A top-blown smelting furnace intelligent control system and method
By using an intelligent control system to monitor and predict the lance parameters of the top-blown smelting furnace in real time and adjust the operating strategy, the problem of relying on manual experience in traditional smelting control has been solved, and efficient, precise control and improved stability of the smelting process have been achieved.
Patent Information
- Application Number
- CN202410284084.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-13
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-03-13
AI Technical Summary
Traditional top-blown smelting furnaces rely on manual experience for control and lack real-time monitoring of the internal state of the molten pool. This leads to problems such as slag splashing, increased energy consumption, fluctuations in raw material quantity, and uneven stirring, affecting metal output and quality. Furthermore, the inaccurate and unreliable adjustment limits the optimization of the smelting process.
An intelligent control system is adopted, which monitors the parameters of the top-blown spray gun in real time through the data acquisition module and uses an intelligent prediction module to predict the distribution and changes of smelting parameters. The control module adjusts the spray gun operation strategy according to the prediction results, including spray gun parameters, gun position height, air flow rate and material supply rate, so as to achieve precise control.
It has enabled a comprehensive analysis of the complex physical environment inside the smelting furnace, improved the accuracy and efficiency of the smelting process, reduced the risk of operational errors, optimized the quantitative control of smelting process parameters, and enhanced the furnace's service life and production stability.
Smart Images

Figure CN118210281B_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present disclosure relate to the technical field of metal smelting, and in particular to an intelligent control system and method for a top-blown smelting furnace. Background Art
[0002] Top-blown molten pool smelting technology is widely used in the field of tin, copper, lead and other metal refining due to its high smelting intensity and outstanding smelting efficiency. With the development of the smelting industry, the requirements for the operating accuracy and efficiency of top-blown smelting furnaces are gradually increasing. The traditional control method that relies on manual experience and intuition can no longer meet the needs of modern smelting.
[0003] At present, the commonly used monitoring methods are limited to the gas flow and pressure at the spray gun, and there is no way to know the velocity field and pressure changes inside the molten pool. This lack of information leads to a series of problems, such as slag splashing, increased energy consumption, fluctuations in raw material quantity, and uneven stirring. These problems not only affect the output and quality of the metal, but also cause furnace stagnation and even the shutdown and scrapping of the entire smelting furnace in severe cases. In terms of regulation, it currently mainly relies on the experience and judgment of the staff, and performs extensive regulation by adjusting the flow of spray gun gas and coal powder. Due to the lack of a complete quantitative regulation mechanism, this regulation method is neither accurate nor reliable, which seriously restricts the further optimization and development of the top-blown smelting furnace process.
[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention
[0005] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.
[0006] Some embodiments of the present disclosure propose an intelligent control system, method and computer-readable medium for a top-blown smelting furnace to solve one or more of the technical problems mentioned in the above background technology section.
[0007] In the first aspect, some embodiments of the present disclosure provide an intelligent control system for a top-blown smelting furnace, comprising: a smelting device, including a top-blowing lance and a smelting furnace; a data acquisition module, used to collect a variety of lance parameters of the top-blowing lance, the lance parameters at least including gas flow, pressure and temperature information; an intelligent prediction module, used to receive the lance parameters and obtain prediction results through a preset algorithm, the prediction results characterizing the distribution and changes of smelting parameters in the molten pool of the smelting furnace, the smelting parameters at least including smelting pressure, smelting speed, slag distribution; a control module, used to control the top-blowing lance to adjust the operation strategy based on the prediction results, wherein the operation strategy includes the following steps: adjusting the lance parameters based on the prediction results to meet preset working conditions; controlling the gun position height of the top-blowing lance based on the prediction results; controlling the ventilation volume and feed rate of the top-blowing lance based on the prediction results.
[0008] In the second aspect, some embodiments of the present disclosure provide an intelligent control method for a top-blown smelting furnace, comprising: collecting the spray gun parameters of a top-blowing spray gun, the spray gun parameters including at least gas flow, pressure and temperature information; receiving the spray gun parameters and obtaining prediction results through a preset algorithm, the prediction results characterizing the distribution and changes of smelting parameters in the molten pool of the smelting furnace, the smelting parameters including at least smelting pressure, smelting speed, and slag distribution; controlling the top-blowing spray gun to adjust the operation strategy based on the prediction results, wherein the operation strategy includes the following steps: adjusting the spray gun parameters based on the prediction results to meet preset working conditions; controlling the gun position height of the top-blowing spray gun based on the prediction results; controlling the ventilation volume and feed rate of the top-blowing spray gun based on the prediction results.
[0009] In a third aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner in the second aspect above is implemented.
[0010] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: a comprehensive analysis of the complex and changeable physical environment in the smelting furnace is achieved through the intelligent prediction module. Mathematical model iteration technology is used to perform high-precision simulation and calculation of various operating conditions, thereby revealing the deep-level dynamic characteristics inside the molten pool, including splashing details, stirring mechanism, stirring dead zone morphology, turbulent structure, dynamic pressure fluctuations, pressure gradient evolution, and fine changes in flow velocity and flow trajectory. The acquisition of this information helps the staff understand the conditions in the furnace and automatically and accurately control the gas and fuel flow in the spray gun according to the actual situation, thereby improving the operating life of the furnace body and optimizing the quantitative joint control of the smelting process parameters. This kind of automated control guided by prediction results ensures the efficiency and accuracy of the smelting process, while reducing the dependence on manual experience and significantly reducing the risk of operating errors. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.
[0012] Figure 1 It is a schematic diagram of the structure of some embodiments of the intelligent control system of the top-blown smelting furnace according to the present disclosure;
[0013] Figure 2 is a flow chart of some embodiments of the top-blown smelting furnace intelligent control method suitable for implementing some embodiments of the present disclosure;
[0014] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION
[0015] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.
[0016] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.
[0017] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0018] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0019] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.
[0020] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.
[0021] Figure 1The schematic diagram of the structure of some embodiments of the top-blown smelting furnace intelligent control system according to the present disclosure is shown. The top-blown smelting furnace intelligent control system 100 includes: a smelting device 101, a data acquisition module 102, an intelligent prediction module 103 and a control module 104.
[0022] In some embodiments, the workflow of the top-blowing smelting furnace intelligent control system 100 can be: the data acquisition module 102 collects the spray gun parameters of the top-blowing spray gun 1011 in real time, such as gas flow, pressure and temperature. The intelligent prediction module 103 receives these data, and uses a preset algorithm to process and analyze them to obtain the distribution and change prediction results of the smelting parameters of the molten pool in the smelting furnace 1012. Based on the prediction results, the control module 104 adjusts the operation strategy of the top-blowing spray gun 1011 in real time, including adjusting the spray gun parameters, controlling the gun position height, ventilation volume and feed amount, etc., to optimize the smelting process and improve the smelting efficiency.
[0023] In some embodiments, the smelting device 101 includes a top-blowing lance 1011 and a smelting furnace 1012. The top-blowing lance 1011 is a key component in the smelting process and is responsible for injecting gas (such as oxygen, nitrogen, etc.) and / or fuel (such as pulverized coal, natural gas, etc.) into the smelting furnace 1012. The design and operation of the top-blowing lance 1011 directly affect the smelting efficiency, energy consumption and product quality.
[0024] In some embodiments, the smelting furnace 1012 is the main equipment for metal smelting, and its internal temperature and chemical environment need to be precisely controlled to achieve an efficient smelting process. The smelting furnace usually has the characteristics of high temperature resistance and corrosion resistance to adapt to extreme smelting conditions. The smelting furnace 1012 can be an Ausmelt furnace, which is mainly composed of a circular reactor, and metal smelting is carried out by continuously delivering fuel to the molten pool through a top-blowing lance 1011 installed in the center of the top. The temperature of the smelting furnace 1012 is usually as high as thousands of degrees Celsius. When the flow rate of gas and pulverized coal sprayed from the top-blowing lance 1011 is too high or too low, it will have a serious impact on the output and quality of the metal and the stable operation of the equipment.
[0025] In some embodiments, the data acquisition module 102 is responsible for real-time monitoring and collecting various parameters of the top-blowing lance 1011, such as gas flow, injection pressure, temperature, etc. These data are important bases for evaluating the working state of the top-blowing lance 1011 and the efficiency of the smelting process.
[0026] In some embodiments, the spray gun parameter may be the flow rate of gas, fuel or oxygen through the spray gun. The flow rate directly affects the spray force and effect produced by the spray gun, and further affects the rate and efficiency of the smelting reaction. If the flow rate is too low, the movement of the material in the furnace may not be fully stimulated; if the flow rate is too high, it may cause energy waste and even lead to uneven temperature distribution in the furnace.
[0027] In some embodiments, pressure is another key parameter. Changes in pressure can reflect the blockage of the top-blowing lance 1011, the stability of the gas supply, and the degree of nozzle wear. Too high or too low pressure may have a negative impact on the performance and life of the lance. Temperature information is also one of the core parameters in the smelting process. For the top-blowing lance 1011, the temperatures that need to be monitored include the temperature of the top-blowing lance 1011 itself and the temperature of different areas in the smelting furnace 1012. Excessively high temperature of the top-blowing lance 1011 may cause material deformation or failure, while uneven temperature in the smelting furnace 1012 will affect the quality and efficiency of the smelting reaction.
[0028] In some embodiments, the real-time collection of the working parameters of the top-blowing lance 1011 also includes heat distribution information, which refers to the heat distribution in different areas of the furnace. By analyzing the heat distribution, it is possible to understand whether the injection pattern of the top-blowing lance 1011 is reasonable, the movement state of the material in the furnace, and the utilization efficiency of the heat energy. Uneven heat distribution may lead to uneven composition of the smelting product or unnecessary energy consumption.
[0029] In some embodiments, the data acquisition module 102 can be implemented by various sensors, such as flow sensors, pressure sensors, and temperature sensors. These sensors need to have high precision and high reliability to ensure the accuracy of the data. For example, the flow sensor is used to measure the gas flow, the pressure sensor is used to measure the injection pressure, and the temperature sensor is used to measure the temperature. These sensors convert the collected analog signals into digital signals and then transmit them to the processor or control system for processing.
[0030] In some embodiments, the intelligent prediction module 103 receives the real-time data provided by the data acquisition module 102, and uses a preset algorithm (such as machine learning, deep learning, etc.) to analyze and process these data. By analyzing historical data and real-time data, the intelligent prediction module 103 can predict the state changes of the molten pool in the smelting furnace, such as smelting pressure, smelting speed, slag distribution, etc. These prediction results are crucial for timely adjustment of the spray gun operation strategy. The intelligent prediction module 103 is the core part of the present disclosure. It is responsible for receiving the spray gun parameters transmitted by the data acquisition module 102, and using a preset algorithm to process and analyze these parameters to obtain the prediction results of the distribution and changes of the smelting parameters of the molten pool in the smelting furnace.
[0031] In some embodiments, the smelting parameters may be smelting pressure, smelting speed, slag distribution, etc. The smelting pressure may be the pressure in the molten pool in the smelting furnace 1012. The change in smelting pressure may reflect the distribution and flow of gas in the molten pool, as well as the intensity of the smelting reaction. The smelting speed may be the oxidation or reduction speed of the metal in the molten pool per unit time. The speed of the smelting speed directly affects the production efficiency of smelting. Slag distribution The distribution and thickness of slag in the molten pool. The distribution of slag may reflect the uniformity of the chemical reaction in the molten pool and the efficiency of heat transfer.
[0032] In some embodiments, the control module 104 is responsible for adjusting the operation strategy of the top-blowing lance 1011 in real time according to the prediction results provided by the intelligent prediction module 103. The adjustment of the operation strategy is intended to optimize the smelting process and ensure that the working conditions in the smelting furnace are always kept in the best state. The control module 104 achieves comprehensive monitoring and real-time adjustment of the smelting process by accurately controlling key factors such as lance parameters, lance height, ventilation volume and feed volume.
[0033] In some embodiments, the control module 104 first determines whether the current spray gun parameters meet the preset working conditions based on the prediction results provided by the intelligent prediction module 103, such as smelting pressure, smelting speed and slag distribution. If not, the control module 104 will calculate the parameter values that need to be adjusted and send instructions to the spray gun actuator for adjustment. The adjusted parameters may include gas flow rate, injection pressure, etc. For example, if the prediction results show that the smelting speed is too slow, the control module 104 may increase the gas flow rate or increase the injection pressure to speed up the smelting reaction.
[0034] In some embodiments, the gun height refers to the distance between the end of the top-blowing spray gun 1011 and the surface of the molten pool. The control module 104 adjusts the gun height according to the change in the state of the molten pool in the prediction results. For example, when the molten pool temperature is too high or the slag is unevenly distributed, the control module 104 may lower the gun height to increase the stirring effect of the gas on the molten pool and promote the uniform distribution of heat and components. The adjustment of the gun height is usually achieved by controlling the spray gun lifting mechanism. The control module 104 calculates the gun height value that needs to be adjusted based on the prediction results and the preset gun height adjustment strategy, and sends instructions to the lifting mechanism to perform the adjustment.
[0035] In some embodiments, the ventilation volume refers to the total amount of gas injected into the smelting furnace 1012 through the top-blowing lance 1011 per unit time, and the feed volume refers to the amount of raw materials put into the smelting furnace 1012 per unit time. The control module 104 will adjust the ventilation volume and feed volume according to the changes in parameters such as smelting speed and molten pool composition in the prediction results to ensure the continuity and stability of the smelting process. The adjustment of the ventilation volume and feed volume can be achieved by controlling the gas supply system and the raw material delivery system. The control module 104 will calculate the parameter values such as the gas flow rate and raw material delivery speed that need to be adjusted based on the prediction results and the preset ventilation volume and feed volume adjustment strategy, and send instructions to the corresponding actuators for adjustment.
[0036] In some embodiments, the intelligent prediction module 103 further includes a data iteration module 1031, which is used to perform iterative training of the mathematical model according to the spray gun parameters, and is configured to perform the following operations: using the spray gun parameters collected in real time as initial conditions, and setting the historical spray gun parameters as training samples. The data iteration module 1031 is an important part of the module, and the accuracy and real-time performance of the prediction can be improved by continuously iterating and optimizing the mathematical model.
[0037] In some embodiments, the initial conditions and historical spray gun parameters are input into a preconfigured data iteration model. The data iteration model may include: a full convolution module, a sequence of residual convolution modules, and a pooling module; a training sample is selected from the training sample set, and the following training steps are performed: the training sample is input into the full convolution module included in the initial data iteration model to obtain full convolution information. The full convolution module is responsible for receiving the initial conditions (spray gun parameters collected in real time) and performing global feature extraction. Through convolution operations, it can capture the spatial structure and correlation in the data.
[0038] In some embodiments, the full amount of convolution information is input into the sequence of residual convolution modules included in the initial data iteration model to obtain residual convolution information, which focuses on capturing local details and changing trends in the data. Through the stacking of multiple residual blocks, it can effectively extract deep features and alleviate the problem of gradient disappearance. The residual convolution information is input into the pooling module included in the initial data iteration model to obtain initial spray gun parameter information, which reduces and integrates the extracted features and outputs more representative information. This helps to reduce the complexity and calculation amount of the model and improve the prediction speed. Based on a preset loss function, determine the abnormal difference value between the initial spray gun parameter feature information and the sample spray gun parameter information included in the training sample; in response to determining that the abnormal difference value is less than the target value, determine the initial data iteration model as the data iteration model; and / or in response to determining that the abnormal difference value is greater than or equal to the target value, adjust the relevant parameters in the initial data iteration model, determine the adjusted initial data iteration model as the initial data iteration model, and select training samples from the training sample set for re-executing the training step.
[0039] In some embodiments, first, the system collects multiple spray gun parameters of the top-blowing spray gun 1011 in real time, such as gas flow, pressure, temperature, etc. These data are original and unprocessed, and need to be further analyzed and processed by the intelligent prediction module 103. At the same time, the system saves historical spray gun parameters as training samples. These historical data contain various situations and corresponding results in the past smelting process, providing rich learning materials for data iteration. During the training process, the system randomly selects a batch of training samples from the training sample set and inputs them into the initial data iteration model. These samples contain various different spray gun parameter combinations and corresponding smelting results. Through the forward propagation process, the model outputs initial spray gun parameter information. This information reflects the model's understanding and prediction results of the current input data. Then, the system uses a preset loss function to calculate the difference between the initial spray gun parameter feature information and the sample spray gun parameter information in the training sample. This difference value reflects the degree of deviation between the model's prediction results and the actual data. Commonly used loss functions include mean square error, cross entropy, etc. If the difference value is less than the target value (i.e., the preset accuracy requirement is met), it means that the model's prediction ability is good enough, and the initial data iteration model can be determined as the final data iteration model; otherwise, it is necessary to adjust the relevant parameters in the model (such as convolution kernel size, learning rate, etc.) to optimize the performance of the model, and perform the training step again until the accuracy requirement is met. This process usually requires multiple iterations to complete.
[0040] In some embodiments, when the data iteration model training is completed and reaches the preset accuracy requirements, it can be applied to the actual smelting process for real-time prediction. During the prediction process, the model receives the real-time collected spray gun parameters as input and outputs the corresponding prediction results (such as smelting pressure, smelting speed, slag distribution, etc.). These prediction results can provide a basis for the decision-making of the control system and achieve precise control of the smelting process. At the same time, as the smelting process progresses and data continues to accumulate, the system can also regularly update and optimize the data iteration model to adapt to new smelting conditions and changes in demand. This can be achieved by adding new training samples, adjusting model parameters, or introducing new algorithms.
[0041] In some embodiments, the intelligent module 103 may also include a physical model module 1032, a parameter calculation module 1033, and a post-processing module 1034. The physical model module 1032 is used to construct a physical three-dimensional coordinate model based on the smelting equipment in actual use. The physical three-dimensional coordinate model at least represents the structure and liquid level information of the smelting furnace and the gun position information of the top-blowing lance 1011, and the physical three-dimensional coordinate model is grid-divided, and each grid unit represents the three-dimensional coordinate. The physical model module 1032 is one of the important components of the intelligent prediction module 103. It constructs a physical three-dimensional coordinate model based on the smelting equipment in actual use. This model at least represents the structure, liquid level information and gun position information of the top-blowing lance 1011 of the smelting furnace 1012. By gridding this physical model, each grid unit is assigned a unique three-dimensional coordinate. In this way, the system can accurately simulate and track the state changes of each position in the smelting furnace. The establishment of the physical model module 1032 provides a basis for subsequent parameter calculation and prediction. Through close integration with the actual smelting process, the module can accurately reflect the physical environment and process conditions inside the smelting furnace, providing an important basis for the decision-making of the control system.
[0042] In some embodiments, the construction of the physical three-dimensional coordinate model can be carried out in the following manner. First, according to the size, shape and structural characteristics of the actual smelting equipment 100, the system will construct a three-dimensional geometric model. This model includes the key parts of the smelting furnace 1012, such as the furnace body, furnace bottom, furnace wall, and the position information of key elements such as the liquid level and the top-blowing lance 1011. In order to better simulate and calculate the physical process and state changes in the smelting furnace, the physical model will be divided into a series of small grid units. Each grid unit is assigned a unique three-dimensional coordinate, so that the state changes in each unit can be accurately located and tracked. When constructing the physical model, various parameters need to be set according to the actual smelting conditions and process requirements. These parameters may include the physical properties of the material (such as density, specific heat capacity, etc.), environmental parameters such as smelting temperature and pressure, and operating parameters such as the blowing speed and angle of the top-blowing lance 1011.
[0043] In some embodiments, the parameter calculation module 1033 can be used to select a suitable preset algorithm based on different spray gun parameters to obtain real-time smelting parameters. The parameter calculation module is a key component in the intelligent prediction module 103 responsible for real-time calculation of smelting parameters. Its workflow can be divided into the following steps: Data reception: The parameter calculation module 1033 first receives real-time spray gun parameter information from the data acquisition module 102. This information includes key parameters such as the blowing speed, pressure, and temperature of the spray gun, which is an important basis for calculating real-time smelting parameters. Algorithm selection: According to different spray gun parameters and smelting conditions, the parameter calculation module will select a suitable preset algorithm for calculation. These algorithms may be numerical simulation algorithms based on physical laws, or they may be machine learning models obtained by training a large amount of historical data. Real-time calculation: Using the selected algorithm and the received real-time spray gun parameters, the parameter calculation module 1033 will perform real-time calculations to obtain the smelting parameters at the current moment. These parameters may include key information such as smelting speed, temperature distribution, and slag composition. Result output: The calculated real-time smelting parameters will be output to the post-processing module for further simulation and prediction. At the same time, these parameters will also be saved in the database for subsequent analysis and optimization.
[0044] In some embodiments, the calculation method used by the calculation model equipped with the parameter calculation module 1033 can be based on the Navier-Stokes equation, which can describe the movement of the fluid in the furnace and is composed of the momentum conservation equation, which is based on Newtonian mechanics and describes the balance relationship between the internal and external forces of the fluid. The momentum conservation equation is usually part of the Navier-Stokes equation group, which can be decomposed into three equations corresponding to the movement of the fluid in three coordinate directions. The following is the form of the momentum conservation equation in three dimensions:
[0045] For the X direction: For the Y direction: For the Z direction:
[0046] where ρ is the fluid density, u, v, w are the velocity components of the fluid in the x, y, z directions, respectively, p is the pressure, τij are the components of the stress tensor (describing the viscous effects inside the fluid), and gx, gy, gz are the gravity components, respectively.
[0047] In some embodiments, the method used by the parameter calculation module 1033 to solve the turbulence in the smelting process is to select the corresponding turbulence model for solution. Commonly used turbulence models include the Reynolds-averaged Navier-Stokes (RANS) model (including the k-ε model and the k-ω model), the Lagrangian turbulence model (large eddy simulation (LES)), the hybrid model (SST k-ω model), etc. These models can be called for calculation at will. Among them, the model commonly used for splashing in the molten pool smelting process is the Lagrangian turbulence model (large eddy simulation (LES)). The advantage of the LES model is that it numerically solves the larger-scale turbulent structure of the Navier-Stokes equations, and treats the smaller-scale turbulent structure as a model. Therefore, both the overall motion state of the molten pool and the local splashing phenomenon can be described by this model, and the computational cost of the LES model is relatively low. The basic equations of LES are obtained by spatially filtering the Navier-Stokes equations. The following is the basic equation of LES under three-dimensional flow conditions:
[0048]
[0049] in is the filtered density, is the velocity component after filtering, is the filtered pressure, is the filtered stress tensor, ρu′ i u′ j is the small-scale turbulence stress represented by the subgrid model. Here, the superscript “bar” indicates the filtering operation, and ρu′ i u′ j It is a subgrid model, which represents the modeling on the small-scale turbulent structure. The specific small eddy model is usually based on the local characteristics of the flow field, such as subgrid scale simulation, energy transfer, etc.
[0050] In some embodiments, the solver in the parameter calculation module 1033 can use a pressure-based solver (Pressure-Based Solver). The SIMPLE algorithm used by the solver is particularly suitable for calculating the motion state of the fluid in the smelting furnace. The error between the prediction results obtained by the solver and the actual parameters of the actual industrial smelting furnace can be kept below 10%, so the accuracy of the solver is considered reliable and true. In addition, the solver updates the velocity field and the pressure field by iteratively solving the continuity equation and the momentum equation. This process is mainly based on numerical discretization and iterative solution of numerical solutions. The specific steps are as follows: 1. Spatial discretization: The physical area of the smelting furnace is divided into a finite number of grid cells. In each grid cell, the partial differential equation is discretized into an algebraic equation by numerical methods, and the average value or node value of the variable on each grid is defined; 2. Time discretization: The running time is also discretized into small time steps, and the total calculation time is divided into multiple discrete time steps. In each time step, the value of the variable is updated by iterative calculation; 3. Iterative solution: For each grid cell in each time step, the discretized algebraic equation needs to be solved iteratively, where the solution gradually converges to a stable solution in each iteration step; 4. Boundary conditions: In the process of iterative solution, boundary conditions, that is, the variable values defined on the boundary of the problem, need to be considered; 5. Convergence criterion: The termination of iteration is usually judged by setting a convergence criterion. The convergence criterion can be based on the relative change of the solution, the size of the residual, etc. The iterative process stops when the convergence criterion is met; 6. Time advancement: After a time step is completed, the system enters the next time step, where the previous solution is used as the new initial condition, and the whole process is repeated until the required simulation time is reached. The iterative process is carried out in the entire spatial area and involves calculations within each grid unit. Iterative solution is an iterative approximation process. By continuously updating the variable values, the system gradually tends to a more realistic solution.
[0051] The post-processing module 1034 is used to confirm the physical three-dimensional coordinates of the required grid unit, and obtain the smelting parameters on the grid unit at any time based on the real-time smelting parameters and the spray gun parameters. The post-processing module 1034 is the final link in the intelligent prediction module 103 responsible for integrating and outputting the prediction results. Its main functions include grid positioning, data mapping, and result output and display. Grid positioning means that the post-processing module 1034 will first confirm the physical three-dimensional coordinates of the required grid unit. These grid units may correspond to different positions or areas in the smelting furnace and are the basic units for simulating and predicting state changes. Data mapping, that is, according to the received real-time smelting parameters and spray gun parameter information, the post-processing module 1034 will map these data to the corresponding grid unit. In this way, each grid unit will have a set of smelting parameter values associated with it. Result output and display, that is, finally, the post-processing module 1034 will display the mapped data in the form of graphics or tables for easy viewing and analysis by operators. At the same time, these prediction results will also be sent to other modules in the control system to guide subsequent smelting operations and adjustments.
[0052] In some embodiments, the intelligent prediction module 103 also includes an image module 1035, the main function of which is to convert the smelting parameters into image display, and to intuitively display the smelting parameters on any grid unit through different preset color scales. This graphical display method helps operators to understand the state changes in the smelting furnace more quickly and accurately, so as to make more reasonable operation decisions. The image module 1035 receives the smelting parameter data from the post-processing module 1034 of the intelligent prediction module 103, which reflects the state changes on each grid unit in the smelting furnace. By converting these data into image form, the image module 1035 realizes data visualization, allowing operators to intuitively see the state distribution and changes in the smelting furnace.
[0053] In some embodiments, in order to more clearly show the changes in smelting parameters, the image module uses different preset color scales to map different smelting parameter values. For example, higher smelting pressure may be represented by red, while lower smelting pressure may be represented by blue. Through this color mapping, the operator can quickly identify high-pressure areas and low-pressure areas in the smelting furnace. The image module 1035 may also support interactive operations, allowing the operator to view detailed smelting parameter information on different grid cells by clicking or sliding the image on the screen. This interactive operation mode improves the ease of use and flexibility of the system.
[0054] In some embodiments, the image module 1035 can also provide a processing interface, which provides the three-dimensional geometric structure of the smelting furnace and the physical quantities that need to be displayed (such as speed, pressure and other information). When it is necessary to obtain macroscopic prediction results, you only need to turn on the check box of the corresponding physical quantity to observe the real-time rendering results of the slag splashing. When it is necessary to obtain microscopic prediction results, you can select the physical quantity check box at any position and time to obtain any physical information at the local position. These physical information are conducive to clearly understanding the gas-slag two-phase movement process and the changes in multiple physical quantities in the molten pool.
[0055] In some embodiments, the control module 104 further includes an adaptive learning submodule 1041 and a dynamic optimization strategy library 1042. The control module 104 can utilize deep learning algorithms, prior knowledge, and real-time data to achieve adaptive adjustment and optimization of the operation strategy.
[0056] In some embodiments, the adaptive learning submodule 1041 can use a deep learning algorithm to construct a nonlinear mapping model between historical operating strategies and smelting parameters. This means that the submodule can learn past operating experience and corresponding smelting results, so as to understand how different operating strategies affect the smelting process. First, the adaptive learning submodule 1041 receives historical operating strategies and corresponding smelting parameters as a training set. These data are records of past operating experience and results and are crucial for training models. Then, a nonlinear mapping model between historical operating strategies and smelting parameters is constructed using a deep learning algorithm (such as a neural network). This model can capture the complex relationship between operating strategies and smelting parameters.
[0057] In some embodiments, in order to improve the convergence speed and prediction accuracy of the model, the adaptive learning submodule 1041 also sets a variety of pre-stored operation strategies in the dynamic optimization strategy library 1042 as prior knowledge. These pre-stored strategies are based on experience or previous research and can provide valuable references for the model. After the model is trained, the adaptive learning submodule 1041 inputs the real-time collected spray gun parameters and smelting parameters into the model for online prediction. According to the prediction results, the operation strategy is dynamically adjusted to achieve the best smelting effect. By comparing the feedback with the actual smelting parameters, the model parameters and operation strategies are continuously optimized. In this way, the adaptive learning submodule 1041 can continue to learn and improve to adapt to the ever-changing smelting conditions.
[0058] In some embodiments, the dynamic optimization strategy library 1042 stores a variety of pre-stored operation strategies, which can provide prior knowledge and references for the adaptive learning submodule. Pre-stored strategies may include rule-based operation strategies (such as performing fixed actions according to specific conditions), operation strategies based on model predictive control (MPC) (using models to predict future states and optimize control actions), and operation strategies obtained based on reinforcement learning algorithms (learning optimal strategies through trial and error). Rule-based strategies are fixed rule sets based on expert knowledge or experience. For example, when the smelting temperature is below a certain threshold, the ventilation volume can be increased to increase the temperature. Rule-based strategies are simple and easy to understand, but may not be flexible enough when dealing with complex or nonlinear problems. The activated strategy will be executed and its effect will be monitored in real time. If the effect is as expected, the strategy will continue to be executed; otherwise, it may be necessary to adjust the strategy or select other alternatives. The strategy based on model predictive control (MPC) is an advanced control method that uses mathematical models to predict the future behavior of the system and optimize control actions to achieve predetermined goals. During the smelting process, MPC can calculate the best operation strategy based on real-time data and prediction models, such as adjusting the position of the spray gun, ventilation volume, and feed rate. MPC-based strategies are able to handle multivariable and nonlinear problems, but require accurate mathematical models and powerful computing power. Reinforcement learning is a machine learning method that learns the optimal strategy through trial and error. During the smelting process, reinforcement learning algorithms can automatically adjust the operation strategy based on historical data and feedback signals (such as smelting efficiency, energy consumption, etc.). Reinforcement learning-based strategies are adaptive and have learning capabilities, and can handle complex and dynamically changing environments.
[0059] In some embodiments, when the control module 104 needs to adjust the operation strategy, it selects one or more suitable strategies from the dynamic optimization strategy library 1042 for activation. The selection basis may include real-time data, prediction results, production targets, etc. The activated strategy will be executed and its effect will be monitored in real time. If the effect is as expected, the strategy will continue to be executed; otherwise, it may be necessary to adjust the strategy or select other alternatives. During the execution process, the control module will evaluate the activated strategy and update the strategy in the dynamic optimization strategy library according to the evaluation results. The evaluation criteria may include smelting efficiency, energy consumption, product quality, etc. By continuously updating and optimizing the strategies in the strategy library, the control system can adapt to the ever-changing smelting conditions and needs. These pre-stored strategies can not only provide valuable reference information during model training, but also provide alternatives for the adaptive learning submodule 1041 in the online prediction and dynamic adjustment stages. When there is a large difference between the real-time collected data and the training set, the pre-stored strategies in the dynamic optimization strategy library 1042 may be activated to deal with abnormal situations.
[0060] In some embodiments, the control module 104 may further include a collaborative optimization submodule 1043, which is a part of the control module 104, and its main function is to achieve collaborative optimization between different operation strategies in the smelting process to achieve more efficient, energy-saving and high-quality production goals. It constructs a multi-objective optimization problem by comprehensively considering the real-time smelting parameters, production target requirements and the influence of various operation strategies, and applies a multi-objective optimization algorithm to solve it, so as to obtain the optimal combination of operation strategies.
[0061] In some embodiments, the working principle of the collaborative optimization submodule 1043 can be divided into the following steps: First, the collaborative optimization submodule 1043 receives the optimized operation strategy as input from the adaptive learning submodule 1041 or other control module 104. Then, it constructs a multi-objective optimization problem based on the smelting parameters (such as smelting temperature, pressure, speed, etc.), spray gun parameters (such as gas flow, pressure, temperature, etc.) and production target requirements (such as maximizing smelting efficiency, minimizing energy consumption or improving product quality) obtained in real time. In this problem, multiple operation strategies are used as decision variables and production targets are used as optimization targets. In order to solve the multi-objective optimization problem, the collaborative optimization submodule adopts a multi-objective optimization algorithm, such as a genetic algorithm, a particle swarm optimization algorithm or an ant colony algorithm. These algorithms use iterative search to find the optimal combination of operation strategies that can simultaneously meet multiple production target requirements. Finally, the collaborative optimization submodule 1043 feeds back the obtained optimal operation strategy combination to the control module 104. The control module 104 adjusts the working state of the top-blowing lance 1011 and the smelting furnace 1012 according to these optimized strategies to achieve collaborative optimization control of the smelting process. The collaborative optimization submodule 1043 works closely with other control modules 104 and the adaptive learning submodule 1041 to jointly achieve intelligent control of the smelting process. It can receive the optimized operation strategy from the adaptive learning submodule 1041 as input, and perform further collaborative optimization according to real-time data and production target requirements. At the same time, the optimization results of the collaborative optimization submodule 1043 can also be fed back to the adaptive learning submodule 1041 as a basis for its continuous optimization and learning.
[0062] In some embodiments, the system further includes a fault warning module 105, which is used to analyze potential fault risks based on the prediction results, and timely issue a warning signal or control the top-blowing lance 1011 to adjust the operation strategy. The main function of the fault warning module 105 is to analyze potential fault risks based on the prediction results obtained by the intelligent prediction module 103, and to timely issue a warning signal or control the top-blowing lance 1011 to adjust the operation strategy when a potential fault is detected, so as to avoid or reduce the impact of the fault on the smelting process.
[0063] In some embodiments, the fault warning module 105 may perform the following operations: The fault warning module 105 first receives the prediction results from the intelligent prediction module 103. These prediction results characterize the distribution and changes of smelting parameters in the molten pool of the smelting furnace, including smelting pressure, smelting speed, slag distribution, etc. Secondly, based on the received prediction results, the fault warning module uses a preset algorithm or model to analyze potential fault risks. For example, it may monitor whether the smelting pressure increases or decreases abnormally, whether the smelting speed changes suddenly, whether the slag distribution is uniform, etc. If the fault warning module 105 detects a potential fault risk, it will issue a warning signal in time. These warning signals can be sounds, lights, text messages, or other forms of notifications so that operators or automated systems can respond quickly. In addition to issuing warning signals, the fault warning module can also directly control the top-blowing lance 1011 to adjust the operating strategy to mitigate or avoid potential fault effects. For example, it may automatically adjust the gas flow, pressure, or temperature, or change the gun height, ventilation, and feed volume of the top-blowing lance 1011. The fault warning module 105 works closely with the intelligent prediction module 103 and the control module 104. It receives prediction results from the intelligent prediction module 103 as input and analyzes potential failure risks based on these results. When a potential failure is detected, it can send a signal to the control module 104 to trigger the corresponding operation strategy adjustment.
[0064] In some embodiments, the data acquisition module 102 also includes at least one pressure sensor for detecting pressure, at least one flow sensor for detecting gas flow, and at least one temperature sensor for detecting temperature. These sensors not only provide real-time spray gun parameter data, but also provide a basis for subsequent intelligent prediction and control. The intelligent prediction module 103 can use these data to predict the distribution and changes of smelting parameters in the smelting furnace 1012, and the control module 104 can adjust the operation strategy of the top-blowing spray gun 1011 according to these prediction results to achieve a more optimized and stable smelting process. The data acquisition module 102 works closely with the intelligent prediction module 103 and the control module 104. It transmits the collected spray gun parameter data to the intelligent prediction module 103 in real time, and the latter uses these data for prediction and analysis. At the same time, the control module 104 can also obtain real-time spray gun parameter data from the data acquisition module 102 as needed to support its decision-making and control functions.
[0065] The above-mentioned embodiments of the present disclosure have the following beneficial effects: by collecting multiple spray gun parameters of the top-blowing spray gun in real time, and combining the intelligent prediction module to predict the smelting parameters of the molten pool in the smelting furnace, the system can accurately adjust the operation strategy of the spray gun, including spray gun parameters, gun height, ventilation volume and feed volume, etc., so as to ensure that the smelting process is carried out in the best state, greatly improving the smelting efficiency and stability. The adaptive learning submodule in the control module can continuously learn and optimize the operation strategy, continuously adjust the model parameters and operation strategy through feedback comparison with the actual smelting parameters, and apply the multi-objective optimization algorithm to obtain the operation strategy combination to achieve the coordinated optimization control of the top-blowing spray gun and the smelting furnace, further improving the overall performance of the smelting process.
[0066] Figure 2 A schematic flow chart of some embodiments of the intelligent control method for a top-blown smelting furnace according to the present disclosure is shown.
[0067] like Figure 2 As shown, the top-blown smelting furnace intelligent control method 200 includes the following steps:
[0068] Step 201: Collect the spray gun parameters of the top-blowing spray gun, which at least include gas flow, pressure and temperature information. This is the first step of the control method, which involves collecting various spray gun parameters of the top-blowing spray gun. These parameters at least include gas flow, pressure and temperature information. These parameters are crucial for understanding the working status of the spray gun and the subsequent smelting process control.
[0069] Step 202: Receive the spray gun parameters and obtain prediction results through a preset algorithm, wherein the prediction results characterize the distribution and changes of smelting parameters in the molten pool of the smelting furnace, wherein the above-mentioned smelting parameters at least include smelting pressure, smelting speed, and slag distribution. After the spray gun parameters are collected, these data are transmitted to an intelligent prediction module. The module uses a preset algorithm to process these data and generate prediction results. These prediction results characterize the distribution and changes of smelting parameters (such as smelting pressure, smelting speed, slag distribution, etc.) in the molten pool of the smelting furnace.
[0070] Step 203: Based on the prediction results, the top-blowing lance is controlled to adjust the operation strategy, wherein the operation strategy includes the following steps: adjusting the lance parameters to meet the preset working conditions based on the prediction results; controlling the gun height of the top-blowing lance based on the prediction results; and controlling the ventilation and feeding amount of the top-blowing lance based on the prediction results. According to the prediction results provided by the intelligent prediction module, the control module generates corresponding operation strategies to adjust the working state of the top-blowing lance. These operation strategies include: adjusting the lance parameters: to meet the preset working conditions. For example, if the prediction results show that the smelting speed is too slow, the control module may increase the gas flow rate or increase the temperature to accelerate the smelting process. Controlling the gun height of the top-blowing lance: The adjustment of the gun height can affect the position and angle of the gas injection into the molten pool, thereby affecting the smelting efficiency and the formation of slag. Controlling the ventilation and feeding amount of the top-blowing lance: The adjustment of the ventilation and feeding amount can directly affect the oxygen and raw material supply in the smelting furnace, which is an important means of controlling the smelting process.
[0071] It is understood that the steps described in the smelting furnace control method 200 are similar to those described in the reference Figure 1 Therefore, the operation, features and beneficial effects of the smelting furnace control system described above are also applicable to the smelting furnace control method 200, and will not be repeated here.
[0072] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present disclosure. The electronic devices in some embodiments of the present disclosure may include but are not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.
[0073] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.
[0074] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.
[0075] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.
[0076] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.
[0077] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.
[0078] The computer-readable medium may be included in the electronic device; or it may exist independently without being assembled into the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: collects the spray gun parameters of the top-blowing spray gun, and the spray gun parameters at least include gas flow, pressure and temperature information; receives the spray gun parameters and obtains prediction results through a preset algorithm, and the prediction results characterize the distribution and changes of smelting parameters in the molten pool of the smelting furnace, and the smelting parameters at least include smelting pressure, smelting speed, and slag distribution; controls the top-blowing spray gun adjustment operation strategy based on the prediction results, wherein the operation strategy includes the following steps: adjust the spray gun parameters based on the prediction results to meet the preset working conditions; control the gun position height of the top-blowing spray gun based on the prediction results; and control the ventilation and feeding amount of the top-blowing spray gun based on the prediction results.
[0079] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0080] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0081] The units described in some embodiments of the present disclosure may be implemented in software or in hardware. The units described may also be provided in a processor, and the functions described above herein may be performed at least in part by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0082] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.
Claims
1. An intelligent control system for a top-blown smelting furnace, comprising: Smelting equipment, including top-blowing lances and smelting furnaces; A data acquisition module, used for acquiring various spray gun parameters of the top-blowing spray gun, wherein the spray gun parameters at least include gas flow, pressure and temperature information; An intelligent prediction module, used to receive the spray gun parameters and obtain prediction results through a preset algorithm, wherein the prediction results characterize the distribution and changes of smelting parameters in the molten pool of the smelting furnace, wherein the smelting parameters at least include smelting pressure, smelting speed, and slag distribution; A control module is used to control the top-blowing lance to adjust an operation strategy based on the prediction result, wherein the operation strategy includes the following steps: Adjusting the spray gun parameters based on the prediction results to meet preset working conditions; Controlling the gun height of the top-blowing spray gun based on the prediction result; Controlling the ventilation volume and feed volume of the top-blowing lance based on the prediction result; Wherein, the intelligent prediction module includes: The data iteration module is used to perform iterative training of the mathematical model according to the spray gun parameters. Configured to do the following: The spray gun parameters collected in real time are used as initial conditions, and historical spray gun parameters are set as training samples; Inputting the initial conditions and the historical spray gun parameters into a preconfigured data iteration model, wherein the data iteration model includes: a full convolution module, a sequence of residual convolution modules, and a pooling module; Select training samples from the training sample set and perform the following training steps: Input the training samples into the full convolution module included in the initial data iteration model to obtain full convolution information; Input the full convolution information into the residual convolution module sequence included in the initial data iteration model to obtain residual convolution information; Input the residual convolution information into the pooling module included in the initial data iteration model to obtain the initial spray gun parameter information; Based on a preset loss function, determining an abnormal difference value between the initial spray gun parameter feature information and the sample spray gun parameter information included in the training sample; In response to determining that the abnormal difference value is less than the target value, determining the initial data iteration model as the data iteration model; and / or In response to determining that the abnormal difference value is greater than or equal to the target value, adjusting relevant parameters in the initial data iteration model, determining the adjusted initial data iteration model as the initial data iteration model, and selecting training samples from the training sample set for performing the training step again; Wherein, the intelligent prediction module also includes: A physical model module is used to construct a physical three-dimensional coordinate model based on a smelting device in actual use, wherein the physical three-dimensional coordinate model represents at least the structure and liquid level information of the smelting furnace and the gun position information of the top-blowing lance, and the physical three-dimensional coordinate model is gridded, and each grid unit represents a three-dimensional coordinate; A parameter calculation module, used for selecting a suitable preset algorithm based on different spray gun parameters to obtain real-time smelting parameters; The post-processing module is used to confirm the physical three-dimensional coordinates of the required grid unit, and obtain the smelting parameters on the grid unit at any time based on the real-time smelting parameters and the spray gun parameters.
2. The intelligent control system for a top-blown smelting furnace according to claim 1, wherein: The intelligent prediction module also includes: The image module converts the smelting parameters into an image display, and displays the smelting parameters on any grid unit according to different preset color levels.
3. The intelligent control system of the top-blown smelting furnace according to claim 1, wherein: The control module comprises: Adaptive learning submodule and dynamic optimization strategy library, the adaptive learning submodule is configured as follows: Receiving historical operation strategies and corresponding smelting parameters as a training set, and using a deep learning algorithm to construct a nonlinear mapping model between the historical operation strategies and the smelting parameters; Setting a plurality of pre-stored operation strategies in a dynamic optimization strategy library as prior knowledge to improve the convergence speed and prediction accuracy of the nonlinear mapping model; Inputting the spray gun parameters and smelting parameters collected in real time into the trained nonlinear mapping model for online prediction, and dynamically adjusting the operation strategy according to the prediction results; By comparing the feedback with the actual smelting parameters, the model parameters and the operation strategy are continuously optimized to achieve adaptive learning and continuous optimization of the control module; Among them, the multiple pre-stored control strategies in the dynamic optimization strategy library include rule-based operation strategies, model predictive control-based operation strategies, and operation strategies obtained based on reinforcement learning algorithms.
4. The intelligent control system for a top-blown smelting furnace according to claim 3, wherein: The control module further includes a collaborative optimization submodule, which is configured as follows: Receiving the optimized operation strategy from the adaptive learning submodule; Based on the smelting parameters and lance parameters acquired in real time and production target requirements, a multi-objective optimization problem is constructed, wherein the production target requirements include maximizing smelting efficiency, minimizing energy consumption or improving product quality; Applying a multi-objective optimization algorithm to obtain an operation strategy combination, wherein the multi-objective optimization algorithm includes a genetic algorithm, a particle swarm optimization algorithm or an ant colony algorithm; The operation strategy combination is fed back to the control module to achieve coordinated optimization control of the top-blowing lance and the smelting furnace.
5. The intelligent control system of the top-blown smelting furnace according to claim 1, wherein: The system further comprises: A fault warning module is used to analyze potential fault risks based on the prediction results, and to issue a warning signal in time or control the top-blowing spray gun to adjust the operation strategy.
6. The intelligent control system for a top-blown smelting furnace according to claim 1, wherein: The data acquisition module also includes: at least one pressure sensor for detecting pressure; at least one flow sensor for detecting gas flow; At least one temperature sensor is provided for detecting a temperature.
7. A top-blown smelting furnace intelligent control method, comprising: Collecting the spray gun parameters of the top-blowing spray gun, wherein the spray gun parameters at least include gas flow, pressure and temperature information; Receiving the spray gun parameters and obtaining prediction results through a preset algorithm, wherein the prediction results characterize the distribution and changes of smelting parameters in the molten pool of the smelting furnace, wherein the smelting parameters at least include smelting pressure, smelting speed, and slag distribution; Controlling the top-blowing lance to adjust the operation strategy based on the prediction result, wherein the operation strategy includes the following steps: adjusting the lance parameters based on the prediction result to meet the preset working conditions; Controlling the gun height of the top-blowing spray gun based on the prediction result; Controlling the ventilation volume and feed volume of the top-blowing lance based on the prediction result; According to the spray gun parameters, mathematical model iterative training is performed. Configured to do the following: The spray gun parameters collected in real time are used as initial conditions, and historical spray gun parameters are set as training samples; Inputting the initial conditions and the historical spray gun parameters into a preconfigured data iteration model, wherein the data iteration model includes: a full convolution module, a sequence of residual convolution modules, and a pooling module; Select training samples from the training sample set and perform the following training steps: Input the training samples into the full convolution module included in the initial data iteration model to obtain full convolution information; Input the full convolution information into the residual convolution module sequence included in the initial data iteration model to obtain residual convolution information; Input the residual convolution information into the pooling module included in the initial data iteration model to obtain the initial spray gun parameter information; Based on a preset loss function, determining an abnormal difference value between the initial spray gun parameter feature information and the sample spray gun parameter information included in the training sample; In response to determining that the abnormal difference value is less than the target value, determining the initial data iteration model as the data iteration model; and / or In response to determining that the abnormal difference value is greater than or equal to the target value, adjusting relevant parameters in the initial data iteration model, determining the adjusted initial data iteration model as the initial data iteration model, and selecting training samples from the training sample set for performing the training step again; A physical three-dimensional coordinate model is constructed based on a smelting device in actual use, wherein the physical three-dimensional coordinate model represents at least the structure and liquid level information of the smelting furnace and the gun position information of the top-blowing lance, and the physical three-dimensional coordinate model is gridded, wherein each grid unit represents a three-dimensional coordinate; Selecting a suitable preset algorithm based on different spray gun parameters to obtain real-time smelting parameters; The physical three-dimensional coordinates of the required grid unit are confirmed, and based on the real-time smelting parameters and the spray gun parameters, the smelting parameters on the grid unit at any time are obtained.
8. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to claim 7 is implemented.
Citation Information
Cited By
Efficient molten pool smelting method and system for low-grade tin raw materials
CN120350244A
