Fluent-based blast furnace tuyere convolute area prediction method

Abstract

The invention discloses a method for predicting a blast furnace tuyere convolute area based on fluent, which comprises the following steps: collecting production data of a target blast furnace, and determining structural parameters of a tuyere of the blast furnace; a Fluent + DEM method is adopted to establish a mathematical model of flow and fuel combustion in a single tuyere convolution area in the simulation of the blast furnace convolution area; the method comprises the following steps: inputting different blast-furnace tuyere parameters, and obtaining corresponding characteristic parameters of a tuyere rotation area through Fluent + DEM; calculating the working condition of a target blast furnace, determining input and output characteristics under different conditions, and taking the input and output characteristics as samples; training the sample by adopting a neural network method to obtain a prediction model of the blast-furnace tuyere convolution area; and extracting real-time tuyere production data of the blast furnace by utilizing the trained prediction model of the tuyere convolute area of the blast furnace, and predicting the characteristic parameters of the current tuyere convolute area of the blast furnace in real time. The prediction method can efficiently complete calculation of the shape of the blast-furnace tuyere convolute area, and is high in accuracy.

Description

Technical Field

[0001] The invention belongs to the technical field of blast furnace ironmaking, and in particular relates to a prediction method for a blast furnace tuyere raceway zone based on FLUENT. Background Art

[0002] Blast furnace smelting is a continuous production process of reducing iron ore into pig iron. Solid raw materials such as iron ore, coke and solvent are fed into the blast furnace in batches from the top of the blast furnace. Hot air is fed into the furnace through the blast furnace tuyere, which is a channel for reacting with coal powder and coke. In front of each tuyere, an evacuated and pear-shaped space is formed, which is usually called the tuyere whirlpool. As an important reaction area for the stable operation of the blast furnace, the tuyere whirlpool provides heat and energy for the entire blast furnace production. The shape of the tuyere whirlpool and the complex physical and chemical reactions inside determine the initial distribution of the blast furnace gas flow and the descending state of the upper charge. It also reflects the combustion state of coke, which is the basis for the smooth operation of the furnace and plays a vital role in the smelting process.

[0003] There are usually two methods to determine the characteristic parameters of the blast furnace tuyere raceway: one is the direct detection method, which uses some special detectors, detection needles and other methods to perform direct measurements, but the instrument equipment is easily affected by the environment, resulting in large fluctuations in the measurement results, and the instrument cost is high, and it is impossible to monitor the characteristic parameters of the tuyere raceway in real time; the other is the indirect method, that is, to study the tuyere raceway through a mathematical model, and to establish an Euler mathematical model to solve it according to the transmission of momentum, mass and heat during the movement of the raceway, but the existing Euler model modeling process is complicated, requires more parameters, is difficult to calculate, takes a long time, and is difficult to achieve the purpose of real-time monitoring. It can be seen that the calculation method of the shape of the blast furnace tuyere raceway in the prior art has the technical problems of complex implementation and low efficiency. Summary of the invention

[0004] The present invention aims to provide a prediction method for the raceway of a blast furnace tuyere based on FLUENT, so as to solve the technical problems of the complexity and low efficiency of the calculation method for the shape of the raceway of a blast furnace tuyere in the prior art.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme:

[0006] A method for predicting the blast furnace tuyere raceway based on FLUENT, comprising the following steps:

[0007] Collect the production data of the target blast furnace and determine the structural parameters of the tuyere of the blast furnace;

[0008] The simulation of the blast furnace raceway zone uses the Fluent+DEM method to establish a mathematical model of flow and fuel combustion in a single tuyere raceway zone. Fluent is used to build a turbulence+discrete phase+component transport model to simulate the combustion of hot air and coal powder and the movement of the generated coal gas. The DEM model is used to simulate the movement of coke in front of the tuyere. Fluent and DEM exchange drag and heat transfer data through the source phase.

[0009] By inputting different blast furnace tuyere parameters, the characteristic parameters of the corresponding tuyere raceway are obtained by Fluent+DEM; the operating conditions of the target blast furnace are calculated, the input and output characteristics under different conditions are determined, and these are used as samples;

[0010] The neural network method is used to train the samples and obtain the prediction model of the blast furnace tuyere raceway area.

[0011] By using the trained prediction model of the blast furnace tuyere raceway, the real-time tuyere production data of the blast furnace is extracted, and the characteristic parameters of the current blast furnace tuyere raceway are predicted in real time.

[0012] Furthermore, the production data of the target blast furnace include but are not limited to the type of fuel, coal powder injection amount, blast furnace oxygen enrichment rate, and air volume; the structural parameters of the blast furnace tuyere include tuyere diameter, length, and position of the coal gun.

[0013] Furthermore, the prediction model adopts a convolutional neural network.

[0014] Furthermore, the flow of gas in the cyclotron of the blast furnace is controlled by the continuity equation and the Navier-Stokes equations: the turbulence model in the computational domain is the k-ε model: the gas components in the fluid are controlled by the component transport equations; and the movement of pulverized coal in the cyclotron is simulated and calculated by the discrete phase model.

[0015] Technical effect:

[0016] Compared with the prior art, the prediction method of the blast furnace tuyere raceway zone of the present invention can efficiently complete the calculation of the shape of the blast furnace tuyere raceway zone with high accuracy. DETAILED DESCRIPTION

[0018] The present invention will now be further described in conjunction with specific implementation methods.

[0019] The present invention provides a method for predicting the raceway of a blast furnace tuyere based on FLUENT, comprising the following steps:

[0020] First, the production data of the target blast furnace is collected, including but not limited to the type of fuel, coal powder injection amount, blast furnace oxygen enrichment rate, air volume and other parameters. At the same time, the structural parameters of the blast furnace tuyere are determined, including tuyere diameter, length, coal gun position, etc.

[0021] The simulation of the blast furnace raceway zone adopts the Fluent+DEM method to establish a mathematical model of the flow and fuel combustion in the raceway of a single tuyere. Among them, Fluent is used to construct a turbulence+discrete phase+component transport model to simulate the combustion of hot air and coal powder and the movement of the generated coal gas. The DEM model is used to simulate the movement of coke in front of the tuyere. The drag and heat transfer data are exchanged between Fluent and DEM by the source phase.

[0022] The flow of gases in the raceway of a blast furnace is governed by the continuity equation and the Navier-Stokes equations:

[0023]

[0024] Where ρ is the gas phase density, u i and u j is the velocity of the gas phase in the x, y and z directions, P is the pressure, g is the gravity, F is the volume force, k and ε are the turbulent kinetic energy and turbulent dissipation rate.

[0025] The turbulence model in the computational domain is the k-ε model:

[0026] μ t =ρC μ k2 / ε (3)

[0027]

[0028]

[0029] Among them, the model constant of the turbulence model is C μ =0.09, C 1ε =1.44, C 2ε =1.92,σ k =1.0 and σ ε =1.3.

[0030] The heat transfer equation in the computational domain is:

[0031]

[0032] Where H is the total enthalpy of the fluid, Г h is the heat transfer coefficient, p is the pressure of the fluid, S h It is a heat source and a radiation heat source for the fluid.

[0033] The gas composition within the fluid is governed by the composition transport equation:

[0034]

[0035] Among them, Y i is the mass ratio of component i, R iis the generation and consumption rate of component i, S i is the source phase, that is, the rate at which component i is obtained or consumed by the reaction.

[0036] Fluent uses a component transport model to calculate the transport and composition changes of the reducing gas.

[0037] The movement of pulverized coal in the raceway is simulated and calculated by the discrete phase model, and its control equation is as follows:

[0038] du p / dt=F D (uu p )+g i (ρ p -ρ) / ρ p +F i (8)

[0039] Among them, u p is the velocity of the particle, u is the velocity of the gas phase in the cyclotron, g i is the gravitational acceleration in the i direction, ρ p is the density of the particle, ρ is the density of the gas phase, F i is the additional force on the particle in the i direction, F D (uu p ) is the drag force on the particle.

[0040] The reaction of coal powder in the cyclotron includes gasification, pyrolysis and combustion of residual carbon. The gasification reaction of coal powder particles in the cyclotron is as follows:

[0041] (Reaction 1) Coal Particles → α1VM1 + (1-α1)Char1 (11)

[0042] (Reaction 2) Coal Particles → α2VM2 + (1-α2)Char2 (12)

[0043] Among them, reaction 1 is a low-temperature reaction, reaction 2 is a high-temperature reaction, Char1 and Char2 are the residual carbon of reactions 1 and 2, α1andα2 are the mass ratios of volatile components gasified by reactions 1 and 2, and α2=1.5α1.

[0044] The generation rate of volatile matter from pulverized coal in the raceway is described by the following equation:

[0045] dVM / dt=(α1k1+α2k2)C0 (13)

[0046] Where VM is the total weight of volatile matter in the coal powder particles, C0 is the initial weight of the coal powder, and k1 and k2 are the rate constants of the gasification reaction.

[0047] Volatile matter of coal powder in the cyclotron zone (Cm H n )'s gas phase combustion reaction is as follows:

[0048] C m H n +(m+n / 4)O2=mCO2+n / 2H2O (14)

[0049] The speed of the above reactions is calculated by finite rate and eddy dissipation model simulation.

[0050] By inputting different blast furnace tuyere parameters such as fuel type, pulverized coal injection rate, blast furnace oxygen enrichment rate, air volume, etc., the characteristic parameters (length, width, height) of the corresponding tuyere raceway are obtained by Fluent+DEM. The working conditions of the target blast furnace are calculated to determine the input and output characteristics under different conditions. Using this as a sample, the neural network method is used to train the sample, and the effect of the trained convolutional neural network is tested with the test set. If the accuracy of the convolutional neural network on the test set is low, continue training until the convolutional neural network performs well on the test set.

[0051] Using the trained neural network, the real-time tuyere production data of the blast furnace is extracted, including but not limited to the type of fuel, coal powder injection amount, blast furnace oxygen enrichment rate, air volume and other parameters, and the characteristic parameters (length, width, height) of the current blast furnace tuyere convolution zone are predicted in real time.

[0052] Compared with the prior art, the prediction method of the blast furnace tuyere raceway zone of the present invention can efficiently complete the calculation of the shape of the blast furnace tuyere raceway zone with high accuracy.

[0053] Although the present invention has been specifically shown and described in conjunction with the preferred embodiments, it should be understood by those skilled in the art that various changes may be made to the present invention in form and details without departing from the spirit and scope of the present invention as defined by the appended claims, all of which are within the scope of protection of the present invention.

Claims

1. A prediction method for blast furnace tuyere raceway based on FLUENT, characterized in that: The following steps are involved: Collect the production data of the target blast furnace and determine the structural parameters of the tuyere of the blast furnace; The simulation of the blast furnace raceway zone uses the Fluent+DEM method to establish a mathematical model of flow and fuel combustion in a single tuyere raceway zone. Fluent is used to build a turbulence+discrete phase+component transport model to simulate the combustion of hot air and coal powder and the movement of the generated coal gas. The DEM model is used to simulate the movement of coke in front of the tuyere. Fluent and DEM exchange drag and heat transfer data through the source phase. By inputting different blast furnace tuyere parameters, the characteristic parameters of the corresponding tuyere raceway are obtained by Fluent+DEM; the operating conditions of the target blast furnace are calculated, the input and output characteristics under different conditions are determined, and these are used as samples; The neural network method is used to train the samples and obtain the prediction model of the blast furnace tuyere raceway area. By using the trained prediction model of the blast furnace tuyere raceway, the real-time tuyere production data of the blast furnace is extracted, and the characteristic parameters of the current blast furnace tuyere raceway are predicted in real time.

2. The method for predicting the blast furnace tuyere raceway based on FLUENT as claimed in claim 1, characterized in that: The production data of the target blast furnace include but are not limited to the type of fuel, the amount of pulverized coal injection, the oxygen enrichment rate of the blast furnace, and the air volume; the structural parameters of the tuyere of the blast furnace include the tuyere diameter, length, and position of the coal gun.

3. The method for predicting the blast furnace tuyere raceway based on FLUENT as claimed in claim 1, characterized in that: The prediction model adopts a convolutional neural network.

4. The method for predicting the blast furnace tuyere raceway based on FLUENT as claimed in claim 1, characterized in that: The flow of gas in the blast furnace raceway is controlled by the continuity equation and the Navier-Stokes equations. The turbulence model in the computational domain is the k-ε model. The gas components in the fluid are controlled by the component transport equations. The movement of pulverized coal in the raceway is simulated and calculated by the discrete phase model.

Images