,N,FRN,FT,CT,HS,SS,Speed,Thick);

[0207] (3) HS_MIN <= HS <= HS_MAX;

[0208] (4) SS_MIN <= SS <= SS_MAX;

[0209] (5) TS_MIN <= TS <= TS_MAX;

[0210] (6) CG_MIN <= CG <= CG_MAX.

[0211] Solving this optimization problem using the gradient descent method yields two optimal design values ​​for furnace temperatures. Rounding these values ​​by 5°C, we obtain the following temperature values:

[0212] HS = hs_c2 SS = ss_c2

[0213] Substituting the new SACL process temperature into the performance prediction model, the new predicted performance is calculated as follows:

[0214] Iron loss forecast value P_TS Magnetic induction prediction value P_CG 2.206953 1.637037 Iron loss standard deviation TS_STD Magnetic induction standard deviation CG_STD 0.025645 0.002556

[0215] Substituting the performance requirements and predicted performance values, the pass rates for different performance indicators are calculated:

[0216] CDF_TS=0.9978; CDF_CG=0.9999;

[0217] Both CDF_TS and CDF_CG are greater than 0.9545, meaning that both performance indicators are qualified after optimizing and adjusting the SACL process temperature. Therefore, the SACL process optimization is successful, and the SACL process can be produced according to the new process.

[0218] Adjusted SACL heating section temperature Adjusted SACL soaking temperature hs_c2 ss_c2

[0219] Example 4

[0220] The actual chemical composition of the selected non-oriented steel coil 3 is shown in the table below:

[0221] Si_ACT AL_ACT Mn_ACT P_ACT 2.5290％ 0.3761％ a3％ b3％ Ti_ACT S_ACT C_ACT N_ACT c3％ d3％ e3％ f3％

[0222] The actual values ​​for the hot rolling process of steel coil 3 are shown in the table below:

[0223] Furnace temperature Final rolling temperature winding temperature frn_r3 ft_r3 ct_r3

[0224] The normalizing process performance values ​​for steel coil 3 are shown in the table below:

[0225] Normalizing heating section temperature Normalizing temperature of the homogenization zone nof_r3 sf_r3

[0226] The SACL process design values ​​and process range requirements for steel coil 3 are as follows:

[0227] HS_AIM T_hs_aim3 HS_MIN T_hs_min3 HS_MAX T_hs_max3 SS_AIM T_ss_aim3 SS_MIN T_ss_min3 SS_MAX T_ss_max3

[0228] The contract performance requirements for steel coil 3 are as follows:

[0229] Iron loss (TS) TS≤3.30 Magnetism (CG) CG≥1.66

[0230] The internal control requirements for steel coil 3 are as follows:

[0231] TS_MIN 2.72 TS_MAX 3.00 CG_MIN 1.67 CG_MAX 1.70

[0232] The target thickness of the finished steel coil 3 is 0.50 mm, and the target speed of the central section of SACL is v3 m / min.

[0233] Since the internal control requirements are within the scope of the contract, only the internal control requirements need to be considered.

[0234] Before the SACL plan is issued, data such as steelmaking composition, hot rolling process, SACL process target value, and finished product thickness target value are input into the magnetic performance prediction model. The predicted values ​​of indicators such as iron loss and magnetic induction are calculated as follows:

[0235] Iron loss forecast value P_TS Magnetic induction prediction value P_CG 2.988642 1.682236

[0236] Calculate the prediction bias of iron loss and magnetic induction:

[0237] Delta_TS=2.988642-3.00=-0.011358

[0238] Delta_CG=1.682236-1.67=0.012236

[0239] The adjustment amounts for heating temperature and soaking temperature are calculated based on the deviation, respectively:

[0240] Delta_HS_TS=Delta_TS / k_hs_ts=-0.011358 / -0.0019525=5.82

[0241] Delta_HS_CG=Delta_CG / k_hs_cg=0.012236 / -0.000401242=-30.49

[0242] Delta_SS_TS=Delta_TS / k_ss_ts=-0.011358 / -0.00301768=3.76

[0243] Delta_SS_CG=Delta_CG / k_ss_cg=0.012236 / -0.000202895=-60.30 As required by the on-site SACL temperature control system, the SACL temperature is set in increments of 5℃.

[0244] Since the calculated heating temperature and the adjustment amount for the homogenization temperature are both less than 5 (Delta_SS_TS), the temperature of the homogenization section cannot be adjusted; while Delta_HS_TS is 5.82, so we can try to reduce the heating section temperature by 5℃.

[0245] Substituting the new SACL process temperature into the performance prediction model, the new predicted performance is calculated as follows:

[0246] Iron loss forecast value P_TS Magnetic induction prediction value P_CG 3.0037304 1.6832504 Iron loss standard deviation TS_STD Magnetic induction standard deviation CG_STD 0.034549 0.002839

[0247] The new iron loss prediction value of 3.003730 is greater than the internal control requirement of 3.00. Therefore, the SACL temperature optimization failed and the heating or soaking temperature of SACL could not be optimized. SACL annealing was carried out according to the original process.

[0248] Those skilled in the art should recognize that the above embodiments are merely illustrative of the present invention and are not intended to limit the present invention. Any variations or modifications to the above embodiments that are within the spirit and essence of the present invention will fall within the scope of the claims of the present invention.

Claims

1. An energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model, characterized in that: Based on the established magnetic property prediction model for non-oriented silicon steel, before the issuance of the SACL continuous annealing plan, the following data are extracted: steelmaking composition, actual steelmaking process values, actual hot rolling process values, actual normalizing process values, SACL process parameter setpoints, SACL process parameter thresholds, performance contract requirements, internal control requirements, and specification parameters. These data are then substituted into the magnetic property prediction model to predict the magnetic property indicators of the non-oriented silicon steel. This allows for the adjustment of the annealing temperature in the SACL annealing furnace for the non-oriented silicon steel, thereby achieving energy conservation and emission reduction in the SACL annealing furnace.

2. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 1, characterized in that: The steelmaking components include Si, Al, Mn, P, Ti, S, C, N, Nb, and V; The actual values ​​of the steelmaking process include ladle slag thickness and final free oxygen after decarburization. The actual values ​​of the hot rolling process include furnace inlet temperature, furnace time, furnace outlet temperature, roughing temperature, finishing temperature, and coiling temperature; The actual values ​​of the normalization process include the furnace temperature of the heating section and the furnace temperature of the soaking section; The SACL process parameters are set as follows: annealing rate, furnace temperature in the annealing soaking zone, furnace temperature in the annealing heating zone, and annealing furnace tension. The SACL process parameter thresholds include the upper and lower limits of the SACL process parameters; The performance contract requirements include the minimum magnetic flux density and the maximum iron loss. The internal control requirements include upper limits, lower limits, and target values ​​for performance indicators; The specifications include the thickness and width of the strip.

3. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 2, characterized in that: The SACL process parameter thresholds are configured individually based on the steel type and grade of the non-oriented silicon steel. The performance contract requirement value is the release requirement set by the user and is determined according to different user needs; The internal control requirements are based on performance control requirements formed through long-term actual production and must be stricter than the performance contract requirements.

4. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 1, characterized in that: The magnetic performance prediction model predicts the magnetic performance indicators of the non-oriented silicon steel, including steel loss and magnetic induction.

5. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 4, characterized in that, The magnetic property prediction model calculation specifically includes: Y = f(steelmaking composition, actual steelmaking process value, actual hot rolling process value, actual normalizing process value, SACL process parameter setting value) Where Y is the magnetic property index to be predicted.

6. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 1, characterized in that, The specific temperature control of the SACL annealing furnace for the non-oriented silicon steel includes: Based on the performance contract requirements and the internal control requirements, and using the magnetic property prediction model to predict the relationship between the magnetic property indicators of the non-oriented silicon steel and the temperature of the SACL annealing furnace, it is determined whether the annealing temperature can be adjusted. If so, optimization is performed; otherwise, production is carried out according to the original process.

7. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 6, characterized in that, Optimizing the annealing temperature specifically includes: Based on the predicted values ​​and boundary values, combined with the magnetic property prediction model and the threshold of the SACL annealing temperature, an optimization objective function and constraints are established, and an optimization algorithm is used to solve the problem to minimize the objective function value, thereby obtaining the optimal value of the annealing temperature of the SACL annealing furnace.

8. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 7, characterized in that, Establishing the optimization objective function specifically includes: Delta_Y1=(P_Y1-Y1_BOUND)^2 Delta_Y2=(P_Y2-Y2_BOUND)^2 Delta_Y3=(P_Y3-Y3_BOUND)^2 ... F_LOSS=Delta_Y1+m1*Delta_Y2+... Wherein, P_Y1 is the predicted value of performance indicator Y1; Y1_BOUND is the boundary value of performance indicator Y1, i.e., the upper or lower limit, the specific upper or lower limit is determined by the indicator and its contract or internal control requirements; Delta_Y1 is used to measure the deviation between the predicted value and the boundary value; the meanings of other performance indicator parameters such as Y2 are the same as those of Y1. F_LOSS is the loss function, which is a weighted sum of the deviation values ​​of each performance index, serving as the optimization objective function; where the parameter m is the weighting coefficient for balancing the deviation values ​​of multiple performance indices, which is determined by the specific steel grade and grade.

9. The energy-saving method for non-oriented silicon steel SACL based on a magnetic property prediction model according to claim 7, characterized in that, Establishing optimization constraints specifically includes: (1) Y1 = f1 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters); (2) Y2 = f2 (steelmaking composition, actual value of steelmaking process, actual value of hot rolling process, actual value of normalizing process, and set value of SACL process parameters); (3) Y3 = f3 (steelmaking composition, actual steelmaking process value, actual hot rolling process value, actual normalizing process value, SACL process parameter setting value) (4) HS_MIN <= HS <= HS_MAX; (5) SS_MIN <= SS <= SS_MAX; (6) Y1_MIN <= Y1 <= Y1_MAX; (7) Y2_MIN <= Y2 <= Y2_MAX; (8) Y3_MIN <= Y3 <= Y3_MAX, Among them, equations (1), (2), and (3) are magnetic property prediction models for Y1, Y2, and Y3, which are three performance indicators; Equations (4) and (5) are the temperature constraints of the SACL annealing furnace, where HS is the heating section temperature and SS is the soaking section temperature; HS_MIN, HS_MAX, SS_MIN, and SS_MAX are the minimum, maximum, minimum, and maximum temperatures of the heating section, respectively, and are all specified by the SACL process parameter thresholds. Equations (6), (7), and (8) define the range of magnetic performance indicators, where Y_MIN is the minimum value of the magnetic performance indicator and Y_MAX is the maximum value of the magnetic performance indicator.

Images