A multi-zone stirring intelligent collaborative control system and method for a side-blown molten bath smelting furnace

Abstract

The present application relates to non-ferrous pyrometallurgical process and core equipment technical field, especially to a kind of side-blown bath smelting furnace multi-zone stirring intelligent collaborative control system and method, method includes at the beginning of each control cycle, multi-zone stirring collaborative control unit triggers data acquisition instruction, synchronously obtains fluid dynamics signal, structural response signal and thermodynamics and process parameter etc. smelting furnace bottom layer operating state data;Identify standing wave risk, assess scour risk and stirring uniformity quantification etc. whole pool state index;State machine dispatcher determines the dominant mode of current control cycle based on whole pool state index according to preset priority logic, and calling response model predictive control algorithm solves the actuator set value of next time, obtains the digital optimal set value;The digital optimal set value obtained is converted into physical control signal, and is issued in parallel to bottom layer actuator;Flow regulation, flow pattern transformation and pulse establishment etc. execution action.

Description

Technical Field

[0001] This invention relates to the field of non-ferrous metal pyrometallurgical processes and core equipment technology, and in particular to a multi-zone stirring intelligent collaborative control system and method for a side-blown molten pool furnace. Background Technology

[0002] Side-blown smelting technology has been widely used in the smelting of heavy non-ferrous metals (copper, lead, nickel, etc.) due to its advantages such as compact equipment structure, high thermal efficiency, and strong adaptability to raw materials. Traditional side-blown furnaces typically have one or more rows of tuyeres (lances) arranged linearly, at equal intervals, and symmetrically along both sides of the furnace length. Oxygen-enriched air and fuel are injected into the molten pool at high speed through these tuyeres, and the melt is stirred by bubble buoyancy and jet entrainment to complete oxidation, reduction, or matte-forming reactions. However, as smelting equipment develops towards ultra-large size and high strength, the traditional linear lance arrangement has gradually revealed a series of irreconcilable contradictions in terms of fluid mechanics and reaction engineering, mainly in the following aspects: limited gas-slag contact area and bubble coalescence effect. In the traditional linear close-packed layout, the bubble plumes generated by adjacent tuyeres are prone to lateral interference and coalescence during their ascent. Studies have shown that when the spacing between adjacent jets is too small, the bubble clusters will merge to form large bubbles or gas cavities, resulting in a sharp decrease in the gas-liquid surface area, which significantly reduces oxygen utilization and reaction rate.

[0003] Dead zones and over-mixing coexist in molten pool stirring. The linear layout results in a two-dimensional dominant flow field within the furnace, with stirring energy concentrated in a narrow region near the tuyere axis. Stagnation zones or dead zones often form in the gaps between the two tuyeres and in the deeper regions along the furnace centerline, leading to uneven material mixing and localized condensation. In the area directly opposite the tuyeres, excessive turbulence causes unnecessary energy dissipation. Severe furnace wall erosion and the destructive nature of asymmetric fluctuations mean that, in pursuit of better stirring, traditional operations often simply increase the injection pressure or velocity. This results in strong gas-liquid two-phase jets directly impacting the opposite furnace lining, causing severe mechanical erosion and thermal spalling. Furthermore, linear lance placement easily induces single-frequency standing wave oscillations (Sloshing) on ​​the molten pool surface. These violent surface fluctuations exacerbate periodic thermal shock damage to the refractory material at the slag line, and may even lead to turbidity runoff or furnace structural vibration. The lack of proactive control methods based on flow field morphology means that existing side-blown furnace control systems are mostly setpoint controls, meaning all tuyeres are set with the same flow rate and pressure. This one-size-fits-all approach cannot cope with the complex unsteady flow field changes within the furnace. It cannot dynamically adjust the local stirring intensity based on spatial differences in melt level, viscosity, and reaction progress, resulting in poor stirring consistency across the entire furnace and hindering the achievement of refined metallurgy. In summary, there is an urgent need for a systematic technical solution that breaks free from the constraints of traditional linear layouts and addresses the contradiction between uneven stirring and excessive scouring from a fundamental fluid dynamics perspective through nonlinear topology optimization of tuyer spatial positions and multivariate collaborative control.

[0004] Prior art 1, Chinese patent application number: 202511474566.2, relates to the field of recycled metal smelting technology, specifically a method and system for improving the uniformity of metal smelting and recycling based on electromagnetic stirring. The method includes: collecting temperature, density, and flow field data during the smelting process; analyzing a multi-physics coupling model using finite element method to identify non-uniform regions within the furnace; formulating a zoned heating strategy, adjusting the heating power and timing of each zone to achieve uniform temperature field distribution; driving the molten metal to circulate and eliminate density stratification by adjusting electromagnetic stirring parameters; and uniformly arranging nozzles at the bottom of the smelting furnace to pulse-inject inert gas to achieve homogenization of the molten metal at the bottom of the furnace. Although applying mechanical vibration promotes the migration of impurities and bubbles, further improving the uniformity of metal mixing and significantly enhancing the quality of recycled metal, the problem of uneven stirring in large side-blown furnaces remains.

[0005] Prior art two, Chinese patent application number: 202510756409.4, provides an aluminum alloy comprising, by mass percentage: Zn 4.0-5.0%, Mg 1.2-1.8%, Cu 0.1-0.3%, Zr 0.08-0.12%, Ti 0.02-0.06%, Fe ≤ 0.25%, with the balance being Al. The alloying elements containing Zn 4.0-5.0%, Mg 1.2-1.8%, Cu 0.1-0.3%, Zr 0.08-0.12%, Ti 0.02-0.06%, Fe ≤ 0.25% are added to a smelting furnace along with pure aluminum at 700-700°C. The process involves melting at 50℃ to form a melt, followed by refining with inert gas or refining agents to remove gases and inclusions. This invention utilizes the different speeds of the upper and lower rolls of a differential speed rolling mill to generate shear force, breaking down coarse structures, refining grains, and increasing dislocation density. Rolling at 480-500℃ reduces deformation resistance, promotes dynamic recrystallization, and maintains processing performance. A 60-70% total deformation further enhances performance. Post-rolling multi-directional forging repeatedly changes the strain direction, breaking down grains, homogenizing the structure, and eliminating anisotropy. While this improves the overall performance and application range of aluminum alloy profiles and solves the problem of insufficient microstructure refinement and dislocation density increase during rolling, it also presents the problem of dead zone crusting.

[0006] Prior art three, Chinese patent application number: 202210085237.9, provides a method for the co-processing of hazardous waste containing metal and cyanide tailings, relating to the field of ecological environment governance technology. This method specifically includes the following steps: S1. Unpacking and drying the hazardous waste containing metal and residual anodes from electrolytic aluminum; the cyanide tailings are stored in bulk; S2. Mixing the dry bases of the hazardous waste containing metal, cyanide tailings, and residual anodes from electrolytic aluminum in a specific ratio to obtain a homogenized material that meets process requirements; S3. Introducing the homogenized material into an oxygen-enriched side-blown melting furnace for oxidation-reduction smelting; S4. Purifying the metal-containing fumes generated during the smelting process. This proposed method can improve the metal recovery rate and achieve harmless disposal. Unrecovered metals exist in the vitrified water fragments, posing no risk of leaching pollution, and can be used as building materials or for sandblasting. Furthermore, the smelting process of this method does not require the addition of flux, saving costs; however, it suffers from severe localized erosion of the furnace wall.

[0007] Currently, existing technologies 1, 2, and 3 suffer from problems such as uneven stirring, dead zone crusting, and severe local erosion of the furnace wall in large side-blown furnaces. To solve these problems, this invention provides an intelligent collaborative control system and method for multi-zone stirring in a side-blown molten pool furnace. Summary of the Invention

[0008] The main objective of this invention is to provide a multi-zone stirring intelligent collaborative control system and method for side-blown molten pool furnaces, in order to solve the problems of uneven stirring, dead zone crusting, and severe local erosion of the furnace wall in existing large side-blown furnaces. This invention relates to a side-blown molten pool furnace for smelting heavy non-ferrous metals such as copper, nickel, and lead. In particular, it relates to a comprehensive technical solution that maximizes the gas-slag contact area, improves the consistency of stirring in the molten pool throughout the furnace, and minimizes the erosion stress of the furnace lining by using a non-uniform, nonlinear spatial topology layout of the side-blown tuyeres (spray guns) combined with online reconstruction of the gas-slag multiphase flow field and multi-zone stirring collaborative control.

[0009] To achieve the above objectives, the present invention provides the following technical solution: A side-blown molten pool furnace multi-zone stirring intelligent collaborative control system, the side-blown molten pool furnace multi-zone stirring intelligent collaborative control system comprising: The multi-dimensional data acquisition and preprocessing component is used at the beginning of each control cycle when the multi-zone stirring and co-control unit triggers the data acquisition command to synchronously acquire fluid dynamics signals, structural response signals, and thermodynamic and process parameters, as well as other data on the bottom operating status of the melting furnace. The risk boundary assessment component is used to identify standing wave risk, assess scouring risk, and measure the overall pool status indicators by utilizing the bottom-level operating status data of the smelting furnace and combining it with the built-in price reduction molten pool mechanism model. The optimal control quantity solution component is used by the state machine scheduler to determine the dominant mode of the current control cycle based on the preset priority logic and the full pool state index, and to call the response model predictive control algorithm to solve for the actuator setpoint at the next moment, thus obtaining the digital optimal setpoint. The collaborative instruction issuance and execution component is used to convert the obtained digital optimal setpoint into physical control signals and issue them in parallel to the underlying actuators; to perform flow regulation, flow pattern transformation and pulse establishment execution actions; after completing the instruction issuance, it enters a waiting state until the next control cycle.

[0010] As a further improvement of the present invention, the multidimensional data acquisition and preprocessing component includes: The module for extracting pulsation feature values ​​is used to... The high sampling rate reads the pressure transmitter signals of each air outlet branch, and the DC component is filtered out by a high-pass filter to extract the pulsating characteristic values ​​that characterize the frequency and intensity of bubble detachment. The real-time power spectral density map generation module is used to synchronously acquire the acoustic vibration monitoring array signals installed in key parts of the furnace shell, perform fast Fourier transform, and generate a real-time power spectral density map. The module for obtaining the bottom operating status data of the smelting furnace is used to read the furnace wall temperature field matrix fed back by the embedded thermocouple, as well as the current actual flow feedback value and valve opening status of each mass flow controller.

[0011] As a further improvement of the present invention, the risk boundary assessment component includes: The module for the price-reducing molten pool mechanism model is used to construct a price-reducing molten pool mechanism model using the bottom operating status data of the smelting furnace; it also constructs a nonlinear lance design topology function model to maximize the gas-slag contact area and homogenize the turbulent dissipation rate; it establishes a molten pool state monitoring and evaluation index model based on tuyer parameters; and it constructs a furnace scour risk assessment model to evaluate the scour intensity based on the relationship between jet penetration depth and furnace width. The whole pool status index module is used to derive the whole pool status index from the price reduction molten pool mechanism model; compare the amplitude of the low-frequency main peak in the vibration power spectrum with the preset safety threshold to identify standing wave risk; and calculate the furnace wall scour risk index of each tuyer area based on the real-time furnace wall temperature and its rate of change, combined with the local heat flow coupling model. The module for calculating the overall pool mixing consistency index is used to analyze the variance consistency of the pressure pulsation signals in each zone and calculate the overall pool mixing consistency index. When the mixing consistency index is close to the preset value, it indicates that the kinetic energy distribution of the molten pool is more uniform.

[0012] As a further improvement of the present invention, the optimal control quantity solving component includes: The priority division module is used by the state machine scheduler to divide the system into first priority, second priority and third priority according to the preset priority logic; the weight coefficient is dynamically adjusted according to the acoustic vibration signal and wall temperature status, and when a sudden vibration alarm is detected, the weight coefficient is instantly assigned a maximum value. The model prediction module is used to divide the furnace body into... Each logical control zone employs a model predictive control strategy. Based on the model predictive control framework, the actuators are driven to switch between different operating modes by real-time observation of the entire pool state. The adjustment path module is used to switch the standing wave oscillation control mode, identify abnormal features, and execute asymmetric pulse control logic; switch the furnace wall scour protection mode, provide scour risk warning, and execute swirl divergence adjustment strategy; and switch the stirring consistency improvement mode, perform logical judgment indicators, and execute the adjustment path.

[0013] As a further improvement of the present invention, the model prediction module includes: The first optimal setpoint formation submodule is used by the multi-zone stirring and co-control unit to perform a fast Fourier transform on the vibration spectrum, extract the energy peak value in the 0.2Hz to 1.5Hz frequency band, and compare it with the preset safety threshold. If the energy peak value exceeds the limit and has frequency consistency, the system is determined to be in the pre-standing wave oscillation region, triggering the wave suppression mode. The multi-zone stirring and co-control unit calculates the natural frequency of the standing wave based on the furnace body characteristic dimensions and real-time liquid level, and sets the asymmetric pulse jet parameters: the pulse frequency and 180° phase difference of the left and right duct arrays, and redistributes the air supply intensity of each duct according to the pulse timing. It generates digital optimal setpoints, including the pulse frequency setpoint, phase difference setpoint, and flow timing allocation value in each pulse cycle for each duct. The second optimal setpoint generation submodule is used to collect pressure transmitter signals and furnace wall thermocouple temperature field data of each tuyere branch in real time. The multi-zone stirring and co-control unit calculates the ratio of the jet penetration depth of each tuyere to the half-width of the furnace based on the jet penetration depth model, and evaluates the local furnace wall scour risk index based on the kinetic energy decay model, while monitoring the wall temperature change rate of the corresponding area. If the local furnace wall scour risk index is close to or exceeds the warning value and the wall temperature rises abnormally, it is determined that there is a high scour risk and the furnace protection logic is triggered. For high-risk tuyeres, the multi-zone stirring and co-control unit increases the setpoint of the swirl number through the spray gun swirl adjustment mechanism without changing the total oxygen content, so that the jet expansion angle increases and the kinetic energy is dissipated in the middle of the molten pool, while keeping the flow rate setpoint of the tuyere unchanged. It generates digital optimal setpoints, including the opening setpoint of the swirl adjustment mechanism of the high-risk tuyere and the command value of the corresponding flow controller to maintain the original flow rate. The third optimal setpoint formation submodule is used to collect back pressure pulsation signals from each vent in real time by the multi-zone stirring and co-control unit, and receive macroscopic process parameters from the DCS. It calculates the overall pool stirring consistency index by inverting local stirring power density through back pressure pulsation, and monitors the pressure fluctuation variance of each multi-zone stirring and co-control unit, comparing the local stirring power density of each zone with the overall pool average. If the local stirring power density of a certain multi-zone stirring and co-control unit is consistently lower than 60% of the overall pool average and the pressure fluctuation variance is consistently lower than the threshold, it is determined that there is a risk of flow stagnation or cold material crusting in that zone, triggering the homogenization mode. With maximizing the overall pool stirring consistency index as the optimization objective, a model predictive control algorithm is used to solve the flow increment distribution of each vent, focusing on increasing the oxygen distribution ratio of the target area and its adjacent nonlinear misaligned vents, inducing secondary circulation to activate the stagnation zone. A digital optimal setpoint is generated, including the flow setpoint increment of each vent and the corresponding mass flow controller opening command.

[0014] As a further improvement of the present invention, the third optimal setting value forming submodule includes: The feature extraction unit is used to input the high-frequency back pressure pulsation signal collected by the pressure transmitter of each vent branch into the multi-zone stirring and co-control unit; the multi-zone stirring and co-control unit reads the signal at a sampling rate of not less than 1 kHz, filters out the steady-state DC component through the built-in high-pass filter, extracts the pulsation main frequency and amplitude generated during the bubble detachment process of each vent, and forms the bubble detachment frequency and pulsation intensity characteristic value of each vent. The inversion processing unit is used to input the bubble detachment frequency and pulsation intensity characteristic values ​​of each tuyere into the flow state inversion program built into the risk boundary assessment component. Using empirical correlations calibrated in advance through cold-state experiments or computational fluid dynamics simulations, the local gas holdup and equivalent bubble diameter of the molten pool region where each tuyere is located are inverted. Then, combined with the real-time flow rate of the tuyere, the tuyere burial depth and the molten pool volume of the corresponding logic control area, the local stirring power density of the region is estimated based on the kinetic energy dissipation theory to form the local stirring power density value of each logic control area. The reciprocal operation unit is used to summarize the local stirring power density values ​​of each logic control zone to the multi-zone stirring collaborative control unit. The multi-zone stirring collaborative control unit calculates the average value of the stirring power density of the whole pool and solves the root mean square of the sum of squares of the relative deviations of the power density of each control zone relative to the average value. Through the normalized variance reciprocal operation, the stirring consistency index of the whole pool between 0 and 1 is obtained, which is used to quantify the uniformity of the kinetic energy distribution of the molten pool.

[0015] As a further improvement of the present invention, the inversion processing unit includes: The power estimation subunit is used by the multi-zone stirring and co-control unit to estimate the mechanical work released by the gas as it expands from the nozzle to the surface of the melt, based on thermodynamic principles and real-time pressure and temperature parameters in the molten pool, and to obtain the power value of gas expansion at each nozzle. The kinetic energy estimation subunit is used by the multi-zone stirring and co-control unit to estimate the effective kinetic energy transferred to the surrounding melt by the high-speed jet during the process of breaking into bubbles, based on the jet dynamics theory, and to obtain the jet kinetic energy transfer power value of each vent. The accumulation processing subunit is used to accumulate the power value of gas expansion and the power value of jet kinetic energy transfer at each vent within the same logic control area. The total power value after accumulation is divided by the preset molten pool volume in the logic control area to obtain the local stirring power density value of the logic control area.

[0016] As a further improvement of the present invention, the accumulation processing subunit includes: The classification component is used to classify the gas expansion power value and jet kinetic energy transfer power value of each vent according to the preset logical control area classification rules; the multi-zone stirring and co-control unit matches each vent to the corresponding logical control area based on the spatial coordinate range of the feeding area, main reaction area and settling area defined during the nonlinear gun design, forming a list of vents belonging to each logical control area and its corresponding two types of power value datasets. The power accumulation component is used by the multi-zone stirring collaborative control unit to traverse the list of air vents belonging to each logical control zone. For all air vents in the same logical control zone, the power value of gas expansion work done by each air vent in the zone is accumulated together to obtain the total power value of gas expansion work done by the logical control zone. At the same time, the power value of jet kinetic energy transfer from each air vent in the zone is accumulated together to obtain the total power value of jet kinetic energy transfer by the logical control zone. The total power acquisition component is used by the multi-zone stirring and co-control unit to add the total work power value of gas expansion in the same logical control zone to the total transfer power value of jet kinetic energy, and obtain the total input mechanical power value of the logical control zone.

[0017] As a further improvement of the present invention, the total power acquisition component includes: The consistency comparison sub-component is used by the multi-zone stirring collaborative control unit to extract two power values ​​corresponding to the same logical control zone from the total power dataset of gas expansion and the total power dataset of jet kinetic energy transfer of the power accumulation component; and to perform consistency comparison based on the logical control zone identifier and timestamp attached to the two power values ​​to confirm that they belong to the same control zone and are in the same control cycle, thus forming a power data pair that has been paired in the logical control zone. The numerical merging sub-component is used by the multi-zone stirring and collaborative control unit to input the paired gas expansion total work power value and the jet kinetic energy transfer total power value into the internal numerical merging module, perform linear superposition calculation of the power values, and obtain the preliminary accumulation result of the input mechanical power of the logic control area. The numerical verification sub-component is used by the multi-zone stirring collaborative control unit to verify the unit dimension of the preliminary accumulation result, confirm that it is consistent with the preset power unit standard, and store the verified value in the total input mechanical power cache of the corresponding logic control area to form the total input mechanical power value for calculating the local stirring power density.

[0018] To achieve the above objectives, the present invention also provides the following technical solution: A method for intelligent and coordinated control of multi-zone stirring in a side-blown molten pool furnace is provided, which is applied to the aforementioned intelligent and coordinated control system for multi-zone stirring in a side-blown molten pool furnace. The method includes: At the beginning of each control cycle, the multi-zone stirring and co-control unit triggers a data acquisition command to simultaneously acquire fluid dynamics signals, structural response signals, and thermodynamic and process parameters of the bottom operating status data of the melting furnace. By utilizing the bottom-level operating status data of the smelting furnace and combining it with the built-in price reduction molten pool mechanism model, the system identifies standing wave risk, assesses scouring risk, and measures the overall pool status indicators for stirring uniformity. The state machine scheduler determines the dominant mode of the current control cycle based on the preset priority logic and the full pool state index, and calls the response model predictive control algorithm to solve for the actuator setpoint at the next moment, thus obtaining the digital optimal setpoint. The obtained digital optimal setpoint is converted into a physical control signal and sent to the underlying actuator in parallel; flow regulation, flow pattern transformation and pulse establishment are performed; after the instruction is issued, it enters a waiting state until the next control cycle.

[0019] To achieve the above objectives, the present invention also provides the following technical solution: An electronic device includes a processor and a memory coupled to the processor, the memory storing program instructions executable by the processor; when the processor executes the program instructions stored in the memory, it implements the intelligent collaborative control method for multi-zone stirring in a side-blown molten pool furnace as described above.

[0020] To achieve the above objectives, the present invention also provides the following technical solution: A storage medium storing program instructions, which, when executed by a processor, implement the intelligent collaborative control method for multi-zone stirring in a side-blown molten pool furnace as described above.

[0021] This invention constructs a technical system encompassing spatial topological nonlinearity, logical mixing zones, digital flow field observation, and intelligent control strategies. Its core is to utilize the nonlinear, staggered layout of spray guns to disperse the coupling effect of bubble plumes, thereby improving the overall mixing uniformity (UI) of the entire pool by leveraging the spatiotemporal difference in momentum input at each tuyer. Simultaneously, it utilizes the coherent eddy shearing dissipation of kinetic energy between jets to suppress direct scouring and damage to the furnace wall from the jets at the source. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the functional modules of an embodiment of the intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace of the present invention; Figure 2 This is a schematic diagram of the spatial topology development of the nonlinear staggered lance arrangement in the side-blown furnace of the present invention; Figure 3 This is a diagram of the architecture of the multi-zone stirring and coordinated control system of the present invention; Figure 4This is a functional block diagram of one embodiment of the multidimensional data acquisition and preprocessing component of the present invention; Figure 5 This is a functional block diagram of one embodiment of the risk boundary assessment component of the present invention; Figure 6 This is a comparison diagram of the gas-slag contact area and bubble coalescence effect in this invention; Figure 7 This is a functional block diagram of one embodiment of the optimal control quantity solving component of the present invention; Figure 8 This is a flowchart of the wave suppression control logic based on furnace wall vibration feedback in this invention. Figure 9 This is a flowchart illustrating the multi-objective collaborative closed-loop control strategy and mode switching process for the side-blown furnace of the present invention. Figure 10 This is a flowchart illustrating the steps of an embodiment of the intelligent collaborative control method for multi-zone stirring in a side-blown molten pool furnace according to the present invention. Figure 11 This is a schematic diagram of the structure of an embodiment of the electronic device of the present invention; Figure 12 This is a schematic diagram of the structure of one embodiment of the storage medium of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] The terms "first," "second," and "third" in this invention are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of those features. In the description of this invention, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this invention are only used to explain the relative positional relationships and movements between components in a specific orientation (as shown in the figures). If the specific orientation changes, the directional indication changes accordingly. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0025] References to embodiments herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0026] like Figure 1 As shown, this embodiment provides an example of a multi-zone stirring intelligent collaborative control system for a side-blown molten pool furnace. In this embodiment, the multi-zone stirring intelligent collaborative control system for a side-blown molten pool furnace includes: The multi-dimensional data acquisition and preprocessing component 1 is used at the beginning of each control cycle. The multi-zone stirring and co-control unit triggers a data acquisition command to synchronously acquire fluid dynamics signals, structural response signals, and thermodynamic and process parameters, as well as other data on the bottom operating status of the melting furnace. Risk boundary assessment component 2 is used to identify standing wave risk, assess scouring risk, and measure the overall pool status indicators such as stirring uniformity by using the bottom operating status data of the smelting furnace and combining it with the built-in price reduction molten pool mechanism model. The optimal control quantity solution component 3 is used by the state machine scheduler to determine the dominant mode of the current control cycle based on the preset priority logic and the full pool state index, and to call the response model predictive control algorithm to solve for the actuator setpoint at the next moment, thus obtaining the digital optimal setpoint. The collaborative instruction issuing and execution component 4 is used to convert the obtained digital optimal setpoint into physical control signals and send them to the underlying actuators in parallel; to perform execution actions such as flow regulation, flow pattern transformation and pulse establishment; after completing the instruction issuance, it enters a waiting state until the next control cycle.

[0027] Preferably, the furnace body includes a furnace shell, furnace lining, charging port, slag discharge port, and side-blowing tuyeres (spray guns) arranged in multiple zones. Unlike the traditional linear, evenly spaced tuyer arrangement, the tuyeres on both sides of the furnace body are no longer arranged on the same horizontal straight line, but according to a specific nonlinear function curve. The tuyeres are arranged in a non-linear topological logic according to the following principles: non-uniformity along the furnace length direction (X-axis), non-uniformity of the horizontal spacing between tuyeres, denser spacing in the feeding zone to accelerate material processing, and sparser spacing in the settling zone to facilitate separation. (Tuyer horizontal coordinates) Based on the flow characteristics of the materials inside the furnace, a sparse-density distribution is adopted. The high-reaction intensity zone below the feed inlet is densely arranged to provide higher energy density for material processing; the settling zone at the furnace tail is sparsely arranged to reduce interference and facilitate slag-metal separation. The furnace height (Z-axis) is wavy, and the tuyere positions are distributed in a wavy or stepped staggered manner. Tuyere installation height... No longer on the same horizontal line, but according to a preset damped sine function. Fluctuation. This tiered staggered design activates the molten pool in layers along the vertical dimension, suppressing the single horizontal backflow ring that is easily generated in traditional layouts, and achieving three-dimensional stirring in the depth direction of the molten pool. The staggered arrangement (Y-axis mapping) on ​​both sides, with the left and right wall vents intersecting in their X-axis projection positions, avoids direct collisions between the jets, using fluid shearing instead of wall collisions to dissipate kinetic energy. The left and right wall vent arrays are completely staggered in their axial projections. When the left jet penetrates to near the centerline of the molten pool, its extended coherent vortex structure precisely enters the gap between the two adjacent right jets, using fluid shear friction to replace the rigid impact of the jet on the wall.

[0028] like Figure 2 The figure illustrates the spatial topology of a side-blown furnace with a nonlinear staggered tuyeres arrangement. The diagram details the distribution of the tuyeres on the left and right side walls along the furnace length (X-axis) and furnace height (Z-axis) planes, showing a sinusoidal distribution pattern. The figure also shows the projection relationship of the tuyeres along the furnace width (Y-axis), demonstrating that the tuyer arrays on the left and right walls are intersecting and completely staggered in axial projection. Furthermore, this layout reflects the non-uniform arrangement of the tuyeres along the furnace length (X-axis), with a denser arrangement in the feeding zone and a sparser arrangement in the settling zone.

[0029] The MCU is the core of the system's computing power, employing an embedded real-time control system. It includes a data acquisition layer, a flow field reconstruction and state observation layer, a multi-objective optimization layer, and an instruction execution layer, achieving real-time data transmission and closed-loop control via industrial Ethernet. The MCU integrates the following core computing layers: a flow field reconstruction layer, which, based on measured pressure pulsations and vibration signals, calls a reduced-order fluid model (ROM) to reconstruct the three-dimensional kinetic energy dissipation rate within the furnace. Layout; Safety indicator calculation layer, which calculates the furnace wall erosion risk index of each tuyeres in real time. and the consistency index of mixing throughout the pool The multi-objective optimization scheduling layer, based on the model predictive control (MPC) algorithm, periodically solves for the optimal jetting vector.

[0030] The system control commands are sent to intelligent execution terminals distributed on both sides of the furnace body, including: a precision mass flow control valve (MFC) for millisecond-level adjustment of the oxygen / air ratio and total momentum of each duct; a spray gun swirl adjustment mechanism for changing the flow pattern (expansion angle) of the jet by adjusting the blade opening or core tube position inside the spray gun; and a pulse generation unit with a fast switching valve in the high-pressure gas supply line to achieve intermittent power input with controlled pulse frequency.

[0031] like Figure 3The diagram shows the architecture of the multi-zone stirring and coordinated control system, illustrating the data flow within the system. The architecture includes a sensor layer, a PLC / DCS control layer, and an upper-level optimization algorithm module. The sensor layer integrates an acoustic vibration monitoring array, an air vent pressure pulsation monitoring network, and a wall temperature integrated monitoring network to provide real-time input. The multi-zone stirring and coordinated control unit (MCU), as the core of the computing power, includes a data acquisition layer, a flow field reconstruction layer based on a reduced-order fluid model (ROM), a safety index calculation layer, and a multi-objective optimization scheduling layer based on model predictive control (MPC). Control commands are ultimately sent to the actuators, such as precision mass flow control valves, spray gun swirl adjustment mechanisms, and pulse generation units.

[0032] The core of the side-blown molten pool melting furnace described in this invention lies in achieving precise control of the kinetic energy distribution of the molten pool through the spatial coordinates of the lance defined by a nonlinear function. Based on the dynamic characteristics of the melting process, the furnace body is divided into three functional logic zones along its length (X-axis), with different nonlinear lance placement strategies employed in each zone. The high-intensity mixing zone (feed end) has the following topological characteristics: the air vents in this area employ a high-frequency sawtooth nonlinear layout. Its vertical coordinates... Based on the formula, the wave number is set. Take the upper limit value (e.g.) ), misalignment amplitude Set as .

[0033] Project Implementation: Shorten the horizontal distance between adjacent air outlets This design creates significant dislocations in the vertical dimension of the jet. The aim is to utilize the coherent vortices generated by the high-frequency dislocation jet to create a strong shear-entrainment effect, forcing the melt to generate a large longitudinal circulation, thereby rapidly entraining and melting the cold material below the feed inlet and eliminating the material dead zone.

[0034] The stable reaction zone (middle of the furnace body) features a standard damped sine wave layout. The left and right tuyeres strictly adhere to the phase angle... The mirror correspondence ensures that the jets on both sides are completely staggered in the axial projection, forming an interlocking flow field structure.

[0035] Engineering Implementation: Calculate the specific surface area of ​​the gas-slag contact using formulas. The coherent vortex interlocking effect induced by the staggered layout prevents bubbles from overlapping and coalescing during their ascent, thus maintaining a high level of gas holdup. This prolongs the residence time of oxygen-enriched air in the melt, thereby improving the efficiency of metallurgical reactions.

[0036] Slag-gold separation / depletion zone (slag discharge end), topological characteristics: Introducing damping coefficient (Value) This results in a sparse and decreasing distribution of air vents, and the installation height... It gradually rises as the X-axis coordinate increases.

[0037] Engineering Implementation: This design can actively induce turbulent energy dissipation rate. It decays rapidly in the area near the slag discharge port, providing a slightly disturbed fluid environment for the settling of gold slag (or lead slag, copper slag), effectively reducing entrainment losses in the waste slag.

[0038] The nozzle structure and actuator: The nozzle assembly (spray gun) is not only a channel for injecting reactive gases, but also the physical execution end for achieving coordinated control of multi-zone stirring in the molten pool. To meet the requirements of nonlinear gun placement and dynamic control, each nozzle integrates a four-in-one execution loop of dynamic sensing, precise flow regulation, flow pattern transformation, and pulse excitation. Each nozzle is equipped with an independent flow control valve (FCV) and pressure sensor (PT). The nozzle contains adjustable swirl vanes or a central cone, driven by an electric actuator, which can adjust the swirl number of the jet online, thereby changing the jet expansion angle. When optimization calculations require reducing the penetration depth... To protect the opposite furnace wall, there is no need to reduce the flow rate; simply increase the swirl intensity to make the jet more diffuse, and the kinetic energy will be dissipated in the middle of the molten pool.

[0039] The dynamic sensing and precision adjustment loop, with each side-blowing branch independently configured with an execution terminal, enables millisecond-level response to the jet momentum: Precision Mass Flow Control Actuator (MFC / FCV): Used for online adjustment of the total momentum and oxygen / gas ratio of oxygen-enriched air at each air outlet. This unit receives data from the MCU. The instructions ensure that the energy gap in the dead zone can be accurately compensated in the homogenization mode, while the flow load in the high scouring zone can be flexibly transferred in the furnace protection mode.

[0040] High-frequency pressure pulsation sensing unit (PT): A high-sensitivity pressure transmitter is installed at the front end of the spray gun air inlet to capture the back pressure pulsation signal during the bubble detachment process in real time. This data is not only used to invert the equivalent diameter of bubbles. It also serves as an MCU for measuring the overall pool mixing consistency index. The core input source for computation.

[0041] The flow pattern adjustable swirl control mechanism is used to control the penetration depth of the jet without changing the total oxygen supply. The present invention incorporates an adjustable swirl unit inside the air vent: Core components: The air vent is equipped with adjustable swirl vanes or a central adjusting cone driven by an electric or pneumatic actuator.

[0042] The principle of swirl control is to dynamically adjust the swirl number of the jet airflow by changing the blade opening or the axial position of the core tube relative to the nozzle end face. .

[0043] Column-cone conversion mechanism, when the furnace wall erosion risk index When the warning threshold is approached, the MCU instruction execution mechanism increases the swirl number. At this point, the jet changes from restricted columnar penetration to divergent conical expansion, significantly increasing the jet expansion angle; this change shortens the kinetic energy attenuation characteristic length of the jet core region. This causes the kinetic energy of the high-speed fluid to dissipate in the multiphase friction in the middle of the molten pool, reducing the residual impact pressure on the opposite side furnace lining from a physical source. ; The pulse generation and suppression execution unit, designed to address the standing wave intervention requirements in suppression mode, integrates a high-response pulse control system into the air supply pipeline at the vent: A fast-switching pulse valve is installed, and an electromagnetic pulse generator is set at the branch of the high-pressure gas supply main line to support intermittent power input at a controlled frequency. Asymmetric pulse execution strategy, upon detection of destructive standing waves At that time, the MCU commands the left and right side air vent arrays to switch to a phase difference of [value missing]. The pulse jet state; through the alternating action of momentum input from both sides, an active damping torque is generated to counteract the single-frequency standing wave oscillation of the molten pool surface, thereby protecting the refractory material in the slag line area from periodic thermal shock damage. To ensure project feasibility, all adjustment parameters of the actuators are mapped to physically measurable values. Flow control vector : Directly corresponds to the feedback opening degree of FCV and standard condition flow rate ; Swirl opening The blade position fed back by the actuator displacement sensor is compared with the preset swirl number. Mapping the association model; pulse frequency The value is determined directly by the frequency of the digital trigger signal output by the MCU, and its value is dynamically matched based on the real-time monitored inherent oscillation frequency of the molten pool.

[0044] In summary, this embodiment utilizes the principles of hydraulic collision and shear dissipation in fluid mechanics. Through precise misalignment and swirling adjustment of the left and right tuyeres, the kinetic energy of the high-speed airflow is primarily dissipated in the fluid interaction at the center of the molten pool, rather than directly acting on the opposite furnace lining. This is combined with an online scour risk prediction model. The closed-loop control effectively eliminates local hotspots and high-risk areas of mechanical scouring. Through a nonlinear, staggered layout in three-dimensional space, the early coalescence of adjacent bubble plumes is geometrically eliminated. The complex three-dimensional vortex structure induced by the staggered jet within the molten pool causes the bubbles to be broken down into smaller, more uniformly distributed particles, significantly increasing the specific surface area of ​​the gas-liquid two-phase system, thereby improving oxygen utilization and smelting intensity.

[0045] Furthermore, such as Figure 4 As shown, the multidimensional data acquisition and preprocessing component 1 includes: The module 11 for extracting pulsation feature values ​​is used to... The high sampling rate reads the pressure transmitter signals of each air outlet branch, and the DC component is filtered out by a high-pass filter to extract the pulsating characteristic values ​​that characterize the frequency and intensity of bubble detachment. The real-time power spectral density map generation module 12 is used to synchronously acquire the acoustic vibration monitoring array signal installed in key parts of the furnace shell, immediately perform fast Fourier transform, and generate a real-time power spectral density map. The bottom operating status data module 13 of the smelting furnace is used to read the furnace wall temperature field matrix fed back by the embedded thermocouple, as well as the current actual flow feedback value and valve opening status of each mass flow controller.

[0046] Preferably, in this embodiment, the fluid dynamics signal acquisition is based on... High sampling rate reading of pressure transmitter signals from each air outlet branch The DC component was filtered out using a high-pass filter, and pulsating characteristic values ​​representing the frequency and intensity of bubble detachment were extracted. Structural response signals were acquired, and acoustic vibration monitoring array signals installed at key parts of the furnace shell were simultaneously acquired. Immediately perform a Fast Fourier Transform (FFT) to generate a real-time power spectral density map, focusing on key areas. Low-frequency energy peak; thermodynamic and process parameter readings of the furnace wall temperature field matrix fed back by embedded thermocouples. and the current actual flow feedback values ​​of each mass flow controller (MFC). The system monitors valve opening status. An acoustic vibration monitoring array, with high-temperature accelerometers and acoustic emission sensors arranged in a grid along the outer surface of the furnace shell, collects vibration spectrum signals of the fluid impacting the furnace wall inside the molten pool. The acoustic signals are used to capture the turbulence intensity distribution and standing wave oscillation frequency within the molten pool. Tubular pressure pulsation monitoring involves installing pressure transmitters in the air inlet branches of each tubular nozzle to monitor the back pressure pulsation frequency, extracting the dominant frequency characteristics of the back pressure pulsation, and using empirical correlation to extrapolate the bubble detachment frequency, diameter, penetration depth, and gas-slag contact surface area in real time. A wall temperature integrated monitoring network uses thermocouples arranged in a matrix along the outer surface of the furnace shell; wall temperature data is used for long-term assessment of the remaining thickness of the local furnace lining. Process feedback inputs are connected to macroscopic process indicators such as material throughput, melt grade, and oxygen consumption from the DCS system. This system achieves digital reconstruction of the unsteady flow field inside the molten pool through a logical chain of multi-source data acquisition, feature extraction, multi-physical quantity fusion, and state inversion. The specific processing flow is as follows: In the multi-source heterogeneous data acquisition layer, the system synchronously acquires the following underlying data at different sampling frequencies: High-frequency dynamic data: obtained through acoustic vibration monitoring array and air outlet pressure sensor, The above frequencies are used to collect furnace wall acceleration signals. and back pressure pulsation signal at the air outlet Long-term status data: Furnace shell matrix temperature is obtained through an integrated wall temperature monitoring network. The monitoring frequency is usually 100%. Macroscopic process parameters: Feed rate read in real time from the DCS system. Oxygen concentration and melt level Multi-dimensional feature extraction and preprocessing: Real-time processing of the original signal to extract key feature values ​​reflecting the flow state: Acoustic spectrum analysis: Fast Fourier Transform (FFT) is used to extract the energy concentration frequency band of furnace wall vibration. If in If a peak characteristic appears in the frequency band, it is identified as a molten pool standing wave (VSWR) warning signal. Back pressure pulsation extraction: This is achieved by extracting the dominant frequency characteristic of the back pressure signal. The detachment frequency and initial equivalent diameter of bubble generation are estimated in real time using empirical correlations. Spatial correlation analysis: Calculate the correlation coefficient of vibration signals in adjacent wind outlet areas to determine whether there is unexpected lateral coalescence of the bubble plume; Data fusion based on Kalman filter: The system uses an extended Kalman filter as the fusion kernel to couple discrete sensor observations with the physical model: Prediction step: Utilize a reduced-order fluid model (ROM) based on the current jet vector... Predicting the kinetic energy distribution of the flow field at the next moment Update step: Using the measured pressure pulsation inversion values ​​and acoustic signature energy distribution as observation constraints, the prediction error is corrected, and the optimal estimated three-dimensional gas holdup of the molten pool is output. and flow velocity distribution field; key indicator calculation and decision output, based on the fused full information, update the following evaluation indicators in real time: scour risk assessment: combined with Estimated remaining furnace lining thickness and real-time kinetic impact pressure Dynamically update the erosion risk index of each trend. Stirring consistency evaluation: Summarize the stirring power density of each logical zone in the entire pool and calculate the normalized stirring consistency index. Mode trigger judgment: MCU based on , Based on the vibration spectrum characteristics, logical switching or multi-objective weighted solution is performed between homogenization mode, wave suppression mode and furnace protection mode.

[0047] In summary, the multi-zone collaborative control strategy in this embodiment breaks away from the traditional egalitarian gas supply model. The system can identify and activate flow dead zones, dynamically adjusting the local stirring power density. This makes the temperature and composition fields of the entire furnace (from the feed end to the slag discharge end) more uniform, effectively preventing the problem of cold material crusting at the furnace bottom and dead corners.

[0048] Furthermore, such as Figure 5 As shown, risk boundary assessment component 2 includes: The price reduction molten pool mechanism model module 21 is used to construct a price reduction molten pool mechanism model using the bottom operating status data of the smelting furnace; construct a nonlinear lance design topology function model to maximize the gas-slag contact area and homogenize the turbulent dissipation rate; establish a molten pool state monitoring and evaluation index model based on tuyer parameters; and construct a furnace scour risk assessment model to assess the scour intensity based on the relationship between jet penetration depth and furnace width. Among them, the degradation molten pool mechanism model includes a nonlinear lance design topology function model, a molten pool condition monitoring and evaluation index model, and a furnace scour risk assessment model. The whole pool status index module 22 is used to derive the whole pool status index from the price reduction molten pool mechanism model; compare the amplitude of the low-frequency main peak in the vibration power spectrum with the preset safety threshold to identify standing wave risk; and calculate the furnace wall scour risk index of each tuyer area based on the real-time furnace wall temperature and its rate of change, combined with the local heat flow coupling model.

[0049] If the amplitude exceeds the limit and there is obvious frequency consistency, the system is determined to be in the precursor region of standing wave oscillation. When the furnace wall erosion risk index falls within the preset range, it is determined to be a high-risk area; The module 23 for calculating the overall pool mixing consistency index is used to analyze the variance consistency of the pressure pulsation signals in each zone and calculate the overall pool mixing consistency index. When the mixing consistency index is close to the preset value, it indicates that the kinetic energy distribution of the molten pool is more uniform.

[0050] Preferably, the nonlinear gun placement design topology function model in this embodiment To maximize the gas-slag contact area and homogenize the turbulent dissipation rate, this invention defines the vent coordinates. Specific topological constraints must be met. Assume the furnace body length is... The total number of wind spots is Definition of the first The height of the wind vent It follows a damped sinusoidal distribution function:

[0051] in: Reference air outlet height (m); Amplitude, range of values ; Wave number determines the number of offset cycles along the furnace length, and is usually taken as... ; Phase angle. Left side wall. right side wall This achieves complete misalignment of the peaks and valleys on both sides; Damping coefficient, used to adjust the attenuation of stirring intensity in the settling zone at the tail of the furnace; This layout creates an interlocking jet field in space, causing the vortex rings generated by adjacent jets to shear each other rather than merge during the ascent. When the jet on the left penetrates to the right, it is located in the gap between the two jets on the right, avoiding energy loss caused by direct collision of the jets. At the same time, it increases the lateral mixing component of the fluid in the horizontal direction. To achieve online control, this invention establishes a simplified fluid dynamics model based on vent parameters for molten pool condition monitoring and evaluation. The control system calculates the following core indicators in real time as input data for closed-loop control: Gas-slag contact specific surface area ( Local gas holdup is inverted through pressure pulsation. The calculation formula is:

[0052] in: Local gas holdup; : Equivalent diameter of the bubble; The aim is to maximize this value in the reaction zone to improve reaction efficiency. This invention utilizes the coherent vortex interlocking effect generated by a staggered layout to enhance reaction efficiency. The gas holdup distribution generated by a single staggered jet in the molten pool can be described by a modified drift flow model. Due to the spatial staggering, bubble groups generated by adjacent vents do not overlap or coalesce on their upward paths. The control system adjusts the flow rate of each zone to... It remains at its maximum in the reaction zone, while rapidly decaying in the settling zone.

[0053] like Figure 6 As shown in the figure, the gas holdup distribution cloud map (b) under the traditional linear layout (a) and the staggered layout proposed in this invention is compared through CFD simulation results to illustrate the difference between the gas-slag contact area and the bubble aggregation effect. Figure 5 As shown, in a traditional linear close-packed layout, the bubble plumes generated by adjacent air outlets are prone to lateral interference and coalescence, forming large bubbles or air cavities, resulting in a decrease in the gas-liquid surface area. However, in the nonlinear staggered layout of this invention, by dispersing the rising path of the bubble plumes in three-dimensional space, bubble coalescence is suppressed, making the bubbles smaller and more evenly distributed, thereby maximizing the gas-slag contact area.

[0054] Furnace wall erosion risk assessment model, based on jet penetration depth With furnace width The relationship between the scouring intensity and the control system is used to assess the scouring intensity. The control system calculates the following key indicators in real time as input data for closed-loop control: Jet penetration depth

[0055] Based on the modified Froude number, the first Jet penetration depth at each vent The calculation is as follows: in: : Air outlet diameter (m); Density of gas and slag / melt (kg / m³); : Airflow velocity (m / s); : Gravitational acceleration; : Empirical constant, usually taken as ; Furnace wall erosion risk index

[0056] Define the furnace wall erosion risk index : in: Furnace width (m); : No. Input kinetic energy power of each air outlet ( ); The characteristic length (m) of the kinetic energy decay of the jet in the melt is determined by the fluid viscosity; : The benchmark value for the erosion resistance of refractory materials; Constraints: The control objective must meet the following conditions. ; Impact pressure acting on the furnace side wall

[0057] The system identifies the jet penetration depth in real time. With half the width of the furnace The proportional relationship is used to assess the scouring intensity; the local energy attenuation coefficient is defined. The impact pressure acting on the furnace sidewall for: MCU according to The ratio of the refractory material yield strength is used to adjust the swirl number of the affected area in real time. Without reducing the total flow rate, the residual kinetic energy of the core area reaching the furnace wall is reduced by increasing the jet expansion angle.

[0058] Uniformity Index (UI) of the entire pool mixing Using the standard deviation of back pressure pulsation signals from different areas To characterize the intensity of local turbulence.

[0059] The overall mixing consistency index is defined as the reciprocal of the normalized variance:

[0060] Among them, local stirring power density The estimated value (W / m³) is:

[0061] Standard gas flow rate; : The wind gap is buried deep; : No. Effective molten pool volume of each control unit; : Kinetic energy transfer efficiency coefficient; System control commands are sent to intelligent execution terminals distributed on both sides of the furnace body, including: Precision mass flow control valve (MFC) is used to adjust the oxygen / air ratio and total momentum of each air outlet in milliseconds; spray gun swirl adjustment mechanism changes the flow pattern (expansion angle) of the jet by adjusting the blade opening or core tube position inside the spray gun; pulse generation unit is set with a fast switching valve in the high-pressure air supply line to realize intermittent power input with controlled pulse frequency.

[0062] In summary, this embodiment introduces acoustic vibration feedback and pulse suppression control for the first time, which can actively suppress destructive standing wave oscillations on the molten pool surface. This not only protects the refractory materials in the slag line area but also avoids splashing and slagging in the flue caused by violent tumbling, ensuring the stable operation of large-scale smelting equipment.

[0063] Furthermore, such as Figure 7 As shown, the optimal control quantity solution component 3 includes: The priority division module 31 is used by the state machine scheduler to divide the priority into first priority, second priority and third priority according to the preset priority logic; the weight coefficient is dynamically adjusted according to the acoustic vibration signal and wall temperature status, and when a sudden vibration alarm is detected, the weight coefficient is instantly assigned a maximum value. Among them, the first priority is the safety bottom line of the wave suppression mode. Standing wave oscillation may cause serious runaway, splashing, or even structural damage to the furnace body. Therefore, as long as the vibration spectrum characteristics meet the triggering conditions, the system will unconditionally interrupt the current homogenization or regular adjustment task and force the system to enter the wave suppression execution sequence. The wave suppression mode has the highest execution authority before the oscillation energy decays to below the safety threshold. Second priority: Furnace protection mode for asset protection. Under the premise of no standing wave risk, the system prioritizes protecting the furnace lining thickness and equipment lifespan; if the risk of localized erosion is high... If the flow rate exceeds the limit, the system will restrict flow rate increase requests in homogenization mode and prioritize vortex divergence control; if there is a conflict between the furnace protection logic and the homogenization logic in the flow rate setting, the system will prioritize the flow rate setting. As a prerequisite constraint, the oxygen content of adjacent misaligned air outlets is redistributed to maintain the total oxygen balance. Third priority: Homogenization mode process optimization. When the system operates within the safe envelope, i.e., there are no standing waves or severe scouring risks, control is transferred to the homogenization mode. At this time, the MPC algorithm maximizes the overall mixing consistency index. To optimize the target, the kinetic energy input of each zone is finely adjusted in order to pursue the highest reaction efficiency and oxygen utilization rate; Model prediction module 32 is used to divide the furnace body into Each logical control zone employs a model predictive control strategy. Based on the model predictive control framework, the actuators are driven to switch between different operating modes by real-time observation of the entire pool state. The adjustment path module 33 is used to switch the standing wave oscillation control mode, identify abnormal features, and execute asymmetric pulse control logic; switch the furnace wall scour protection mode, perform scour risk warning, and execute swirl divergence adjustment strategy; switch the stirring consistency improvement mode, perform logical judgment indicators, and execute the adjustment path.

[0064] Preferably, in this embodiment, the furnace body is divided into: There are three logical control regions; these include the feed preparation region, main reaction region, and depletion region, which employ Model Predictive Control (MPC) strategy. The objective function is optimized. The aim is to maximize mixing consistency while minimizing the risk of furnace wall erosion and gas supply energy consumption.

[0065] in: : Airflow velocity control vector for each air outlet : Safety scouring threshold; Weighting coefficients; The dynamic increase occurs when abnormal furnace wall vibration is detected. The multi-zone stirring and co-control unit (MCU) serves as the system's logical hub, performing full-pool state assessments with a decision cycle of 10-30 seconds. Based on the model predictive control (MPC) algorithm, the system dynamically switches between or executes the following three core control modes in parallel by solving multi-objective weighted solutions to the real-time sensed flow field characteristics, thereby achieving the optimal balance between production efficiency, energy efficiency, and furnace lifespan throughout the entire cycle: The standing wave oscillation suppression mode, also known as the wave suppression mode, is designed to address the single-frequency standing wave oscillation (Sloshing) on ​​the liquid surface that is easily induced by high-intensity blowing in the side-blown molten pool. The system intervenes through active damping technology.

[0066] Anomaly identification: The acoustic signature monitoring array captures the low-frequency vibration spectrum of the furnace shell surface in real time. When anomalies are detected... When a sustained energy peak occurs within the frequency band and the fluctuation amplitude of the liquid surface exceeds the safety threshold, the system determines that a destructive standing wave has been induced.

[0067] Asymmetric pulse control logic: The MCU instruction execution mechanism switches to pulsed blowing mode. The system forces the air inlet arrays on the left and right sides of the furnace to increase the air supply intensity. The time phase difference. Its pulse frequency. Based on furnace body dimensions and real-time liquid level The calculated natural frequency of the standing wave is dynamically matched.

[0068] Physical effect: By utilizing the time difference of momentum input on both sides, a reverse damping torque is generated to actively counteract the undulation component of the liquid surface. This mode can effectively suppress violent churning on the surface of the molten pool and prevent the refractory material at the slag line from suffering periodic thermal shock damage and metal splashing.

[0069] Furthermore, the model prediction module 32 specifically includes: The first optimal setpoint formation submodule is used by the multi-zone stirring collaborative control unit MCU to perform a fast Fourier transform on the vibration spectrum and extract... The system calculates the peak energy value of the frequency band and compares it with a preset safety threshold. If the peak energy value exceeds the limit and has frequency consistency, the system is determined to be in the precursor region of standing wave oscillation, triggering the wave suppression mode. The multi-zone stirring and co-control unit (MCU) calculates the natural frequency of the standing wave based on the furnace body's characteristic dimensions and real-time liquid level, and uses this to set the asymmetric pulse jet parameters: the pulse frequency of the left and right side air vent arrays and... Phase difference, and at the same time, redistribute the air supply intensity of each air outlet according to the pulse timing; generate digital optimal settings, including the pulse frequency setting value, phase difference setting value, and flow timing allocation value of each air outlet in each pulse cycle; The second optimal setpoint generation submodule is used to collect pressure transmitter signals and furnace wall thermocouple temperature field data of each tuyere branch in real time. The multi-zone stirring and co-control unit (MCU) calculates the ratio of the jet penetration depth of each tuyere to the half-width of the furnace based on the jet penetration depth model, and evaluates the local furnace wall scour risk index based on the kinetic energy decay model, while monitoring the wall temperature change rate of the corresponding area. If the local furnace wall scour risk index is close to or exceeds the warning value and the wall temperature rises abnormally, it is determined that there is a high scour risk and the furnace protection logic is triggered. For high-risk tuyeres, the multi-zone stirring and co-control unit (MCU) increases the setpoint of the swirl number S through the spray gun swirl adjustment mechanism without changing the total oxygen content, so that the jet expansion angle increases and the kinetic energy is dissipated in the middle of the molten pool, while keeping the flow rate setpoint of the tuyere unchanged. The MCU generates digital optimal setpoints, including the opening setting value of the swirl adjustment mechanism of the high-risk tuyere and the command value of the corresponding flow controller to maintain the original flow rate. The third optimal setpoint formation submodule is used by the multi-zone stirring co-control unit MCU to collect the back pressure pulsation signal of each duct in real time and receive macroscopic process parameters from the DCS; it calculates the overall stirring consistency index by inverting the local stirring power density through back pressure pulsation, and monitors the pressure fluctuation variance of each multi-zone stirring co-control unit, comparing the local stirring power density of each zone with the overall average value of the whole pool; if the local stirring power density of a certain multi-zone stirring co-control unit is continuously lower than 60% of the overall average value of the whole pool and the pressure fluctuation variance is continuously lower than the threshold, it is determined that there is a risk of flow stagnation or cold material crusting in that zone, triggering the homogenization mode; with the goal of maximizing the overall stirring consistency index of the whole pool, the model predictive control algorithm is used to solve the flow increment distribution of each duct, focusing on increasing the oxygen distribution ratio of the target area and its adjacent nonlinear misaligned ducts, inducing secondary circulation to activate the stagnation zone; and generating digital optimal setpoints, including the flow setpoint increment of each duct and the corresponding mass flow controller opening command.

[0070] Preferably, this embodiment extracts the energy peak value of a specific frequency band using Fast Fourier Transform and compares it with a safety threshold to promptly detect and suppress potential standing wave oscillation precursors, thereby improving system stability. Based on the furnace body's characteristic dimensions and real-time liquid level, the natural frequency of the standing wave is calculated, and asymmetric pulse jet parameters, including pulse frequency, phase difference, and flow timing distribution, are set to achieve a better furnace stirring effect. The risk of local furnace wall erosion is assessed using jet penetration depth and kinetic energy attenuation models, and the set value of the spray gun swirl number S is adjusted in a timely manner to reduce the risk of local erosion. The pressure fluctuation variance of each multi-zone stirring and co-control unit is monitored in real time, and the flow distribution of each tuyer is optimized using a model predictive control algorithm, particularly focusing on increasing oxygen distribution in target areas to activate stagnant zones and improve overall pool stirring consistency. Through back pressure pulsation signal analysis and monitoring of macroscopic process parameters, potential risks of flow stagnation or cold material crusting are identified, and a homogenization mode is triggered to increase oxygen distribution in specific areas to avoid these risks.

[0071] In summary, the combined use of the adjustment path module in this embodiment enables more precise and effective furnace stirring control, optimizes the furnace environment, reduces the risk of local scouring and crust formation, and improves overall stirring efficiency and uniformity, thereby enhancing the safety and operating efficiency of the industrial furnace.

[0072] Furthermore, the third optimal setting value formation submodule specifically includes: The feature extraction unit is used to input the high-frequency back pressure pulsation signal collected by the pressure transmitter of each vent branch into the multi-zone stirring and co-control unit; the multi-zone stirring and co-control unit reads the signal at a sampling rate of not less than 1 kHz, filters out the steady-state DC component through the built-in high-pass filter, extracts the pulsation main frequency and amplitude generated during the bubble detachment process of each vent, and forms the bubble detachment frequency and pulsation intensity characteristic value of each vent. The inversion processing unit is used to input the bubble detachment frequency and pulsation intensity characteristic values ​​of each tuyere into the flow state inversion program built into the risk boundary assessment component. Using empirical correlations calibrated in advance through cold-state experiments or computational fluid dynamics simulations, the local gas holdup and equivalent bubble diameter of the molten pool region where each tuyere is located are inverted. Then, combined with the real-time flow rate of the tuyere, the tuyere burial depth and the molten pool volume of the corresponding logic control area, the local stirring power density of the region is estimated based on the kinetic energy dissipation theory to form the local stirring power density value of each logic control area. The reciprocal operation unit is used to summarize the local stirring power density values ​​of each logic control zone to the multi-zone stirring collaborative control unit. The multi-zone stirring collaborative control unit calculates the average value of the stirring power density of the whole pool and solves the root mean square of the sum of squares of the relative deviations of the power density of each control zone relative to the average value. Through the normalized variance reciprocal operation, the stirring consistency index of the whole pool between 0 and 1 is obtained, which is used to quantify the uniformity of the kinetic energy distribution of the molten pool.

[0073] Preferably, this embodiment, through the flow state inversion program of the high-frequency back pressure pulsation signal analysis and inversion processing unit using feature extraction, can accurately obtain key parameters such as local gas holdup and bubble diameter in each tuyer region within the molten pool, achieving precise control over the internal state of the molten pool. By calculating the whole-pool stirring consistency index using reciprocal operations, the uniformity of kinetic energy distribution within the molten pool can be quantified, providing an important reference for molten pool stirring control. By combining the local stirring power density values ​​of each logic control zone with the average value of the whole-pool stirring power density, the stirring efficiency of the molten pool can be optimized, reducing energy consumption and improving smelting efficiency. Precise control and uniformity assessment of the state of each region within the molten pool can prevent abnormal phenomena such as local overheating and coking, reducing safety risks and extending furnace life. Collecting and analyzing key parameters in each region within the molten pool can provide rich data support for intelligent furnace condition diagnosis, improving the accuracy and reliability of the diagnosis. Precise control and assessment of the molten pool stirring state can improve product quality and reduce production costs.

[0074] Furthermore, the inversion processing unit specifically includes: The power estimation subunit is used by the multi-zone stirring and co-control unit to estimate the mechanical work released by the gas as it expands from the nozzle to the surface of the melt, based on thermodynamic principles and real-time pressure and temperature parameters in the molten pool, and to obtain the power value of gas expansion at each nozzle. The kinetic energy estimation subunit is used by the multi-zone stirring and co-control unit to estimate the effective kinetic energy transferred to the surrounding melt by the high-speed jet during the process of breaking into bubbles, based on the jet dynamics theory, and to obtain the jet kinetic energy transfer power value of each vent. The accumulation processing subunit is used to accumulate the power value of gas expansion and the power value of jet kinetic energy transfer at each vent within the same logic control area. The total power value after accumulation is divided by the preset molten pool volume in the logic control area to obtain the local stirring power density value of the logic control area.

[0075] Preferably, this embodiment enables precise control over the work done by gas expansion and the effective kinetic energy transfer of the jet during the molten pool stirring process. The power estimation subunit accurately estimates the mechanical work released by the gas expansion in the melt, helping to understand the impact of the gas on the molten pool stirring. The kinetic energy estimation subunit estimates the effective kinetic energy transferred to the surrounding melt during the high-speed jet breaking into bubbles, which is crucial for optimizing the stirring effect within the furnace. Combining these estimation results, the accumulation processing subunit adds the gas expansion work power value and the jet kinetic energy transfer power value, calculating the local stirring power density value. This process allows for the assessment of the stirring intensity in different areas of the molten pool, and then the adjustment of the stirring parameters at each tuyeres according to the actual situation, ensuring uniform and effective stirring within the molten pool. This provides strong support for achieving refined control of the stirring process within the molten pool, contributing to improved smelting quality and increased production efficiency.

[0076] Furthermore, the accumulation processing subunit specifically includes: The classification component is used to classify the gas expansion power value and jet kinetic energy transfer power value of each vent according to the preset logical control area classification rules; the multi-zone stirring and co-control unit matches each vent to the corresponding logical control area based on the spatial coordinate range of the feeding area, main reaction area and settling area defined during the nonlinear gun design, forming a list of vents belonging to each logical control area and its corresponding two types of power value datasets. The power accumulation component is used by the multi-zone stirring collaborative control unit to traverse the list of air vents belonging to each logical control zone. For all air vents in the same logical control zone, the power value of gas expansion work done by each air vent in the zone is accumulated together to obtain the total power value of gas expansion work done by the logical control zone. At the same time, the power value of jet kinetic energy transfer from each air vent in the zone is accumulated together to obtain the total power value of jet kinetic energy transfer by the logical control zone. The total power acquisition component is used by the multi-zone stirring and co-control unit to add the total work power value of gas expansion in the same logical control zone to the total transfer power value of jet kinetic energy, and obtain the total input mechanical power value of the logical control zone.

[0077] Preferably, in this embodiment, firstly, the classification component categorizes the power values ​​of the air vents according to preset rules, ensuring the organization and classification of information. Next, the multi-zone stirring and co-control unit accurately matches the air vents to their corresponding logical control zones based on the spatial coordinates of the feeding zone, main reaction zone, and settling zone, forming an air vent list and a power value dataset, providing accurate basic data for subsequent power accumulation. Through the operation of the power accumulation component, the gas expansion work power and jet kinetic energy transfer power of all air vents within each logical control zone can be accumulated to obtain the total power value of each logical control zone. Finally, the total power acquisition component integrates the total gas expansion work power and the total jet kinetic energy transfer power to obtain the total input mechanical power value of the logical control zone. This effectively and accurately calculates and manages the air vent power, optimizes air vent power control, and improves system efficiency and accuracy.

[0078] Furthermore, the total power acquisition component specifically includes: The consistency comparison sub-component is used by the multi-zone stirring collaborative control unit to extract two power values ​​corresponding to the same logical control zone from the total power dataset of gas expansion and the total power dataset of jet kinetic energy transfer of the power accumulation component; and to perform consistency comparison based on the logical control zone identifier and timestamp attached to the two power values ​​to confirm that they belong to the same control zone and are in the same control cycle, thus forming a power data pair that has been paired in the logical control zone. The numerical merging sub-component is used by the multi-zone stirring and collaborative control unit to input the paired gas expansion total work power value and the jet kinetic energy transfer total power value into the internal numerical merging module, perform linear superposition calculation of the power values, and obtain the preliminary accumulation result of the input mechanical power of the logic control area. The numerical verification sub-component is used by the multi-zone stirring collaborative control unit to verify the unit dimension of the preliminary accumulation result, confirm that it is consistent with the preset power unit standard, and store the verified value in the total input mechanical power cache of the corresponding logic control area to form a total input mechanical power value that can be used for local stirring power density calculation.

[0079] Preferably, in this embodiment like Figure 8 The diagram illustrates the wave suppression control logic flow based on furnace wall vibration feedback. The flow first acquires furnace wall vibration signals using an acoustic vibration monitoring array. When a specific low frequency (such as...) is detected inside the furnace... When the overall synchronous fluctuation of the signal exceeds the first-level alarm threshold, the system determines that a standing wave oscillation has been induced and immediately triggers the suppression mode. In this mode, the control system switches to asymmetric pulse jet logic, introducing amplification into the air supply intensity of the left and right side air vent arrays. The time phase difference between the two sides is used to generate a reverse damping torque, which actively counteracts the fluctuations in the liquid surface.

[0080] Furnace wall scour protection mode, also known as furnace protection mode, will detect scour of a certain side wall. When the warning threshold is approached, the system adjusts the kinetic energy distribution of the jet flow pattern in space by adjusting the actuator (such as the dual-channel spray gun regulating valve) to achieve precise protection of the sidewall, while keeping the total smelting intensity unchanged.

[0081] Scour risk warning: The system identifies the jet penetration depth in real time. With half the width of the furnace The proportional relationship. If the calculated local furnace wall erosion risk index... Approaching or exceeding the warning value (usually set at 0.8), accompanied by localized wall temperature If the temperature rises abnormally, the furnace protection logic will be triggered.

[0082] Swirl divergence regulation strategy: The MCU does not reduce the total oxygen content at the air outlet to maintain regional thermal balance, but instead increases the swirl number of the jet airflow through the spray gun swirl regulation mechanism. .

[0083] Energy dissipation mechanism: As the swirl intensity increases, the jet expansion angle also increases, transforming the jet from a confined cylindrical penetration to a divergent conical expansion. This significantly shortens the kinetic energy decay characteristic length of the jet core region. This causes the momentum of the high-speed fluid to be mainly dissipated in the fluid shearing and multiphase friction in the middle of the molten pool, rather than directly and rigidly impacting the furnace lining on the opposite side.

[0084] The stirring uniformity enhancement mode, also known as the homogenization mode, occurs when the system identifies the turbulent energy dissipation rate in a local region through the state observation layer. When the concentration is below 60% of the average value of the entire pool, the stirring and homogenization control logic is automatically triggered, such as in the dead zone of the material or the sedimentation zone at the bottom of the furnace.

[0085] Logical judgment index: Local stirring power density inverted using back pressure pulsation signal and the overall mixing consistency index If a pressure fluctuation variance is detected in a certain control unit... If the flow rate remains below the threshold, it is determined that there is a risk of flow stagnation or cold material crusting in the area.

[0086] Execution adjustment path: The MCU dynamically increases the oxygen distribution ratio of the target area and its adjacent nonlinear misaligned air outlets. By synergistically increasing the input kinetic energy of the misaligned jets on both sides, an enhanced secondary circulation is induced, and the material in the stagnant zone is forcibly activated by utilizing the coherent vortex interlocking effect.

[0087] The improved mixing consistency mode breaks the traditional proportional gas supply mode and realizes the on-demand distribution of mixing energy, fundamentally eliminating the dead zone phenomenon in ultra-large molten pools.

[0088] The engineering implementation logic for mode switching, and the system's logical switching cycle are set as follows: Seconds, through a multi-objective weighted function Achieve a smooth transition. Among these, the weighting coefficients... It will dynamically adjust based on acoustic vibration signals and wall temperature.

[0089] When a sudden vibration alarm is detected, It is instantaneously assigned a maximum value, enabling near real-time switching from averaging mode to suppression mode.

[0090] During the exit process of each mode, the system introduces a damping element to prevent the actuator flow valve and swirl vane from oscillating due to frequent logic switching, thus ensuring the long-term stable operation of the side-blowing process.

[0091] like Figure 9 As shown, a state machine task scheduling system based on the Model Predictive Control (MPC) framework is presented. Based on real-time data from sensors such as acoustic vibration monitoring arrays, the system makes logical judgments and automatically switches between three control modes with clearly defined priorities: the highest priority standing wave oscillation suppression mode serves as a safety baseline, triggering immediately upon detecting a standing wave risk at a specific frequency, employing asymmetric pulse jetting, and forcibly interrupting other modes; the second priority furnace wall erosion protection mode focuses on asset protection, intervening when the furnace wall erosion index or temperature abnormally rises; and the third priority stirring consistency enhancement mode is a routine process optimization method when the system is within its safety envelope, aiming to adjust the flow rate to improve the homogenization effect of the molten pool.

[0092] The MCU converts the calculated digital optimal setpoint into physical control signals and sends them down to the underlying actuators in parallel: Flow adjustment, setting a new flow rate. Sending data to each branch MFC to drive and control valves for precise adjustment of opening; flow pattern transformation to achieve the target swirl number. The displacement command is mapped to the actuator of the vortex mechanism, adjusting the blade angle or the position of the central cone to achieve the morphological switching of the jet from columnar to conical. In pulse excitation only suppression mode, a high-frequency switching signal is sent to the pulse valve group to execute asymmetric pulse jet action. After the command is issued, the system enters a waiting state until the next control cycle. Upon arrival, the above process is repeated, thereby achieving continuous, dynamic, and collaborative optimization control of the side-blown molten pool smelting process.

[0093] In summary, the control model proposed in this invention is based on the classical dimensionless number (Froude number) and the law of conservation of energy in fluid mechanics. All required parameters can be directly measured or obtained through conventional industrial instruments. These parameters include flow rate, pressure, and geometric dimensions. It does not require a complex black-box model and is easy to program and implement in existing DCS / PLC systems, thus having extremely high engineering application value.

[0094] like Figure 10 As shown, this embodiment also provides an embodiment of a method for intelligent coordinated control of multi-zone stirring in a side-blown molten pool furnace. In this embodiment, the method is applied to a system for intelligent coordinated control of multi-zone stirring in a side-blown molten pool furnace. The method specifically includes the following steps: Step S1: At the beginning of each control cycle, the multi-zone stirring and co-control unit triggers a data acquisition command to synchronously acquire fluid dynamics signals, structural response signals, and thermodynamic and process parameters, as well as other data on the bottom operating status of the melting furnace. Step S2: Using the bottom operating status data of the smelting furnace, combined with the built-in price reduction molten pool mechanism model, identify standing wave risk, assess scouring risk, and measure the overall pool status indicators such as stirring uniformity. Step S3: The state machine scheduler determines the dominant mode of the current control cycle based on the preset priority logic and the full pool state index, and calls the response model predictive control algorithm to solve for the actuator setpoint at the next moment, and obtains the digital optimal setpoint. Step S4: Convert the obtained digital optimal setpoint into a physical control signal and send it to the underlying actuator in parallel; perform actions such as flow regulation, flow pattern transformation and pulse establishment; after completing the instruction issuance, enter the waiting state until the next control cycle.

[0095] Preferably, this embodiment employs spatial topology reconstruction, abandoning the traditional linear, equidistant jet arrangement method. Instead, it utilizes a staggered layout strategy with sinusoidal waveforms, sawtooth patterns, or discrete non-uniformity to disperse the rising path of the bubble plume in three-dimensional space, suppressing bubble coalescence and maximizing the gas-slag contact area. Multi-zone flow field decoupling utilizes the coherent vortex structure generated by the staggered jets to construct several interlocking but functionally independent micro-stirring zones within the furnace. Dead zones are eliminated through the interlocking effect of the jets, while fluid damping attenuates the impact kinetic energy on the side furnace walls. Dynamic homogenization control, based on the spectral analysis of molten pool fluctuations and nozzle back pressure feedback, constructs a full-pool stirring consistency evaluation model, adjusting the injection parameters of each tuyeres in real time to achieve spatiotemporal homogenization of stirring intensity. By using a nonlinear spray gun staggered layout to break up the coupling effect of the bubble plume, the spatiotemporal difference of momentum input at each tuyer is used to improve the overall mixing consistency (UI) of the entire pool; at the same time, the coherent eddy shear between the jets dissipates kinetic energy, thereby suppressing the direct scouring and damage of the jets to the furnace wall at the source.

[0096] In summary, this embodiment addresses key technical bottlenecks in large side-blown furnaces, such as uneven stirring, dead zone crusting, and severe local erosion of the furnace wall, by nonlinear reconstruction of the tuyeres' spatial topology, combined with online state observation and multi-objective closed-loop optimization of the gas-liquid-slag multiphase flow field.

[0097] like Figure 11 As shown, this embodiment provides an example of an electronic device, which includes a processor and a memory coupled to the processor.

[0098] The memory stores program instructions for implementing the intelligent collaborative control system for multi-zone stirring of a side-blown molten pool furnace in any of the above embodiments.

[0099] The processor is used to execute program instructions stored in memory to configure the multi-zone stirring intelligent collaborative control system for the side-blown molten pool furnace.

[0100] The processor can also be called a CPU (Central Processing Unit). A processor may be an integrated circuit chip with signal processing capabilities. A processor can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor.

[0101] Furthermore, Figure 12This is a schematic diagram of the structure of a storage medium according to an embodiment of this application. The storage medium of this embodiment stores program instructions capable of implementing all the above-described methods. These program instructions can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or terminal devices such as computers, servers, mobile phones, and tablets.

[0102] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0103] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

[0104] The specific embodiments of the invention have been described in detail above, but these are merely examples, and the invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this invention. Therefore, all equivalent transformations, modifications, and improvements made without departing from the spirit and principles of this invention should be included within the scope of this invention.

Claims

1. A multi-zone stirring intelligent collaborative control system for a side-blown molten pool furnace, characterized in that, The intelligent collaborative control system for multi-zone stirring in the side-blown molten pool furnace includes: The multi-dimensional data acquisition and preprocessing component is used at the beginning of each control cycle. The multi-zone stirring and co-control unit triggers the data acquisition command to synchronously acquire fluid dynamics signals, structural response signals, and thermodynamic and process parameters of the bottom operating status data of the melting furnace. The risk boundary assessment component is used to identify standing wave risk, assess scouring risk, and measure the overall pool status indicators by utilizing the bottom-level operating status data of the smelting furnace and combining it with the built-in price reduction molten pool mechanism model. The optimal control quantity solution component is used by the state machine scheduler to determine the dominant mode of the current control cycle based on the preset priority logic and the full pool state index, and to call the response model predictive control algorithm to solve for the actuator setpoint at the next moment, thus obtaining the digital optimal setpoint. The collaborative instruction issuance and execution component is used to convert the obtained digital optimal setpoint into physical control signals and issue them in parallel to the underlying actuators; to perform flow regulation, flow pattern transformation and pulse establishment execution actions; after completing the instruction issuance, it enters a waiting state until the next control cycle.

2. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 1, characterized in that, Multidimensional data acquisition and preprocessing components include: The module for extracting pulsation feature values ​​is used to... The high sampling rate reads the pressure transmitter signals of each air outlet branch, and the DC component is filtered out by a high-pass filter to extract the pulsating characteristic values ​​that characterize the frequency and intensity of bubble detachment. The real-time power spectral density map generation module is used to synchronously acquire the acoustic vibration monitoring array signals installed in key parts of the furnace shell, perform fast Fourier transform, and generate a real-time power spectral density map. The module for obtaining the bottom operating status data of the smelting furnace is used to read the furnace wall temperature field matrix fed back by the embedded thermocouple, as well as the current actual flow feedback value and valve opening status of each mass flow controller.

3. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 1, characterized in that, Risk boundary assessment components include: The module for the price-reducing molten pool mechanism model is used to construct a price-reducing molten pool mechanism model using the bottom operating status data of the smelting furnace; construct a nonlinear lance design topology function model to maximize the gas-slag contact area and homogenize the turbulent dissipation rate; establish a molten pool state monitoring and evaluation index model based on tuyer parameters; and construct a furnace scour risk assessment model to evaluate the scour intensity based on the relationship between jet penetration depth and furnace width. The whole pool status index module is used to derive the whole pool status index from the price reduction molten pool mechanism model; compare the amplitude of the low-frequency main peak in the vibration power spectrum with the preset safety threshold to identify standing wave risk; and calculate the furnace wall scour risk index of each tuyer area based on the real-time furnace wall temperature and its rate of change, combined with the local heat flow coupling model. The module for calculating the overall pool mixing consistency index is used to analyze the variance consistency of the pressure pulsation signals in each zone and calculate the overall pool mixing consistency index. When the mixing consistency index is close to the preset value, it indicates that the kinetic energy distribution of the molten pool is more uniform.

4. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 1, characterized in that, The optimal control quantity solution component includes: The priority division module is used by the state machine scheduler to divide the system into first priority, second priority and third priority according to the preset priority logic; the weight coefficient is dynamically adjusted according to the acoustic vibration signal and wall temperature status, and when a sudden vibration alarm is detected, the weight coefficient is instantly assigned a maximum value. The model prediction module is used to divide the furnace body into... Each logical control zone employs a model predictive control strategy. Based on the model predictive control framework, the actuators are driven to switch between different operating modes by real-time observation of the entire pool state. The adjustment path module is used to switch the standing wave oscillation control mode, identify abnormal features, and execute asymmetric pulse control logic; switch the furnace wall scour protection mode, provide scour risk warning, and execute swirl divergence adjustment strategy; and switch the stirring consistency improvement mode, perform logical judgment indicators, and execute the adjustment path.

5. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 4, characterized in that, The model prediction module includes: The first optimal setpoint formation submodule is used by the multi-zone stirring and co-control unit to perform a fast Fourier transform on the vibration spectrum, extract the energy peak value in the 0.2Hz to 1.5Hz frequency band, and compare it with the preset safety threshold. If the energy peak value exceeds the limit and has frequency consistency, the system is determined to be in the pre-standing wave oscillation region, triggering the wave suppression mode. The multi-zone stirring and co-control unit calculates the natural frequency of the standing wave based on the furnace body characteristic dimensions and real-time liquid level, and sets the asymmetric pulse jet parameters: the pulse frequency and 180° phase difference of the left and right duct arrays, and redistributes the air supply intensity of each duct according to the pulse timing. It generates digital optimal setpoints, including the pulse frequency setpoint, phase difference setpoint, and flow timing allocation value in each pulse cycle for each duct. The second optimal setpoint generation submodule is used to collect pressure transmitter signals and furnace wall thermocouple temperature field data of each tuyere branch in real time. The multi-zone stirring and co-control unit calculates the ratio of the jet penetration depth of each tuyere to the half-width of the furnace based on the jet penetration depth model, and evaluates the local furnace wall scour risk index based on the kinetic energy decay model, while monitoring the wall temperature change rate of the corresponding area. If the local furnace wall scour risk index is close to or exceeds the warning value and the wall temperature rises abnormally, it is determined that there is a high scour risk and the furnace protection logic is triggered. For high-risk tuyeres, the multi-zone stirring and co-control unit increases the setpoint of the swirl number through the spray gun swirl adjustment mechanism without changing the total oxygen content, so that the jet expansion angle increases and the kinetic energy is dissipated in the middle of the molten pool, while keeping the flow rate setpoint of the tuyere unchanged. It generates digital optimal setpoints, including the opening setpoint of the swirl adjustment mechanism of the high-risk tuyere and the command value of the corresponding flow controller to maintain the original flow rate. The third optimal setpoint formation submodule is used to collect back pressure pulsation signals from each vent in real time by the multi-zone stirring and co-control unit, and receive macroscopic process parameters from the DCS. It calculates the overall pool stirring consistency index by inverting local stirring power density through back pressure pulsation, and monitors the pressure fluctuation variance of each multi-zone stirring and co-control unit, comparing the local stirring power density of each zone with the overall pool average. If the local stirring power density of a certain multi-zone stirring and co-control unit is consistently lower than 60% of the overall pool average and the pressure fluctuation variance is consistently lower than the threshold, it is determined that there is a risk of flow stagnation or cold material crusting in that zone, triggering the homogenization mode. With maximizing the overall pool stirring consistency index as the optimization objective, a model predictive control algorithm is used to solve the flow increment distribution of each vent, focusing on increasing the oxygen distribution ratio of the target area and its adjacent nonlinear misaligned vents, inducing secondary circulation to activate the stagnation zone. A digital optimal setpoint is generated, including the flow setpoint increment of each vent and the corresponding mass flow controller opening command.

6. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 5, characterized in that, The third optimal setting value formation submodule includes: The feature extraction unit is used to input the high-frequency back pressure pulsation signal collected by the pressure transmitter of each vent branch into the multi-zone stirring and co-control unit; the multi-zone stirring and co-control unit reads the signal at a sampling rate of not less than 1 kHz, filters out the steady-state DC component through the built-in high-pass filter, extracts the pulsation main frequency and amplitude generated during the bubble detachment process of each vent, and forms the bubble detachment frequency and pulsation intensity characteristic value of each vent. The inversion processing unit is used to input the bubble detachment frequency and pulsation intensity characteristic values ​​of each tuyere into the flow state inversion program built into the risk boundary assessment component. Using empirical correlations calibrated in advance through cold-state experiments or computational fluid dynamics simulations, the local gas holdup and equivalent bubble diameter of the molten pool region where each tuyere is located are inverted. Then, combined with the real-time flow rate of the tuyere, the tuyere burial depth and the molten pool volume of the corresponding logic control area, the local stirring power density of the region is estimated based on the kinetic energy dissipation theory to form the local stirring power density value of each logic control area. The reciprocal operation unit is used to summarize the local stirring power density values ​​of each logic control zone to the multi-zone stirring collaborative control unit. The multi-zone stirring collaborative control unit calculates the average value of the stirring power density of the whole pool and solves the root mean square of the sum of squares of the relative deviations of the power density of each control zone relative to the average value. Through the normalized variance reciprocal operation, the stirring consistency index of the whole pool between 0 and 1 is obtained, which is used to quantify the uniformity of the kinetic energy distribution of the molten pool.

7. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 6, characterized in that, The inversion processing unit includes: The power estimation subunit is used by the multi-zone stirring and co-control unit to estimate the mechanical work released by the gas as it expands from the nozzle to the surface of the melt, based on thermodynamic principles and real-time pressure and temperature parameters in the molten pool, and to obtain the power value of gas expansion at each nozzle. The kinetic energy estimation subunit is used by the multi-zone stirring and co-control unit to estimate the effective kinetic energy transferred to the surrounding melt by the high-speed jet during the process of breaking into bubbles, based on the jet dynamics theory, and to obtain the jet kinetic energy transfer power value of each vent. The accumulation processing subunit is used to accumulate the power value of gas expansion and the power value of jet kinetic energy transfer at each vent within the same logic control area. The total power value after accumulation is divided by the preset molten pool volume in the logic control area to obtain the local stirring power density value of the logic control area.

8. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 7, characterized in that, The accumulation processing subunit includes: The classification component is used to classify the gas expansion power value and jet kinetic energy transfer power value of each vent according to the preset logical control area classification rules; the multi-zone stirring and co-control unit matches each vent to the corresponding logical control area based on the spatial coordinate range of the feeding area, main reaction area and settling area defined during the nonlinear gun design, forming a list of vents belonging to each logical control area and its corresponding two types of power value datasets. The power accumulation component is used by the multi-zone stirring collaborative control unit to traverse the list of air vents belonging to each logical control zone. For all air vents in the same logical control zone, the power value of gas expansion work done by each air vent in the zone is accumulated together to obtain the total power value of gas expansion work done by the logical control zone. At the same time, the power value of jet kinetic energy transfer from each air vent in the zone is accumulated together to obtain the total power value of jet kinetic energy transfer by the logical control zone. The total power acquisition component is used by the multi-zone stirring and co-control unit to add the total work power value of gas expansion in the same logical control zone to the total transfer power value of jet kinetic energy, and obtain the total input mechanical power value of the logical control zone.

9. The intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace according to claim 8, characterized in that, Total power acquisition components include: The consistency comparison sub-component is used by the multi-zone stirring collaborative control unit to extract two power values ​​corresponding to the same logical control zone from the total power dataset of gas expansion and the total power dataset of jet kinetic energy transfer of the power accumulation component; and to perform consistency comparison based on the logical control zone identifier and timestamp attached to the two power values ​​to confirm that they belong to the same control zone and are in the same control cycle, thus forming a power data pair that has been paired in the logical control zone. The numerical merging sub-component is used by the multi-zone stirring and collaborative control unit to input the paired gas expansion total work power value and the jet kinetic energy transfer total power value into the internal numerical merging module, perform linear superposition calculation of the power values, and obtain the preliminary accumulation result of the input mechanical power of the logic control area. The numerical verification sub-component is used by the multi-zone stirring collaborative control unit to verify the unit dimension of the preliminary accumulation result, confirm that it is consistent with the preset power unit standard, and store the verified value in the total input mechanical power cache of the corresponding logic control area to form the total input mechanical power value for calculating the local stirring power density.

10. A method for intelligent collaborative control of multi-zone stirring in a side-blown molten pool furnace, applied to the intelligent collaborative control system for multi-zone stirring in a side-blown molten pool furnace as described in any one of claims 1 to 9, characterized in that, The intelligent collaborative control method for multi-zone stirring in a side-blown molten pool furnace includes: At the beginning of each control cycle, the multi-zone stirring and co-control unit triggers a data acquisition command to simultaneously acquire fluid dynamics signals, structural response signals, and thermodynamic and process parameters of the bottom operating status data of the melting furnace. By utilizing the bottom-level operating status data of the smelting furnace and combining it with the built-in price reduction molten pool mechanism model, the system identifies standing wave risk, assesses scouring risk, and measures the overall pool status indicators for stirring uniformity. The state machine scheduler determines the dominant mode of the current control cycle based on the preset priority logic and the full pool state index, and calls the response model predictive control algorithm to solve for the actuator setpoint at the next moment, thus obtaining the digital optimal setpoint. The obtained digital optimal setpoint is converted into a physical control signal and sent to the underlying actuator in parallel; flow regulation, flow pattern transformation and pulse establishment are performed; after the instruction is issued, it enters a waiting state until the next control cycle.

Images