Method for controlling distribution of concentrate in flash furnace based on intelligent prediction

Abstract

The invention provides an intelligent prediction-based concentrate distribution control method in a flash furnace, which comprises the following steps of: acquiring the temperature and smoke concentration data of a reaction area in real time through a temperature sensor and a smoke online detector, and synchronously acquiring the supply flow of mineral aggregate, oxygen-enriched gas and air by a distributed control system. And the intelligent prediction module performs three-dimensional grid division and mathematical model iterative calculation based on the data, and outputs multi-physical quantity prediction results of a velocity field, a temperature field, a turbulence field and a particle distribution field. And the distributed control system comprehensively adjusts the flow parameters of the feeding controller and the gas controller according to the predicted data. And the display module displays prediction data in real time in a color grading form, and the data storage module records the corresponding relation between historical prediction and operation parameters, so that dynamic optimization control of the smelting process is realized. The control precision and the energy utilization efficiency of the flash smelting process can be improved, and dust emission and equipment loss are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of nonferrous metal smelting, and more specifically, to a method for controlling concentrate distribution in a flash furnace based on intelligent prediction. Background Art

[0002] Flash smelting is a widely used, efficient smelting technology in modern industrial production, primarily used to process various ores to extract metals. The operation of existing flash smelting furnaces relies heavily on operator experience and limited real-time monitoring data. Typically, operators manually adjust key parameters such as the ore feed rate, oxygen-enriched gas flow rate, and air flow rate by monitoring basic parameters such as furnace temperature and smoke concentration. However, this experience-based control approach has numerous limitations. First, the complex physical and chemical reactions within the furnace are difficult to fully understand using simple monitoring data, resulting in insufficient control accuracy. Second, the furnace interacts with multiple physical variables, such as velocity, temperature, turbulence, and particle distribution. The distribution of these fields has a crucial impact on smelting efficiency and product quality, but existing technologies struggle to accurately predict and comprehensively control these fields in real time. Furthermore, existing monitoring and control methods lack a visual display of furnace conditions, making it difficult for operators to quickly and comprehensively grasp dynamic changes within the furnace, hindering the timely detection and resolution of abnormalities.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the control method of the existing flash smelting furnace cannot fully consider the complex multi-physical quantity interaction in the furnace, and it is difficult to achieve accurate prediction and effective regulation of the furnace state, resulting in low smelting efficiency and unstable product quality. It is also impossible to intuitively display the dynamic changes in the furnace, which is not conducive to operators to promptly discover and deal with abnormal situations. Summary of the Invention

[0004] The present invention provides a method for controlling concentrate distribution in a flash furnace based on intelligent prediction, comprising:

[0005] Step S1: collecting temperature data of the flash smelting furnace reaction zone in real time through a temperature sensor, and collecting smoke concentration data of the ascending flue in real time through a smoke online detector;

[0006] Step S2: The distributed control system receives and displays the real-time data of step S1, and simultaneously obtains the ore supply flow rate of the ore silo, the oxygen-enriched gas supply flow rate of the oxygen-enriched gas pump, and the air supply flow rate of the air pump;

[0007] Step S3: The intelligent prediction module receives the gas flow rate and ore supply flow rate data from step S2, performs three-dimensional space grid division and mathematical model iterative calculation, and outputs multi-physical quantity prediction data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field inside the flash smelting furnace;

[0008] Step S4: Based on the multi-physical quantity prediction data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field obtained in step S3, the flow parameters of the feed controller of the ore silo, the gas controller of the oxygen enrichment pump, and the gas controller of the air pump are comprehensively adjusted through the distributed control system;

[0009] Step S5: The display module displays the prediction data of step S3 in real time in a color-graded form, and the data storage module records the corresponding relationship between the historical prediction data and the operating parameters.

[0010] As a further improvement of the present application, the velocity field distribution in step S3 is obtained by solving the three-dimensional Navier-Stokes equation, which is expressed as:

[0011] For the X direction:

[0012]

[0013] For the Y direction:

[0014]

[0015] For the Z direction:

[0016]

[0017] Where ρ represents the gas density, u represents the velocity component in the x direction, v represents the velocity component in the y direction, w represents the velocity component in the z direction, p represents the pressure, τ represents the viscous stress component, and g x Represents the gravitational acceleration component in the x direction, g y Represents the gravitational acceleration component in the y direction, g z Represents the gravitational acceleration component in the z direction.

[0018] As a further improvement of the present application, the turbulence field distribution in step S3 is obtained by solving a large eddy simulation model, and the model equation is:

[0019]

[0020] Where ρ represents the filtered gas density and is expressed in kg / m 3 ,u i represents the velocity component after filtering and the unit is m / s, P represents the pressure after filtering and the unit is Pa, ρu′ i u′j Represents the subgrid Reynolds stress component and is in Pa, g i Represents the acceleration due to gravity and its unit is m / s 2 .

[0021] As a further improvement of the present application, the particle distribution field in step S3 is obtained by solving the multiphase particle unit method, and the equation includes the particle momentum equation:

[0022]

[0023] Among them, u s represents the particle velocity vector, ρ s represents the particle density, θ s represents the particle volume fraction, τ s represents the particle stress tensor, D s represents the gas-solid drag coefficient, u g represents the gas velocity vector, and g represents the gravitational acceleration vector.

[0024] As a further improvement of the present application, the temperature field distribution of step S3 is obtained by solving the gas-solid two-phase heat conservation equation, which is expressed as:

[0025]

[0026] Among them, θ g represents the gas volume fraction, h g represents the gas enthalpy value, q represents the heat flux vector, Q D represents the enthalpy diffusion rate, S g,s Indicates the gas-solid heat transfer, ΔH rg Indicates the heat released by the oxidation reaction.

[0027] As a further improvement of the present application, step S4 includes:

[0028] Step S4.1: Regulating the gas controller of the air pump based on the velocity field distribution data, specifically generating an air supply flow rate adjustment instruction based on the deviation of the gas flow rate in the central area of ​​the reaction tower from a preset velocity threshold;

[0029] Step S4.2: adjusting the gas controller of the oxygen-enriched gas pump based on the temperature field distribution data, specifically generating an oxygen-enriched gas supply flow adjustment instruction based on the deviation between the wall temperature and a preset temperature safety threshold;

[0030] Step S4.3: Jointly adjusting the gas controllers of the air pump and the oxygen-enriched gas pump based on the turbulence field distribution data, specifically, synchronously generating flow ratio adjustment instructions for air and oxygen-enriched gas based on the deviation of the turbulence intensity coefficient from a preset optimal range;

[0031] Step S4.4: Adjust the feed controller of the ore bin based on the particle distribution field data, specifically, generate an ore supply flow adjustment instruction according to the difference between the particle residence time and the process requirement.

[0032] As a further improvement of the present application, the color gradation display in step S5 includes:

[0033] Step S5.1: Velocity field distribution data is displayed using the dual dimensions of streamline arrow length and color depth, with red representing high-speed areas and blue representing low-speed areas;

[0034] Step S5.2: The temperature field distribution data is rendered using isothermal surface technology, with red representing high-temperature areas and blue representing low-temperature areas, and the color scale is divided according to the preset temperature gradient;

[0035] Step S5.3: The particle distribution field data is displayed using a density cloud map. The transparency is inversely proportional to the particle concentration, and the high-concentration area is displayed as a dark dense dot matrix.

[0036] As a further improvement of the present application, the three-dimensional space grid division in step S3 includes:

[0037] Step S3.1: Generate a hexahedral structured mesh that matches the geometry of the flash smelting furnace;

[0038] Step S3.2: Perform first-level mesh refinement in the gas injection port and mineral injection port areas;

[0039] Step S3.3: Perform the second level of mesh refinement in the molten pool reaction zone and the flue transition zone;

[0040] Step S3.4: Dynamically optimize the mesh size according to the fluid dynamic characteristics.

[0041] As a further improvement of the present application, the data storage module operation in step S5 includes:

[0042] Step S5.4: Establish a multi-physical quantity prediction database to store time series data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field;

[0043] Step S5.5: Construct an operation parameter-prediction data mapping relationship table. When it is detected that the matching degree between the current operation parameter and the historical record exceeds a preset threshold, directly call the historical prediction data and output it to the display module.

[0044] As a further improvement of the present application, the iterative calculation in step S3 adopts a heterogeneous computing architecture, including:

[0045] Step S3.5: The central processing unit performs velocity field distribution and temperature field distribution calculations, and solves the Navier-Stokes equations and the heat conservation equation;

[0046] Step S3.6: The graphics processing unit performs turbulence field distribution and particle distribution field calculations, completing large eddy simulation and parallel solution of multiphase particle unit equations;

[0047] Step S3.7: The central processing unit and the graphics processing unit exchange boundary condition data via a high-speed bus to achieve dynamic coupling of the calculation process.

[0048] The above embodiments of the present invention have at least the following beneficial effects:

[0049] 1. Based on multi-physics field coupling modeling (velocity field, temperature field, turbulence field, particle distribution field), dynamic prediction of the smelting process is achieved. By adjusting the ore supply and gas flow parameters in real time, the problems of strong hysteresis and insufficient adjustment accuracy of traditional manual control are solved, and the reaction stability and process index compliance rate are improved.

[0050] 2. Through heterogeneous computing architecture (CPU processes Navier-Stokes equations / GPU parallel calculations of turbulence and particle fields), complex three-dimensional models can be quickly solved, solving the problem of low efficiency of traditional single-machine computing and inability to meet real-time control requirements, ensuring that the system completes full-process simulation and decision-making in milliseconds.

[0051] 3. Through color-graded visualization and historical data mapping functions, streamline arrows display velocity fields, isothermal surfaces render temperature fields, and density cloud maps display particle distribution. This solves the problem that manual monitoring is difficult to capture the dynamic correlation of multi-dimensional parameters, and provides operators with an intuitive basis for judging process status. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0053] Figure 1 Schematic diagram of the flash smelting furnace provided by the present invention. DETAILED DESCRIPTION

[0054] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0055] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0056] It should be noted that any number of elements in the drawings is for illustration only and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0057] like Figure 1 As shown in Figure 1, the flash smelting furnace consists of a molten pool 1, a distributed control system 2, an intelligent prediction module 3 (including a model iterative calculation module 31, a data storage module 32, and a display module 33), a ore silo 4, an oxygen-enriched gas pump 5, an air pump 6, a feed controller ① (including a sensor), an oxygen-enriched gas controller ② (including a sensor), an air controller ③ (including a sensor), a temperature sensor ④, and an online smoke detector ⑤. The ore silo 4 stores ore, with the feed controller ① regulating its supply flow rate. The oxygen-enriched gas pump 5 and the air pump 6 provide oxygen-enriched gas and air, respectively, with their flow rates regulated by the oxygen-enriched gas controller ② and the air controller ③. The temperature sensor ④ collects real-time temperature data within the reaction tower, and the online smoke detector ⑤ measures the smoke concentration in the ascending flue. The distributed control system 2 receives and integrates this real-time data and flow information, and adjusts the flow parameters of each controller based on the prediction results of the intelligent prediction module 3. The intelligent prediction module 3 uses a model iteration calculation module 31 to perform three-dimensional spatial gridding and mathematical model iteration on gas flow and ore feed flow data, outputting multi-physics prediction data for velocity, temperature, turbulence, and particle distribution. The data storage module 32 records the correspondence between historical prediction data and operating parameters, while the display module 33 displays the prediction data in real time using a color-coded gradient. This enables intelligent prediction and control of the flash smelting furnace, ensuring an efficient and stable smelting process. The coordinated operation of these components makes the entire system more intelligent and precise.

[0058] A method for controlling concentrate distribution in a flash furnace based on intelligent prediction includes:

[0059] Step S1: collecting temperature data of the flash smelting furnace reaction zone in real time through a temperature sensor, and collecting smoke concentration data of the ascending flue in real time through a smoke online detector;

[0060] Step S2: The distributed control system receives and displays the real-time data of step S1, and simultaneously obtains the ore supply flow rate of the ore silo, the oxygen-enriched gas supply flow rate of the oxygen-enriched gas pump, and the air supply flow rate of the air pump;

[0061] Step S3: The intelligent prediction module receives the gas flow rate and ore supply flow rate data from step S2, performs three-dimensional space grid division and mathematical model iterative calculation, and outputs multi-physical quantity prediction data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field inside the flash smelting furnace;

[0062] Step S4: Based on the multi-physical quantity prediction data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field obtained in step S3, the flow parameters of the feed controller of the ore silo, the gas controller of the oxygen enrichment pump, and the gas controller of the air pump are comprehensively adjusted through the distributed control system;

[0063] Step S5: The display module displays the prediction data of step S3 in real time in a color-graded form, and the data storage module records the corresponding relationship between the historical prediction data and the operating parameters.

[0064] It should be noted that the present invention provides a method for controlling the distribution of concentrate in a flash furnace based on intelligent prediction, which aims to achieve accurate prediction and effective control of the complex physical processes inside the flash smelting furnace through intelligent means. In step S1, the temperature data of the reaction zone of the flash smelting furnace is collected in real time by a temperature sensor. The temperature sensor is a device that can convert temperature changes into electrical signals and is used to monitor the temperature changes in the reaction zone in the furnace in real time. At the same time, the smoke concentration data of the rising flue is collected in real time by an online smoke detector. The online smoke detector is an instrument for real-time monitoring of the concentration of particulate matter in the flue gas and can reflect the content of solid particles in the flue gas in real time. These data are crucial for understanding the progress of the reaction in the furnace and provide basic data support for subsequent intelligent prediction and control.

[0065] Specifically, the distributed control system mentioned in step S2 is a system that combines centralized management and decentralized control. It can receive and display the real-time data collected in step S1, including temperature data and smoke concentration data. At the same time, the system can also obtain key parameters such as the ore supply flow rate of the ore bin, the oxygen-enriched gas supply flow rate of the oxygen-enriched pump, and the air supply flow rate of the air pump. The ore supply flow rate refers to the speed at which the ore is transported from the ore bin to the smelting furnace, and the unit is usually tons / hour; the oxygen-enriched gas supply flow rate refers to the speed at which the oxygen-enriched gas pump transports oxygen-enriched gas into the furnace, and the unit is cubic meters / hour; the air supply flow rate refers to the speed at which the air pump transports air into the furnace, and the unit is also cubic meters / hour. These flow parameters play a decisive role in controlling the speed and effect of the reaction in the furnace.

[0066] Preferably, in step S3, after the intelligent prediction module receives the gas flow rate and ore supply flow rate data in step S2, it will perform three-dimensional space grid division and mathematical model iterative calculation. Three-dimensional space grid division refers to dividing the internal space of the flash smelting furnace into multiple small grid units in order to more accurately simulate and calculate the distribution of various physical quantities in the furnace. Mathematical model iterative calculation refers to predicting and calculating multiple physical quantities such as velocity field, temperature field, turbulence field and particle distribution field in the furnace by establishing corresponding physical models, such as fluid mechanics models, thermodynamic models, etc. For example, the prediction of velocity field distribution can be achieved by solving the three-dimensional Navier-Stokes equations, which describe the motion law of fluid in three-dimensional space. By inputting parameters such as gas density, velocity component, pressure, and combining boundary conditions for iterative solution, the velocity distribution of each point in the furnace can be obtained.

[0067] In some embodiments, the velocity field distribution in step S3 is obtained by solving the three-dimensional Navier-Stokes equation, which is expressed as:

[0068] For the X direction:

[0069]

[0070] For the Y direction:

[0071]

[0072] For the Z direction:

[0073]

[0074] Where ρ represents the gas density and its unit is kg / m 3 , u represents the velocity component in the x direction and the unit is m / s, v represents the velocity component in the y direction and the unit is m / s, w represents the velocity component in the z direction and the unit is m / s, p represents the pressure and the unit is Pa, τ represents the viscous stress component and the unit is Pa, g x Represents the acceleration due to gravity in the x direction and its unit is m / s 2 , g y Represents the gravitational acceleration component in the y direction and the unit is m / s 2 , g z Represents the acceleration due to gravity in the z direction and its unit is m / s 2 .

[0075] It should be noted that the velocity field distribution mentioned in the present invention is obtained by solving the three-dimensional Navier-Stokes equations. The Navier-Stokes equations are a set of basic equations that describe fluid motion, which can accurately reflect the velocity changes of the fluid in three-dimensional space. These equations take into account the interaction between multiple physical quantities such as fluid density, velocity, pressure, and viscous stress. By solving these equations, the velocity distribution at different positions in the furnace can be obtained, which is of great significance for optimizing the airflow organization in the furnace and improving the smelting efficiency. In practical applications, the solution of these equations needs to be carried out with the help of numerical calculation methods, combined with the specific geometric structure and boundary conditions in the furnace.

[0076] Specifically, the three-dimensional Navier-Stokes equations include equations in three directions, corresponding to the three coordinate axes X, Y, and Z. In the equation in the X direction, the parameters involved include gas density ρ, x-direction velocity component u, y-direction velocity component v, z-direction velocity component w, pressure p, viscous stress component τ, and x-direction gravity acceleration component g. x These parameters together determine the motion state of the fluid in the X direction. Similarly, the equations for the Y and Z directions also describe the motion laws of the fluid in the corresponding directions. The gas density ρ represents the mass of the gas per unit volume, and the unit is kilograms per cubic meter kg / m 3 The velocity components u, v, and w represent the velocity of the fluid in the three coordinate axes, respectively, and are expressed in meters per second (m / s). The pressure p represents the pressure inside the fluid, and is expressed in Pascals (Pa). The viscous stress component τ reflects the viscous resistance inside the fluid, and is also expressed in Pascals (Pa). The gravitational acceleration component g x 、g y 、g z Represents the components of gravity in three directions, in meters per second squared (m / s) 2 The precise measurement and reasonable setting of these parameters are the basis for solving the Navier-Stokes equations.

[0077] Preferably, when solving the three-dimensional Navier-Stokes equations, a detailed numerical model needs to be constructed. This model first needs to generate a three-dimensional grid based on the specific geometric structure of the flash smelting furnace, dividing the space inside the furnace into multiple small calculation units. Then, the various parameters mentioned above are used as input, combined with the boundary conditions in the furnace, such as inlet velocity, outlet pressure, etc., and the equations are solved step by step through numerical calculation methods. During the calculation process, numerical calculation methods such as finite difference method, finite volume method or finite element method can be used. These methods can discretize continuous partial differential equations and solve them on a computer. In this way, the velocity distribution at different positions in the furnace can be obtained, thereby providing accurate prediction data for subsequent control strategies.

[0078] In some embodiments, the turbulence field distribution in step S3 is obtained by solving a large eddy simulation model, and the model equation is:

[0079]

[0080] Where ρ represents the filtered gas density and is expressed in kg / m 3 ,u i represents the velocity component after filtering and the unit is m / s, P represents the pressure after filtering and the unit is Pa, ρu′ i u′ j Represents the subgrid Reynolds stress component and is in Pa, g i Represents the acceleration due to gravity and its unit is m / s 2 .

[0081] It should be noted that the turbulent field distribution is obtained by solving the large eddy simulation model. Large Eddy Simulation (LES) is an advanced numerical method for simulating turbulent flows. It can capture the large-scale vortex structure in turbulent flows and model the small-scale vortex structure. This method has high accuracy and reliability when dealing with complex industrial flow problems, such as turbulence phenomena in flash smelting furnaces. The large eddy simulation model can more accurately predict the distribution of the turbulent field in the furnace, thereby providing important data support for optimizing the flow characteristics in the furnace and improving the smelting efficiency.

[0082] Specifically, the parameters involved in the large eddy simulation model equation include filtered gas density, filtered velocity component, filtered pressure, subgrid Reynolds stress component, and gravity acceleration component. The filtered gas density represents the gas density value after filtering, and the unit is kilograms per cubic meter kg / m 3 The filtered velocity component refers to the component of the velocity vector in each direction after filtering, and the unit is meter per second m / s. The filtered pressure refers to the pressure value after filtering, and the unit is Pascal Pa. The sub-grid Reynolds stress component represents the stress generated by small-scale turbulent motion, and the unit is Pascal Pa. The gravity acceleration component represents the component of gravity in each direction, and the unit is meter per second squared m / s 2 The accurate measurement and reasonable setting of these parameters are the basis for constructing large eddy simulation models, through which the motion characteristics of fluids at different scales can be described.

[0083] Preferably, when constructing a large eddy simulation model, the original Navier-Stokes equations need to be filtered first to distinguish between large-scale and small-scale turbulent structures. In the filtered equations, the large-scale turbulent structure is solved directly, while the small-scale turbulent structure is simulated by a sub-grid model. The sub-grid model is usually constructed based on parameters such as turbulent kinetic energy and dissipation rate, which can be determined by experimental data or empirical formulas. In actual calculations, it is necessary to use the filtered gas density, velocity component, pressure and other parameters as input, and combine the boundary conditions in the furnace, such as inlet velocity and outlet pressure, to gradually solve the filtered equations through numerical calculation methods. In this way, the turbulent field distribution at different positions in the furnace can be obtained, thereby providing accurate prediction data for subsequent control strategies.

[0084] In some embodiments, the particle distribution field of step S3 is obtained by solving a multiphase particle unit method, and the equation includes a particle momentum equation:

[0085]

[0086] Among them, u s represents the particle velocity vector and is expressed in m / s, ρ s Indicates particle density in kg / m 3 ,θ s represents the particle volume fraction and is dimensionless, τ s represents the particle stress tensor and its unit is Pa, D s represents the air-solid drag coefficient and its unit is s -1 ,u g represents the gas velocity vector and its unit is m / s, g represents the gravity acceleration vector and its unit is m / s 2 .

[0087] It should be noted that the particle distribution field is obtained using the multiphase particle element method. The multiphase particle element method is a numerical method for simulating particle motion in multiphase flows, accurately describing the particle trajectories and distribution within the flow field. This method takes into account the interaction between particles and gas, including drag, pressure gradients, and gravity. This method allows for the prediction of particle distribution within the furnace, which is crucial for optimizing the residence time of particles within the furnace and improving smelting efficiency.

[0088] Specifically, the particle momentum equation in the multiphase particle unit method describes the motion of particles in the flow field. The parameters in the equation include the particle velocity vector, particle density, particle volume fraction, particle stress tensor, gas-solid drag coefficient, gas velocity vector, and gravity acceleration vector. The particle velocity vector represents the velocity of the particle in space, with units of meters per second (m / s). The particle density represents the mass of the particle per unit volume, with units of kilograms per cubic meter (kg / m 3 The particle volume fraction represents the volume ratio of particles in the mixture and is a dimensionless quantity. The particle stress tensor reflects the interaction force between particles and is expressed in Pascals. The gas-solid drag coefficient represents the drag force of the gas on the particles and is expressed in units of seconds. -1 The gas velocity vector represents the speed of the gas in space, and its unit is m / s. The gravity acceleration vector represents the effect of gravity in space, and its unit is m / s 2 The accurate setting and measurement of these parameters are the basis for solving the particle momentum equation, and these parameters can be used to describe the motion state of particles in the flow field.

[0089] Preferably, when constructing a multiphase particle unit method model, it is first necessary to determine the initial position and velocity of the particles. Then, based on the particle momentum equation, the motion trajectory of the particles in the flow field is gradually solved by numerical calculation methods. During the calculation process, it is necessary to input the physical parameters of the particles, such as particle density, particle diameter, etc., as well as the parameters of the flow field, such as gas velocity, pressure distribution, etc. The gas-solid drag coefficient can be determined by experimental data or empirical formulas, which reflects the resistance of the gas to the particles. Through these input parameters, combined with numerical calculation methods, such as the Euler-Lagrange method, the motion trajectory and distribution of the particles can be gradually solved. Ultimately, this method can obtain the particle distribution at different positions in the furnace, thereby providing accurate prediction data for subsequent control strategies.

[0090] In some embodiments, the temperature field distribution in step S3 is obtained by solving the gas-solid two-phase heat conservation equation, which is expressed as:

[0091]

[0092] Among them, θ g represents the gas volume fraction and is dimensionless, h g represents the gas enthalpy value and its unit is J / kg, q represents the heat flux density vector and its unit is W / m 2 , Q D represents the enthalpy diffusion rate and its unit is W / m 3 , S g,s Indicates the gas-solid heat transfer and the unit is W / m 3 , ΔH rgIt represents the heat released by the oxidation reaction and has the unit of J / kg.

[0093] It should be noted that the temperature field distribution is obtained by solving the gas-solid two-phase heat conservation equation. This equation describes the energy transfer and conversion during the interaction between gas and solid particles. It takes into account multiple physical processes, such as heat exchange between gas and solid particles, enthalpy diffusion, and heat release from oxidation reactions. Solving this equation reveals the temperature distribution at different locations within the furnace, which is crucial for optimizing the furnace's thermal balance and improving melting efficiency.

[0094] Specifically, the parameters involved in the gas-solid two-phase heat conservation equation include gas volume fraction, gas enthalpy, heat flux density vector, enthalpy diffusivity, gas-solid heat exchange and oxidation reaction heat release. The gas volume fraction represents the volume ratio of the gas in the mixture and is a dimensionless quantity. The gas enthalpy represents the enthalpy per unit mass of gas, with the unit of joule per kilogram (J / kg). The heat flux density vector represents the direction and magnitude of heat transfer in space, with the unit of watt per square meter (W / m 2 The enthalpy diffusivity is the diffusion rate of enthalpy per unit volume, and its unit is watt per cubic meter W / m 3 The gas-solid heat exchange rate refers to the heat exchanged between gas and solid particles, and its unit is watt per cubic meter W / m 3 The heat released by the oxidation reaction represents the amount of heat released per unit mass of gas during the oxidation reaction, expressed in joules per kilogram (J / kg). Accurate measurement and proper setting of these parameters are fundamental to solving the gas-solid two-phase heat conservation equation. These parameters can be used to describe the heat transfer and conversion process within the furnace.

[0095] Preferably, when constructing a gas-solid two-phase heat conservation equation model, it is first necessary to determine the initial temperature and position of the gas and solid particles. Then, based on the gas-solid two-phase heat conservation equation, the temperature distribution at different positions in the furnace is gradually solved by numerical calculation methods. During the calculation process, it is necessary to input the physical parameters of the gas and solid particles, such as gas density, solid particle density, specific heat capacity, etc., as well as flow field parameters, such as gas velocity, pressure distribution, etc. The enthalpy diffusivity can be determined by experimental data or empirical formulas, which reflects the diffusion of heat in the gas. Through these input parameters, combined with numerical calculation methods, such as the finite difference method or the finite element method, the temperature distribution at different positions in the furnace can be gradually solved. Ultimately, the temperature distribution at different positions in the furnace can be obtained by this method, thereby providing accurate prediction data for subsequent control strategies.

[0096] In some embodiments, step S4 includes:

[0097] Step S4.1: Regulating the gas controller of the air pump based on the velocity field distribution data, specifically generating an air supply flow rate adjustment instruction based on the deviation of the gas flow rate in the central area of ​​the reaction tower from a preset velocity threshold;

[0098] Step S4.2: adjusting the gas controller of the oxygen-enriched gas pump based on the temperature field distribution data, specifically generating an oxygen-enriched gas supply flow adjustment instruction based on the deviation between the wall temperature and a preset temperature safety threshold;

[0099] Step S4.3: Jointly adjusting the gas controllers of the air pump and the oxygen-enriched gas pump based on the turbulence field distribution data, specifically, synchronously generating flow ratio adjustment instructions for air and oxygen-enriched gas based on the deviation of the turbulence intensity coefficient from a preset optimal range;

[0100] Step S4.4: Adjust the feed controller of the ore bin based on the particle distribution field data, specifically, generate an ore supply flow adjustment instruction according to the difference between the particle residence time and the process requirement.

[0101] It should be noted that step S4 uses the multi-physics prediction data obtained in step S3, including velocity, temperature, turbulence, and particle distribution, to comprehensively adjust the flow parameters of the feed controller for the ore silo, the gas controller for the oxygen enrichment pump, and the gas controller for the air pump, using a distributed control system. A distributed control system combines centralized management with decentralized control, enabling precise flow regulation of equipment based on predicted data. This regulation method optimizes reaction conditions within the furnace, improving smelting efficiency and product quality.

[0102] Specifically, in step S4.1, the velocity field distribution data is used to adjust the gas controller of the air pump. Air supply flow rate adjustment instructions are generated based on the deviation between the gas flow rate in the center of the reaction tower and a preset velocity threshold. The velocity field distribution data reflects the velocity of gas flow within the furnace. The preset velocity threshold is a gas flow rate standard set according to process requirements. In step S4.2, the temperature field distribution data is used to adjust the gas controller of the oxygen enrichment pump. Oxygen-enriched gas supply flow rate adjustment instructions are generated based on the deviation between the wall temperature and a preset temperature safety threshold. The wall temperature refers to the temperature of the furnace wall. The preset temperature safety threshold is a temperature standard set according to process requirements. In step S4.3, the turbulence field distribution data is used to jointly adjust the gas controllers of the air pump and oxygen enrichment pump. Air and oxygen-enriched gas flow rate ratio adjustment instructions are generated synchronously based on the deviation between the turbulence intensity coefficient and a preset optimal range. The turbulence intensity coefficient is a parameter that measures the degree of turbulence. The preset optimal range is a turbulence intensity range set according to process requirements. In step S4.4, the particle distribution field data is used to adjust the feed controller of the ore silo. The difference between the pellet residence time and the process requirement is used to generate feed flow control instructions. Particle residence time refers to the time the pellets remain in the furnace, while the process requirement is the standard pellet residence time set based on process requirements.

[0103] Preferably, when implementing step S4, it is first necessary to determine the deviation between the actual value of each physical quantity and the preset threshold value based on the predicted data. For the velocity field distribution data, by monitoring the gas flow rate in the central area of ​​the reaction tower, the deviation from the preset velocity threshold is calculated, thereby determining the adjustment amount of the air supply flow rate. For the temperature field distribution data, by monitoring the wall temperature, the deviation from the preset temperature safety threshold is calculated, thereby determining the adjustment amount of the oxygen-enriched gas supply flow rate. For the turbulence field distribution data, by calculating the deviation between the turbulence intensity coefficient and the preset optimal interval, the flow ratio of air and oxygen-enriched gas is synchronously adjusted. For the particle distribution field data, by calculating the difference between the particle residence time and the process requirement time, the adjustment amount of the ore supply flow rate is determined. In actual operation, these adjustment instructions can be automatically sent to the corresponding controller through the distributed control system to achieve real-time optimization of the reaction conditions in the furnace.

[0104] In some embodiments, the color gradation display in step S5 includes:

[0105] Step S5.1: Velocity field distribution data is displayed using the dual dimensions of streamline arrow length and color depth, with red representing high-speed areas and blue representing low-speed areas;

[0106] Step S5.2: The temperature field distribution data is rendered using isothermal surface technology, with red representing high-temperature areas and blue representing low-temperature areas, and the color scale is divided according to the preset temperature gradient;

[0107] Step S5.3: The particle distribution field data is displayed using a density cloud map. The transparency is inversely proportional to the particle concentration, and the high-concentration area is displayed as a dark dense dot matrix.

[0108] It should be noted that the color-graded display in step S5 displays the predicted data from step S3 in an intuitive, real-time color format. This display method allows operators to quickly understand the distribution of different physical quantities within the furnace, enabling more effective monitoring and adjustment. The color-graded display includes displays of velocity field distribution, temperature field distribution, and particle distribution field. Each display method uses specific colors and graphic techniques to distinguish different physical quantities and their intensities.

[0109] Specifically, the velocity field distribution data is displayed in two dimensions: the length of the streamline arrow and the depth of the color. The red color represents the high-speed area, and the blue color represents the low-speed area. The length of the streamline arrow and the depth of the color together reflect the speed and direction of the gas flow. The temperature field distribution data uses isothermal surface rendering technology, red represents the high-temperature area, blue represents the low-temperature area, and the color scale is divided according to the preset temperature gradient. The isothermal surface rendering technology uses color gradients to represent temperature changes, making the temperature distribution more intuitive. The particle distribution field data is displayed using a density cloud map. The transparency is inversely proportional to the particle concentration, and the high-concentration area is displayed as a dark dense dot matrix. The density cloud map represents the distribution of particles by the depth of color and the density of the dot matrix. The change in transparency makes the difference in particle concentration more obvious.

[0110] Preferably, when implementing step S5, it is first necessary to generate a corresponding color-graded display image based on the predicted data. For the velocity field distribution, the predicted velocity data can be converted into streamline arrows, the length of the arrows is scaled according to the velocity, and the color is selected according to the velocity range. For example, the area with a velocity of 0-10m / s is represented by light blue, the area with a velocity of 10-20m / s is represented by dark blue, and so on. For the temperature field distribution, the predicted temperature data can be converted into an isothermal surface, and different colors are selected according to the temperature range. For example, the area with a temperature of 800-1000℃ is represented by light red, and the area with a temperature of 1000-1200℃ is represented by dark red. For the particle distribution field, the predicted particle concentration data can be converted into a density cloud map, and different colors and transparency are selected according to the particle concentration. For example, the particle concentration is 0-100kg / m 3 The area is indicated in light grey, 100-200 kg / m 3 These display images can be generated by specialized software tools and updated in real time on the monitoring screen so that operators can keep abreast of the dynamic changes in the furnace.

[0111] In some embodiments, the three-dimensional space grid division in step S3 includes:

[0112] Step S3.1: Generate a hexahedral structured mesh that matches the geometry of the flash smelting furnace;

[0113] Step S3.2: Perform first-level mesh refinement in the gas injection port and mineral injection port areas;

[0114] Step S3.3: Perform the second level of mesh refinement in the molten pool reaction zone and the flue transition zone;

[0115] Step S3.4: Dynamically optimize the mesh size according to the fluid dynamic characteristics.

[0116] It should be noted that the three-dimensional meshing in step S3 is intended to address the complex physical field distribution within the flash smelting furnace. By dividing the furnace space into multiple small grid cells, the distribution of various physical quantities within the furnace can be more accurately simulated and calculated. This meshing method can adapt to the flow characteristics and reaction intensity of different areas within the furnace, thereby improving the accuracy and efficiency of the calculation. Three-dimensional meshing is a critical step in numerical simulation, directly affecting the accuracy of the simulation results and the consumption of computing resources.

[0117] Specifically, the three-dimensional space meshing includes generating a hexahedral structured grid that matches the geometric structure of the flash smelting furnace. The hexahedral structured grid is a regular meshing method that can adapt well to complex geometric shapes and is easy to calculate. The first level of mesh encryption is performed in the gas injection port and ore injection port areas because the flow in these areas changes dramatically and requires a finer mesh to capture the flow details. The second level of mesh encryption is performed in the molten pool reaction zone and the flue transition zone. These areas are key areas for reaction and flow in the furnace and also require high-precision meshing. Finally, the mesh size is dynamically optimized according to the fluid dynamics characteristics, which means that during the simulation process, the size and density of the mesh are dynamically adjusted according to the changes in the flow field to adapt to different flow conditions.

[0118] Preferably, when implementing the three-dimensional space grid division of step S3, it is first necessary to generate an initial hexahedral structured grid based on the specific geometric model of the flash smelting furnace. This can be done through professional grid generation software, which will automatically generate a regular grid based on the geometric shape of the furnace. Next, in the gas injection port and ore injection port area, the grid is divided into finer parts according to parameters such as injection velocity and flow rate to capture the complex flow during the injection process. In the molten pool reaction zone and flue transition zone, the grid is further encrypted according to the reaction intensity and flow changes. Finally, the entire grid is dynamically optimized by fluid dynamics simulation software, and the size and density of the grid are adjusted in real time according to the changes in physical quantities such as velocity field and pressure field during the simulation process. For example, if the turbulence intensity in a certain area increases, the grid will be automatically encrypted to improve the simulation accuracy. This dynamic optimization process can ensure that the most accurate simulation results are obtained when computing resources are limited.

[0119] In some embodiments, the data storage module operation in step S5 includes:

[0120] Step S5.4: Establish a multi-physical quantity prediction database to store time series data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field;

[0121] Step S5.5: Construct an operation parameter-prediction data mapping relationship table. When it is detected that the matching degree between the current operation parameter and the historical record exceeds a preset threshold, directly call the historical prediction data and output it to the display module.

[0122] It should be noted that the data storage module operation in step S5 is intended to effectively manage and utilize the correspondence between historical prediction data and operating parameters. By establishing a multi-physics prediction database, time series data for velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field can be stored. This data not only facilitates real-time monitoring and analysis of furnace conditions, but can also be used to optimize control strategies and predict future operating conditions. Constructing a mapping table between operating parameters and prediction data enables rapid access to historical prediction data that matches the current operating parameters, thereby improving the system's response speed and decision-making efficiency.

[0123] Specifically, the multi-physical quantity prediction database is a system for storing and managing prediction data of various physical quantities in the furnace. It can record the changes of data such as velocity field, temperature field, turbulence field and particle distribution field over time. These data are stored in the form of time series to facilitate subsequent analysis and query. The operating parameter-prediction data mapping relationship table is a data structure used to store the correspondence between operating parameters and prediction data. When the system detects that the current operating parameters match a parameter in the historical records and the degree of match exceeds the preset threshold, the corresponding historical prediction data can be directly called without the need to re-calculate complex data. The establishment of this mapping relationship table, based on the analysis and mining of historical data, can quickly provide prediction results and improve the real-time performance and efficiency of the system.

[0124] Preferably, when implementing the data storage module operation of step S5, it is first necessary to design and construct a multi-physical quantity prediction database. This includes defining the structure of the database, determining the data type and format of the storage, and setting the index and query mechanism of the data. For example, a relational database or a non-relational database can be used to store time series data, and a suitable storage scheme can be selected according to actual needs. For the construction of the operating parameter-prediction data mapping relationship table, it is necessary to analyze the historical data, extract the key operating parameters, and establish the association between these parameters and the prediction data. For example, a machine learning algorithm can be used to analyze the historical data, find the best matching relationship between the operating parameters and the prediction results, and store these relationships in the mapping table. When the system is running, by monitoring the operating parameters in real time and comparing them with the data in the mapping table, the corresponding prediction data can be quickly called, thereby achieving rapid response and optimized control of the state in the furnace.

[0125] In some embodiments, the iterative calculation in step S3 adopts a heterogeneous computing architecture, including:

[0126] Step S3.5: The central processing unit performs velocity field distribution and temperature field distribution calculations, and solves the Navier-Stokes equations and the heat conservation equation;

[0127] Step S3.6: The graphics processing unit performs turbulence field distribution and particle distribution field calculations, completing large eddy simulation and parallel solution of multiphase particle unit equations;

[0128] Step S3.7: The central processing unit and the graphics processing unit exchange boundary condition data via a high-speed bus to achieve dynamic coupling of the calculation process.

[0129] It should be noted that the iterative calculations in step S3 utilize a heterogeneous computing architecture to improve computational efficiency and processing power. This architecture combines the strengths of the central processing unit (CPU) and the graphics processing unit (GPU). By rationally allocating computational tasks, it can rapidly complete complex three-dimensional meshing and iterative mathematical model calculations. This architecture is particularly well-suited for handling the complex calculations of multiple physical quantities within a flash smelting furnace, as it can simultaneously handle a large number of parallel computing tasks, significantly reducing computation time and improving the system's real-time performance and responsiveness.

[0130] Specifically, the central processing unit (CPU) performs velocity field distribution and temperature field distribution calculations, and solves the Navier-Stokes equations and the heat conservation equation. The CPU performs well in processing complex logical operations and sequential execution tasks, and is suitable for processing computing tasks that require precise control and complex algorithms. The graphics processing unit (GPU) performs turbulence field distribution and particle distribution field calculations, and completes large eddy simulation and parallel solution of multiphase particle unit equations. The GPU has a large number of cores and can process multiple computing tasks at the same time, and is suitable for processing parallel computing tasks such as turbulence simulation and particle distribution calculation. The central processing unit and the graphics processing unit exchange boundary condition data through a high-speed bus to achieve dynamic coupling of the computing process. The high-speed bus ensures the efficiency and real-time performance of data transmission, allowing the CPU and GPU to work together to complete complex computing tasks.

[0131] Preferably, when implementing the iterative calculation in step S3, it is first necessary to select an appropriate algorithm and model based on the specific computational task. For the calculation of velocity and temperature fields, numerical calculation methods such as the finite difference method or the finite element method can be used. These methods can accurately solve the Navier-Stokes equations and the heat conservation equation. For the calculation of turbulent flow and particle distribution fields, advanced numerical simulation methods such as large eddy simulation (LES) and multiphase particle unit method (MPU) can be used. These methods can effectively handle complex turbulent and multiphase flow problems. During the calculation process, it is necessary to rationally allocate computational tasks to the CPU and GPU. For example, complex logical operations and sequential execution tasks can be assigned to the CPU, while parallel computing tasks can be assigned to the GPU. In this way, the advantages of the CPU and GPU can be fully utilized to improve computational efficiency. At the same time, it is necessary to ensure efficient and accurate data transmission between the CPU and GPU, and to exchange boundary condition data via a high-speed bus to achieve dynamic coupling of the computational process. For example, in each iterative calculation, the CPU and GPU need to exchange the latest boundary condition data to ensure the accuracy and consistency of the calculation results. This heterogeneous computing architecture can significantly improve the efficiency and accuracy of multi-physical quantity calculations within the flash smelting furnace.

[0132] The above embodiments of the present invention have the following beneficial effects:

[0133] 1. Dynamic prediction of the smelting process can be achieved based on multi-physics field coupling modeling (velocity field, temperature field, turbulence field, particle distribution field). By adjusting the mineral supply and gas flow parameters in real time, the problems of strong hysteresis and insufficient adjustment accuracy of traditional manual control can be solved, and the reaction stability and process index compliance rate can be improved.

[0134] 2. The heterogeneous computing architecture (CPU processes Navier-Stokes equations / GPU parallel calculates turbulence and particle fields) can achieve rapid solutions to complex three-dimensional models, solving the problem of low computing efficiency and inability to meet real-time control requirements of traditional single-machine computing, and ensuring that the system completes full-process simulation and decision-making in milliseconds.

[0135] 3. The color-graded visualization and historical data mapping functions (streamline arrows display velocity fields, isothermal surfaces render temperature fields, and density cloud maps display particle distribution) can solve the problem that manual monitoring is difficult to capture the dynamic correlation of multi-dimensional parameters, and provide operators with an intuitive basis for judging process status.

[0136] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a terminal device such as a computer, a server, a mobile phone, or a tablet.

[0137] The above descriptions are merely some preferred embodiments of the present invention and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also encompass other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned inventive concept. For example, a technical solution formed by mutually replacing the above-mentioned features with (but not limited to) technical features having similar functions disclosed in the embodiments of the present invention.

Claims

1. A method for controlling concentrate distribution in a flash furnace based on intelligent prediction, characterized in that: The following steps are involved: Step S1: collecting temperature data of the flash smelting furnace reaction zone in real time through a temperature sensor, and collecting smoke concentration data of the ascending flue in real time through a smoke online detector; Step S2: The distributed control system receives and displays the real-time data of step S1, and simultaneously obtains the ore supply flow rate of the ore silo, the oxygen-enriched gas supply flow rate of the oxygen-enriched gas pump, and the air supply flow rate of the air pump; Step S3: The intelligent prediction module receives the gas flow rate and ore supply flow rate data from step S2, performs three-dimensional space grid division and mathematical model iterative calculation, and outputs multi-physical quantity prediction data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field inside the flash smelting furnace; Step S4: Based on the multi-physical quantity prediction data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field obtained in step S3, the flow parameters of the feed controller of the ore silo, the gas controller of the oxygen enrichment pump, and the gas controller of the air pump are comprehensively adjusted through the distributed control system; Step S5: The display module displays the prediction data of step S3 in real time in a color-graded form, and the data storage module records the corresponding relationship between the historical prediction data and the operating parameters.

2. The method according to claim 1, characterized in that The velocity field distribution in step S3 is obtained by solving the three-dimensional Navier-Stokes equation, which is expressed as: For the X direction: For the Y direction: For the Z direction: Where ρ represents the gas density, u represents the velocity component in the x direction, v represents the velocity component in the y direction, w represents the velocity component in the z direction, p represents the pressure, τ represents the viscous stress component, and g x Represents the gravitational acceleration component in the x direction, g y Represents the gravitational acceleration component in the y direction, g z Represents the gravitational acceleration component in the z direction.

3. The method according to claim 2, characterized in that The turbulence field distribution in step S3 is obtained by solving the large eddy simulation model, and the model equation is: Where ρ represents the gas density after filtering, u i represents the velocity component after filtering, P represents the pressure after filtering, ρu′ i u′ j represents the subgrid Reynolds stress component, g i represents the gravitational acceleration component.

4. The method according to claim 3, characterized in that The particle distribution field in step S3 is obtained by solving the multiphase particle unit method, and the equation includes the particle momentum equation: Among them, u s represents the particle velocity vector, ρ s represents the particle density, θ s represents the particle volume fraction, τ s represents the particle stress tensor, D s represents the gas-solid drag coefficient, u g represents the gas velocity vector, and g represents the gravitational acceleration vector.

5. The method according to claim 4, characterized in that The temperature field distribution in step S3 is obtained by solving the gas-solid two-phase heat conservation equation, which is expressed as: Among them, θ g represents the gas volume fraction, h g represents the gas enthalpy value, q represents the heat flux vector, Q D represents the enthalpy diffusion rate, S g,s Indicates the gas-solid heat transfer, ΔH rg Indicates the heat released by the oxidation reaction.

6. The method according to claim 1, characterized in that The step S4 comprises: Step S4.1: Regulating the gas controller of the air pump based on the velocity field distribution data, specifically generating an air supply flow rate adjustment instruction based on the deviation of the gas flow rate in the central area of ​​the reaction tower from a preset velocity threshold; Step S4.2: adjusting the gas controller of the oxygen-enriched gas pump based on the temperature field distribution data, specifically generating an oxygen-enriched gas supply flow adjustment instruction based on the deviation between the wall temperature and a preset temperature safety threshold; Step S4.3: Jointly adjusting the gas controllers of the air pump and the oxygen-enriched gas pump based on the turbulence field distribution data, specifically, synchronously generating flow ratio adjustment instructions for air and oxygen-enriched gas based on the deviation of the turbulence intensity coefficient from a preset optimal range; Step S4.4: Adjust the feed controller of the ore bin based on the particle distribution field data, specifically, generate an ore supply flow adjustment instruction according to the difference between the particle residence time and the process requirement.

7. The method according to claim 1, characterized in that The color gradation display in step S5 includes: Step S5.1: Velocity field distribution data is displayed using the dual dimensions of streamline arrow length and color depth, with red representing high-speed areas and blue representing low-speed areas; Step S5.2: The temperature field distribution data is rendered using isothermal surface technology, with red representing high-temperature areas and blue representing low-temperature areas, and the color scale is divided according to the preset temperature gradient; Step S5.3: The particle distribution field data is displayed using a density cloud map. The transparency is inversely proportional to the particle concentration, and the high-concentration area is displayed as a dark dense dot matrix.

8. The method according to claim 1, characterized in that The three-dimensional space grid division in step S3 includes: Step S3.1: Generate a hexahedral structured mesh that matches the geometry of the flash smelting furnace; Step S3.2: Perform first-level mesh refinement in the gas injection port and mineral injection port areas; Step S3.3: Perform the second level of mesh refinement in the molten pool reaction zone and the flue transition zone; Step S3.4: Dynamically optimize the mesh size according to the fluid dynamic characteristics.

9. The method according to claim 1, characterized in that The data storage module operation in step S5 includes: Step S5.4: Establish a multi-physical quantity prediction database to store time series data of velocity field distribution, temperature field distribution, turbulence field distribution, and particle distribution field; Step S5.5: Construct an operation parameter-prediction data mapping relationship table. When it is detected that the matching degree between the current operation parameter and the historical record exceeds a preset threshold, directly call the historical prediction data and output it to the display module.

10. The method according to claim 1, characterized in that The iterative calculation in step S3 adopts a heterogeneous computing architecture, including: Step S3.5: The central processing unit performs velocity field distribution and temperature field distribution calculations, and solves the Navier-Stokes equations and the heat conservation equation; Step S3.6: The graphics processing unit performs turbulence field distribution and particle distribution field calculations, completing large eddy simulation and parallel solution of multiphase particle unit equations; Step S3.7: The central processing unit and the graphics processing unit exchange boundary condition data via a high-speed bus to achieve dynamic coupling of the calculation process.

Images