Method for predicting optimal blast temperature of blast furnace based on multi-objective genetic algorithm

Abstract

The invention relates to a method for predicting the optimal blast temperature of a blast furnace based on a multi-objective genetic algorithm, which comprises the following steps of: taking factors such as hot air pressure, pressure difference, blast temperature, coke amount, sulfur content of molten iron of a previous furnace, material speed, blast volume, top pressure, silicon content of molten iron of the previous furnace and the like as input variables; training the RBF neural network to obtain a molten iron physical temperature prediction model, a molten iron silicon content prediction model, a fuel ratio prediction model and a coal ratio prediction model; constructing a multi-objective optimization model by taking the air temperature as a decision variable and taking the molten iron physical temperature, the molten iron silicon content, the fuel ratio and the coal ratio as constraint conditions; and solving the multi-objective optimization model by using a genetic algorithm to obtain the optimal wind temperature. Based on the multi-target genetic algorithm, different factors can be taken into consideration, so that prediction of the optimal wind temperature of the target blast furnace is realized, and energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of blast furnaces, and particularly to a method for predicting the optimal blast temperature of a blast furnace based on a multi-objective genetic algorithm. Background Art

[0002] A blast furnace is an important device for ironmaking in the metallurgical industry, and its performance and efficiency directly affect the economy and environmental friendliness of steel production. During the ironmaking process in a blast furnace, the control of blast temperature is crucial because an appropriate blast temperature can fully improve the combustion efficiency of fuel, promote the reduction of iron ore, increase the output and quality of hot metal, and at the same time reduce energy consumption. However, too high or too low blast temperature will cause a series of problems such as frosting, incomplete in-furnace reactions, and reduced finished product quality, posing challenges to the stable operation and production efficiency of the blast furnace.

[0003] In traditional blast furnace blast temperature prediction methods, researchers usually rely on empirical formulas or single-variable models. These methods are difficult to consider the mutual influence of various complex factors during the operation of the blast furnace, such as raw material properties, in-furnace temperature distribution, gas flow characteristics, etc., and are difficult to provide real-time optimal blast temperature prediction. Summary of the Invention

[0004] To solve the above problems, an object of an embodiment of the present invention is to provide a method for predicting the optimal blast temperature of a blast furnace based on a multi-objective genetic algorithm.

[0005] A method for predicting the optimal blast temperature of a blast furnace based on a multi-objective genetic algorithm includes:

[0006] Taking hot blast pressure, differential pressure, blast temperature, coke amount, and sulfur content of the previous furnace of hot metal as input variables, training an RBF neural network to obtain a prediction model for the physical temperature of hot metal;

[0007] Taking air volume, differential pressure, blast temperature, coke amount, and manganese content of the previous furnace of hot metal as input variables, training an RBF neural network to obtain a prediction model for the silicon content of hot metal;

[0008] Taking charging rate, air volume, top pressure, blast temperature, coke amount, and silicon content of the previous furnace of hot metal as input variables, training an RBF neural network to obtain a prediction model for fuel ratio;

[0009] Taking charging rate, air volume, top pressure, blast temperature, coke amount, and silicon content of the previous furnace of hot metal as input variables, training an RBF neural network to obtain a prediction model for coal ratio;

[0010] Using blast temperature as a decision variable, and taking the physical temperature of hot metal, silicon content of hot metal, fuel ratio, and coal ratio as constraint conditions to construct a multi-objective optimization model;

[0011] Using a genetic algorithm to solve the multi-objective optimization model to obtain the optimal blast temperature.

[0012] Preferably, the physical temperature prediction model of hot metal is:

[0013]

[0014] where m 3 represents the hot blast pressure, m 5 represents the differential pressure, m 7 represents the hot blast temperature, n 1 represents the amount of coke, m 11 represents the sulfur content of the previous heat of hot metal, b 1 represents the bias of the physical temperature prediction model of hot metal, wi 1 represents the weight of the i-th hidden neuron of the physical temperature prediction model of hot metal, φ i (X p1 ) represents the output of the j-th neuron in the hidden layer of the physical temperature prediction model of hot metal.

[0015] Preferably, the silicon content prediction model of hot metal is:

[0016]

[0017] where m 3 represents the hot blast pressure, m 5 represents the differential pressure, m 7 represents the hot blast temperature, n 3 represents the manganese content of the previous heat of hot metal, m 11 represents the sulfur content of the previous heat of hot metal, b 2 represents the bias of the silicon content prediction model of hot metal, w i2 represents the weight of the i-th hidden neuron of the silicon content prediction model of hot metal, φ i (X p2 ) represents the output of the j-th neuron in the hidden layer of the silicon content prediction model of hot metal.

[0018] Preferably, the fuel ratio prediction model is:

[0019]

[0020] where m 1 represents the feeding rate, m 2 represents the air volume, m 4 represents the top pressure, m 7 represents the hot blast temperature, m 11 represents the amount of coke, n 2 represents the silicon content of the previous heat of hot metal, b 3 represents the bias of the fuel ratio prediction model, wi 3 represents the weight of the i-th hidden neuron of the fuel ratio prediction model, φ i (X p3) represents the output of the j-th neuron in the hidden layer of the fuel ratio prediction model.

[0021] Preferably, the coal ratio prediction model is:

[0022]

[0023] where m 1 represents the material feeding speed, m 2 represents the air volume, m 4 represents the top pressure, m 7 represents the air temperature, m 11 represents the amount of coke, n 2 represents the silicon content of the molten iron in the previous furnace, b 4 represents the bias of the coal ratio prediction model, wi 4 represents the weight of the i-th hidden neuron of the coal ratio prediction model, φ i (X p4 ) represents the output of the j-th neuron in the hidden layer of the coal ratio prediction model.

[0024] Preferably, the multi-objective optimization model is:

[0025]

[0026] where m 7 represents the air temperature.

[0027] According to the specific embodiments provided by the present invention, the following technical effects are disclosed:

[0028] The present invention relates to a method for predicting the optimal air temperature of a blast furnace based on a multi-objective genetic algorithm. Compared with the prior art, the present invention is based on a multi-objective genetic algorithm, which can take different factors into consideration, thereby realizing the prediction of the optimal air temperature of the target blast furnace and reducing energy consumption.

[0029] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following preferred embodiments are specifically exemplified below and described in detail in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0031] Figure 1 It is a schematic diagram of the prediction effect of the temperature prediction model provided by the present invention;

[0032] Figure 2 Schematic diagram of the prediction effect of the molten iron silicon content provided by the present invention;

[0033] Figure 3 Schematic diagram of the prediction effect of the fuel ratio prediction model provided by the present invention;

[0034] Figure 4 Schematic diagram of the prediction effect of the coal ratio prediction model provided by the present invention;

[0035] Figure 5 Schematic diagram of the Pareto front of the multi-objective optimization model provided by the present invention;

[0036] Figure 6 Industrial test result diagram provided by the present invention;

[0037] Figure 7 Schematic diagram of the influence of different blast temperatures on the K value provided by the present invention. Detailed implementation manners

[0038] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.

[0039] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality" means two or more unless otherwise specifically defined.

[0040] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected with", "fixed" and other terms should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0041] Please refer to Figure 1-7, a method for predicting the optimal blast temperature of a blast furnace based on a multi-objective genetic algorithm, including:

[0042] Step 1: Using hot blast pressure, differential pressure, blast temperature, coke amount, and sulfur content of the previous furnace of hot metal as input variables, train an RBF neural network to obtain a prediction model for the physical temperature of hot metal; the prediction model for the physical temperature of hot metal is:

[0043]

[0044] where, m 3 represents the hot blast pressure, m 5 represents the differential pressure, m 7 represents the blast temperature, n 1 represents the coke amount, m 11 represents the sulfur content of the previous furnace of hot metal, b 1 represents the bias of the prediction model for the physical temperature of hot metal, wi 1 represents the weight of the i-th hidden neuron of the prediction model for the physical temperature of hot metal, φ i (X p1 ) represents the output of the j-th neuron in the hidden layer of the prediction model for the physical temperature of hot metal.

[0045] Step 2: Using blast volume, differential pressure, blast temperature, coke amount, and manganese content of the previous furnace of hot metal as input variables, train an RBF neural network to obtain a prediction model for the silicon content of hot metal; the prediction model for the silicon content of hot metal is:

[0046]

[0047] where, m 3 represents the hot blast pressure, m 5 represents the differential pressure, m 7 represents the blast temperature, n 3 represents the manganese content of the previous furnace of hot metal, m 11 represents the sulfur content of the previous furnace of hot metal, b 2 represents the bias of the prediction model for the silicon content of hot metal, w i2 represents the weight of the i-th hidden neuron of the prediction model for the silicon content of hot metal, φ i (X p2 ) represents the output of the j-th neuron in the hidden layer of the prediction model for the silicon content of hot metal.

[0048] Step 3: Using charging rate, blast volume, top pressure, blast temperature, coke amount, and silicon content of the previous furnace of hot metal as input variables, train an RBF neural network to obtain a prediction model for fuel ratio; the prediction model for fuel ratio is:

[0049]

[0050] where, m 1 represents the charging rate, m2 Denotes the air volume, m 4 Denotes the top pressure, m 7 Denotes the blast temperature, m 11 Denotes the coke amount, n 2 Denotes the silicon content of the previous heat of hot metal, b 3 Denotes the bias of the fuel ratio prediction model, wi 3 Denotes the weight of the i-th hidden neuron of the fuel ratio prediction model, φ i (X p3 ) Denotes the output of the j-th neuron in the hidden layer of the fuel ratio prediction model.

[0051] Step 4: Use the material feeding speed, air volume, top pressure, blast temperature, coke amount, and silicon content of the previous heat of hot metal as input variables to train an RBF neural network to establish a coal ratio prediction model; the coal ratio prediction model is:

[0052]

[0053] where, m 1 Denotes the material feeding speed, m 2 Denotes the air volume, m 4 Denotes the top pressure, m 7 Denotes the blast temperature, m 11 Denotes the coke amount, n 2 Denotes the silicon content of the previous heat of hot metal, b 4 Denotes the bias of the coal ratio prediction model, wi 4 Denotes the weight of the i-th hidden neuron of the coal ratio prediction model, φ i (X p4 ) Denotes the output of the j-th neuron in the hidden layer of the coal ratio prediction model;

[0054] Step 5: Use the blast temperature as the decision variable, and use the physical temperature of hot metal, silicon content of hot metal, fuel ratio, and coal ratio as constraint conditions to construct a multi-objective optimization model; the multi-objective optimization model is:

[0055]

[0056] where, m 7 Denotes the blast temperature.

[0057] Step 6: Use the genetic algorithm to solve the multi-objective optimization model to obtain the optimal blast temperature.

[0058] Next, the present invention will be further described in conjunction with specific embodiments of a method for predicting the optimal blast temperature of a blast furnace based on a multi-objective genetic algorithm:

[0059] ① Constraint condition modeling

[0060] In blast furnace production, the hearth temperature is often used to measure whether the control of fuel ratio and coal ratio is appropriate. When the fuel ratio and coal ratio increase, the heat generated by fuel combustion increases, and the hearth temperature rises; while when the fuel ratio and coal ratio decrease, the heat generated by fuel combustion decreases, and the hearth temperature drops. There are two methods to measure the hearth temperature, namely the physical temperature of hot metal and the silicon content of hot metal.

[0061] An RBF neural network is used to establish the correlation model between the physical temperature of hot metal and the silicon content of hot metal. Through model testing, it is determined that the hot blast pressure (m 3 ), differential pressure (m 5 ), hot blast temperature (m 7 ), coke quantity (m 11 ), and sulfur content of the previous heat of hot metal (n 5 ) are used as 5 input variables to establish the prediction model of the physical temperature of hot metal; it is determined that the blast volume (m 2 ), differential pressure (m 5 ), hot blast temperature (m 7 ), coke quantity (m 11 ), and manganese content of the previous heat of hot metal (n 3 ) are used as 5 input variables to establish the prediction model of silicon content. By introducing production experience, the correlation model between the physical temperature of hot metal and silicon content based on the RBF network is established, as shown in Equation 1.

[0062]

[0063] The prediction results of the prediction models for the physical temperature of hot metal and silicon content are as Figure 1-2 shown, and the regression fitting degree of the model is relatively high. The relative error between the predicted output and the actual output of the prediction model for the physical temperature of hot metal is 0.13%, and the relative error between the predicted output and the actual output of the prediction model for silicon content is 3.31%. The relative errors are both less than 5.0%, indicating that the test effect of the prediction model is good.

[0064] ② Modeling of target conditions

[0065] Combined with production practice, an RBF neural network is used to establish the correlation model between fuel ratio and coal ratio, as shown in Equation 2. After debugging the model, according to the consistency between the predicted value and the actual value, the charging speed (m 1 ), blast volume (m 2 ), top pressure (m 4 ), hot blast temperature (m 7 ), coke quantity (m 11 ), and silicon content of the previous heat of hot metal (n 2 ) are used as 6 input variables to establish the prediction model of fuel ratio. It is determined that the charging speed (m 1 ), blast volume (m 2 ), top pressure (m 4 ), hot blast temperature (m 7)、Coke quantity (m 11 )、Silicon content of the previous heat of hot metal (n 2 ), a total of 6 variables are used as input variables to establish a coal ratio prediction model.

[0066]

[0067] The model prediction results of fuel ratio and coal ratio are as Figure 3-4 shown, and the model regression fitting degree is relatively high. The relative error between the predicted output and the actual output of the fuel ratio prediction model is 1.57%, while the relative error between the predicted output and the actual output of the coal ratio prediction model is 0.71%. The relative errors are all less than 5.0%, indicating that the test performance of the prediction model is good.

[0068] The performance of the prediction model is verified by the absolute error, relative error and hit rate of the prediction model. As shown in Table 1, the relative error and hit rate of the model prediction are within reasonable control ranges. The hit rates of the coal ratio and hot metal physical temperature models are 90.11% and 97.80% respectively, and the relative errors are 1.57% and 0.13% respectively. The prediction effect of the model is relatively good. The hit rates of the fuel ratio and hot metal silicon content models are 81.31% and 86.81% respectively, and the relative errors are 0.71% and 3.31% respectively. The prediction effect of the model is good. According to production experience, the absolute error of the hot metal physical temperature model prediction is 2.58 °C, the absolute error of the hot metal silicon content model prediction is 0.05%, and the absolute error of the coal ratio model prediction is 2.71 kg·tHM -1 , and the absolute error of the fuel ratio model prediction is 5.01 kg·tHM -1 . The above absolute errors are relatively small, and the prediction performance of the constraint conditions and target conditions model is good.

[0069] Table 1 Model performance indicators

[0070]

[0071] ③ Multi-objective optimization results and analysis of NSGA-II algorithm

[0072] NSGA-II multi-objective optimization model: During the optimization process, blast temperature is used as the decision variable, hot metal physical temperature and silicon content are used as constraint conditions. When the reasonable constraint range of hot metal physical temperature is controlled at 1500 ± 10 °C and the reasonable constraint range of hot metal silicon content is controlled at 0.45% - 0.55%, it can ensure the smooth operation of the furnace condition while maintaining a low fuel consumption level. In addition, according to actual production experience, the order of the variables of the model is adjusted. The reasonable constraint range of blast temperature in the decision variables is 1180 °C - 1210 °C. The multi-objective optimization model is shown in Equation 3.

[0073]

[0074] Given that there are many decision variables and objective functions, different tests were conducted on the population size ranging from 100 to 1000. The population size corresponding to the model was set to 90, the maximum number of iterations was 200, the crossover probability was set to 0.85, and the mutation probability was 0.15. To solve the Pareto solution set of the decision variables, Figure 5 The Pareto front of the multi-objective optimization model shown indicates that the fuel at point A is relatively low and the fuel at point B is relatively high. According to production experience, the fuel ratio values of the 3 points on the Pareto front closest to point A were selected as the optimal solution set, as shown in Table 2. Among the 3 optimal solution sets, under the constraint of ensuring good hearth temperature, the obtained fuel ratio index was significantly improved. Among them, for the 3rd solution set, with approximately the same coal ratio, the fuel ratio decreased the most, that is, when the blast temperature was 1210.00 °C, the optimal fuel ratio was 534.28 kg·tHM -1 , and the hearth temperature index was at an appropriate level. The model predicted that the fuel ratio was reduced by 5.72 kg·tHM compared to the current fuel ratio -1 , and the predicted coal ratio was reduced by 0.81 kg·tHM compared to the current coal ratio -1 .

[0075] Table 2 Multi-objective optimization results

[0076] Variable Current value Multi-objective optimization value 1 Multi-objective optimization value 2 Multi-objective optimization value 3 Hot blast temperature / °C 1195.00 1199.00 1209.50 1210.00 Physical temperature of hot metal / °C 1499.75 1500.60 1501.00 1500.80 Silicon content of hot metal / % 0.50 0.51 0.51 0.52 <![CDATA[Fuel ratio / kg·tHM -1 > 540.00 538.46 534.40 534.28 <![CDATA[Coal ratio / kg·tHM -1 > 132.00 131.76 131.21 131.19 <![CDATA[Coke ratio / kg·tHM -1 > 408.00 406.70 403.19 403.09

[0077] To verify the prediction effect of the model, 146 groups of industrial tests were carried out on 8 # blast furnaces, and the changes in fuel ratio and coal ratio under different blast temperature conditions were tested, as Figure 6 shown

[0078] Through Figure 6 the data in it, it is difficult to obtain the relationship between fuel ratio, coal ratio and blast temperature change. It is necessary to convert the fuel ratio and coal ratio into the index K first before comparison. In addition, Figure 6 the hearth temperatures and gas utilization rates corresponding to the fuel ratio and coal ratio in it are different, and it is necessary to convert the fuel ratio and coal ratio data in the figure to the same hearth temperature and gas utilization rate conditions

[0079] Let the ratio of fuel ratio to coal ratio be K, then the calculation method of K is:

[0080] K = Fr / PCR Equation 5

[0081] In the formula: Fr - fuel ratio; PCR - coal ratio

[0082] It can be seen from Equation 5 that the smaller the K value, the lower the fuel ratio of the blast furnace, the higher the coal ratio, the more optimized the fuel structure, and the better the effect of reducing CO 2 emissions

[0083] The trend line fitted by the least squares method is asFigure 7 as shown

[0084] From Figure 7 Among the 146 groups of data, 37 groups of data with the smallest K value are selected (about 25% of the total amount of data), and the arithmetic mean of the blast temperature and the K value of these 37 groups of data is used as the optimal solution in the industrial data. After calculation, when the arithmetic mean of the blast temperature is 1200.31 °C, the arithmetic mean of the K value is 3.98, and the average fuel ratio at this time is 534.41 kg·t -1 , and the average coal ratio is 134.24 kg·t -1 . The absolute error between the optimal blast temperature setting value predicted by the model and the industrial test result is 9.69 °C. The deviation between the blast temperature setting value obtained by the industrial test and the blast temperature setting value predicted by the model is small, and the model prediction performance is good. The fuel ratio obtained from the industrial test is 534.41 kg·t -1 is lower than 540.00 kg·t -1 by 5.59 kg·t -1 , and the actual cost reduction is 58.3451 million yuan. The calculation process is shown in Equation 6:

[0085] 2.113×4150×365×(3571.25×4.78 - 1430.22×0.81)

[0086] = 58.3451 million yuan Equation 6

[0087] In Equation 6: Calculated based on the coke price of 3571.25 yuan / ton and the pulverized coal injection of 1430.22 yuan / ton, the effective volume of the blast furnace is 4150 m 3 , and the average utilization coefficient is 2.113 t·m -3 ·d -1 (considering the influence of the blow-down effect), the average carbon content of the fuel is 81.33%, the molar mass of carbon atom is 12 g·mol -1 , and the molar mass of oxygen atom is 16 g·mol -1 .

[0088] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of the technical solutions of changes or substitutions, which should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims described above.

Claims

1. A method for predicting the optimal air temperature of a blast furnace based on a multi-objective genetic algorithm, characterized in that: include: The hot air pressure, pressure difference, air temperature, coke quantity and sulfur content of the previous furnace of molten iron are used as input variables to train the RBF neural network to obtain the molten iron physical temperature prediction model; The air volume, pressure difference, air temperature, coke amount, and manganese content of the previous furnace molten iron are used as input variables to train the RBF neural network to obtain a prediction model for silicon content in molten iron. The material speed, air volume, top pressure, air temperature, coke quantity, and silicon content of the previous furnace of molten iron are used as input variables to train the RBF neural network to obtain the fuel ratio prediction model; The material speed, air volume, top pressure, air temperature, coke quantity, and silicon content of the previous furnace molten iron are used as input variables to train the RBF neural network to establish the coal ratio prediction model; Using wind temperature as the decision variable, the physical temperature of molten iron, silicon content of molten iron, fuel ratio and coal ratio as constraints to build a multi-objective optimization model; The multi-objective optimization model is solved using a genetic algorithm to obtain the optimal wind temperature.

2. The method for predicting the optimal air temperature of a blast furnace based on a multi-objective genetic algorithm according to claim 1, characterized in that: The molten iron physical temperature prediction model is: Among them, m3 represents the hot air pressure, m5 represents the pressure difference, m7 represents the air temperature, n1 represents the coke amount, m 11 represents the sulfur content of the previous molten iron, b1 represents the bias of the molten iron physical temperature prediction model, wi1 represents the weight of the ith hidden neuron in the molten iron physical temperature prediction model, φ i (X p1 ) represents the output of the jth neuron in the hidden layer of the molten iron physical temperature prediction model.

3. The method for predicting the optimal air temperature of a blast furnace based on a multi-objective genetic algorithm according to claim 1, characterized in that: The prediction model for silicon content in molten iron is: Among them, m3 represents the hot air pressure, m5 represents the pressure difference, m7 represents the air temperature, n3 represents the manganese content of the previous furnace molten iron, and m 11 represents the sulfur content of the previous hot metal, b2 represents the bias of the hot metal silicon content prediction model, and w i2 represents the weight of the ith hidden neuron in the prediction model of molten iron silicon content, φ i (X p2 ) represents the output of the jth neuron in the hidden layer of the molten iron silicon content prediction model.

4. The method for predicting the optimal blast furnace temperature based on a multi-objective genetic algorithm according to claim 1, characterized in that: The fuel ratio prediction model is: Among them, m1 represents material speed, m2 represents air volume, m4 represents top pressure, m7 represents air temperature, m 11 represents the amount of coke, n2 represents the silicon content of the previous hot metal, b3 represents the bias of the fuel ratio prediction model, wi3 represents the weight of the i-th hidden neuron in the fuel ratio prediction model, φ i (X p3 ) represents the output of the jth neuron in the hidden layer of the fuel ratio prediction model.

5. The method for predicting the optimal blast furnace temperature based on a multi-objective genetic algorithm according to claim 1, characterized in that: The coal ratio prediction model is: Among them, m1 represents material speed, m2 represents air volume, m4 represents top pressure, m7 represents air temperature, m 11 represents the amount of coke, n2 represents the silicon content of the previous furnace of molten iron, b4 represents the bias of the coal ratio prediction model, wi4 represents the weight of the i-th hidden neuron in the coal ratio prediction model, φ i (X p4 ) represents the output of the jth neuron in the hidden layer of the coal ratio prediction model.

6. A method for predicting the optimal blast furnace temperature based on a multi-objective genetic algorithm according to any one of claims 2 to 5, characterized in that: The multi-objective optimization model is: Among them, m7 represents wind temperature.

Images