Gas-liquid two-phase flow regulation and control method and system for oxygen-enriched side-blown smelting furnace

Abstract

The invention discloses a gas-liquid two-phase flow regulation and control method and system for an oxygen-enriched side-blown smelting furnace, and relates to the technical field of dynamic regulation and control. A high-risk area where a refractory material is peeled off is positioned by obtaining state data of the oxygen-enriched side-blown smelting furnace; determining an associated spray gun according to the high-risk area, performing dynamic game decision on the high-risk stripping area, outputting a control parameter set of the associated spray gun, adjusting gas-liquid two-phase flow of the oxygen-enriched side-blown smelting furnace based on the control parameter set of the spray gun, and outputting a regulation effect according to an adjusted erosion energy drop value and an efficiency drop ratio; if the regulation and control effect is effective, continuously carrying out gas-liquid two-phase flow regulation on the smelting furnace; therefore, by adjusting the parameters of the spray gun of the oxygen-enriched side-blown smelting furnace, the risks of surface peeling and crack propagation of the refractory material of the furnace lining are reduced, and meanwhile, the energy consumption and the maintenance cost are reduced; and the smelting efficiency is improved as much as possible while it is guaranteed that the refractory material is not prone to stripping in the gas-liquid two-phase flow regulation and control process in the molten pool.

Description

Technical Field

[0001] The present invention relates to the field of dynamic control technology, and in particular to a gas-liquid two-phase flow control method and system for an oxygen-enriched side-blown smelting furnace. Background Art

[0002] An oxygen-enriched side-blown smelting furnace is a highly efficient metallurgical device that injects oxygen-enriched air into the molten pool at high speeds through sidewall lances, utilizing gas-liquid two-phase flow to enhance melt mixing and reaction. Its core principle is to accelerate the oxidation reaction of metal sulfides and improve heat transfer efficiency through the intense interaction between oxygen and the melt. Traditional gas-liquid two-phase flow control technology enhances molten pool agitation by adjusting lance parameters (such as increasing oxygen flow rate and adjusting the injection angle) or by adopting a fixed operating mode. This allows for sufficient oxygen and melt contact, forming a dispersed bubble cluster and expanding the gas-liquid reaction interface area, thereby accelerating the oxidation and decomposition of sulfides and the enrichment of the metal phase.

[0003] However, in actual use, the traditional control method has significant defects: although the violent stirring of the gas-liquid two-phase flow in the molten pool can increase the reaction rate, it will form a high-intensity turbulent zone, causing continuous mechanical scouring and thermal stress shock to the furnace lining, that is, lining erosion. The refractory material of the furnace lining is exposed to high-temperature melt and chemical corrosion environment for a long time. Under the repeated impact of gas-liquid flow, it is prone to surface peeling, crack expansion and other problems, which seriously shorten the life of the furnace body; at the same time, lining erosion will also destroy the thermal field stability of the molten pool, resulting in increased local heat dissipation and decreased thermal efficiency, further increasing energy consumption and maintenance costs; therefore, how to ensure that the refractory material is not easy to peel off during the control of the gas-liquid two-phase flow in the molten pool while improving the melting efficiency as much as possible is a problem to be solved. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned problem of how to ensure that the refractory material is not easy to peel off during the gas-liquid two-phase flow control process in the molten pool while improving the smelting efficiency as much as possible, and to provide a gas-liquid two-phase flow control method and system for an oxygen-enriched side-blown smelting furnace.

[0005] In a first aspect of the present invention, a method for controlling gas-liquid two-phase flow in an oxygen-enriched side-blown smelting furnace is first proposed, the method comprising: S1: Acquire status data of the oxygen-enriched side-blown smelting furnace and locate high-risk areas of refractory spalling in the oxygen-enriched side-blown smelting furnace based on the status data; S2: Determine the associated spray guns according to the high-risk areas, perform dynamic game decision-making on the high-risk spalling areas, and output a set of control parameters for the associated spray guns, the set of control parameters including the injection oxygen flow rate and injection angle; S3: adjusting the control parameters of the associated lances based on the lance control parameter set, regulating the gas-liquid two-phase flow of the oxygen-enriched side-blowing smelting furnace, and outputting the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio; S4: Determine whether the control effect is effective. If effective, continue to adjust the gas-liquid two-phase flow of the smelting furnace.

[0006] Optionally, the steps of obtaining status data of the oxygen-enriched side-blowing smelting furnace and locating the risk of refractory spalling of the oxygen-enriched side-blowing smelting furnace according to the status data are: Arrange multiple acoustic emission sensors on the outer wall of the furnace to collect acoustic wave signals in the molten pool; An industrial camera is installed outside the furnace to capture optical bubble images on the molten pool surface and obtain bubble characteristics; All data are timestamped based on the IEEE1588 protocol; The spatial coordinates of the acoustic wave source are determined by triangulation and acoustic wave signals, and the spatial coordinates of the acoustic wave source are spatially mapped with the bubble contours in the optical bubble image to obtain a three-dimensional thermal map of the molten pool. Extract the position of each bubble in the optical bubble image, divide the molten pool surface into several small areas, and count the position distribution of bubbles in each area. Calculate the variance of the bubble position in each area, and use the inverse of each variance as the uniformity of the bubble distribution. The three-dimensional heat map is divided into several regions. For each region, the corresponding regional acoustic wave signal is obtained, and the frequency and amplitude of the acoustic wave signal are extracted through spectrum analysis. The signal energy of the acoustic wave signal is calculated and divided by the regional volume to obtain the erosion energy density. If the bubble distribution uniformity is less than the preset bubble distribution uniformity threshold or the erosion energy is not less than the preset erosion energy density, the corresponding area is defined as a high-risk area for refractory material spalling.

[0007] Optionally, the steps of determining associated spray guns according to high-risk areas and making dynamic game decisions for high-risk spalling areas are as follows: Mark the high-risk areas in the three-dimensional thermal map, identify the high-risk areas, and determine the corresponding three-dimensional coordinates and associated spray guns of the high-risk areas; Coverage association rule: If the center of the high-risk area is within the coverage of spray gun i, then mark spray gun i as an associated spray gun; Construct a game decision model, define the game participants, the attackers, and the attacker strategy: increase the flow rate of the associated spray gun, increase the spray angle, and maximize the smelting efficiency; the payoff function is: , where is the attacking party's profit value, is the mean value of the uniformity of bubble distribution in all regions, is the injection oxygen flow rate of the associated spray gun, is the overall maximum injection oxygen flow rate; and is the weight of the attacker’s payoff function; The defensive strategy of the game participants is to reduce the flow rate of the associated spray gun, reduce the spray angle, and minimize the erosion risk; the payoff function is: , where is the profit value of the defending side, is the erosion energy density, is the maximum erosion energy density; is the weight of the defender’s payoff function; Set constraints: Adjusted oxygen flow rate for the associated spray gun Within the restricted flow rate range, The sum of the spray flow rate of the adjusted associated spray gun and the spray flow rates of the remaining non-associated spray guns is not greater than the preset maximum flow rate; The spray angle of the associated spray gun is adjusted to not more than the maximum adjustable angle; pass The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun.

[0008] Optionally, by The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun in the following steps: Population initialization and constraint filtering, input the injection oxygen flow rate and injection angle of the current associated spray gun, the erosion energy density and bubble uniformity of the high-risk area; Based on the current parameters, the initial population is randomly generated, and the infeasible solutions that exceed the flow velocity angle, total flow rate and maximum adjustable angle constraints are filtered out to obtain the feasible solution set; Calculate the attacking benefit of the solution by non-dominated sorting of the feasible solution set and defender's gain , hierarchical sorting by non-dominated relations, hierarchical sorting by non-dominated relations, and taking the first layer of the hierarchical solution set as the optimal frontier; And process the optimal frontier solution to obtain the attacking party's benefits and defender's benefits Perform weighted summation to obtain the target value, normalize the target value, generate a uniform reference point, perform crossover mutation to generate offspring, and output a new population; For the new population, the solutions with excessive total flow are eliminated, the elite solutions are retained, the Pareto solution set is output, and the solutions are selected according to the priority of the Pareto solution set, and the injection oxygen flow rate and injection angle of the associated spray gun are output.

[0009] Optionally, the step of outputting the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio is: Obtain the overall erosion energy of the oxygen-enriched side-blown smelting furnace after regulation and the erosion energy of the high-risk area, and subtract the overall erosion energy before regulation from the overall erosion energy after regulation to obtain an overall erosion energy reduction value, and divide the overall erosion energy reduction value by the overall erosion energy before regulation to obtain an overall regulation effective value; The erosion energy of the high-risk area before regulation is subtracted from the erosion energy of the high-risk area after regulation to obtain the erosion energy reduction value of the high-risk area, and the erosion energy reduction value is divided by the overall erosion energy before regulation to obtain the effective value of regulation in the high-risk area; Add the overall control effective value and the high-risk area control effective value to obtain the erosion energy reduction value; Obtaining the efficiency of the oxygen-enriched side-blowing smelting furnace before and after the regulation, subtracting the efficiency of the oxygen-enriched side-blowing smelting furnace after the regulation from the efficiency of the oxygen-enriched side-blowing smelting furnace before the regulation to obtain an efficiency decrease value, and dividing the efficiency decrease value by the efficiency of the oxygen-enriched side-blowing smelting furnace before the regulation to obtain an efficiency decrease ratio; If the erosion energy reduction value is not less than the preset erosion energy reduction value threshold, and the efficiency reduction ratio is less than the preset efficiency reduction ratio threshold, it indicates that the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously regulated according to the injection oxygen flow rate and injection angle of the regulated associated spray gun; If the erosion energy decrease value is less than the preset erosion energy decrease value threshold or the efficiency decrease ratio is less than the preset efficiency decrease ratio threshold, it means that the control effect is low, and the S2-S3 dynamic game decision is performed again, and the new injection oxygen flow rate and injection angle of the associated spray gun are output until the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted.

[0010] In a second aspect of the present invention, a gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace is provided, the system comprising: Spalling risk module: obtains status data of the oxygen-enriched side-blown smelting furnace and locates high-risk areas of refractory spalling in the oxygen-enriched side-blown smelting furnace based on the status data; Game decision module: determines the associated spray gun according to the high-risk area, performs dynamic game decision on the high-risk spalling area, and outputs a set of control parameters for the associated spray gun, the control parameter set including the injection oxygen flow rate and injection angle; Regulation module: This module adjusts the control parameters of the associated lances based on the lance control parameter set, regulates the gas-liquid two-phase flow of the oxygen-enriched side-blown smelting furnace, and outputs the regulation effect based on the erosion energy reduction value and efficiency reduction ratio after adjustment; Continuous control module: determines whether the control effect is effective. If it is effective, it continuously adjusts the gas-liquid two-phase flow of the smelting furnace.

[0011] Optionally, the spalling risk module includes: Acoustic signal acquisition module: multiple acoustic emission sensors are arranged on the outer wall of the furnace to collect acoustic signals in the molten pool; Image acquisition module: An industrial camera is installed outside the furnace to capture optical bubble images on the molten pool surface and obtain bubble characteristics; Data alignment module: aligns timestamps of all data based on the IEEE1588 protocol; 3D thermal map module: The spatial coordinates of the acoustic wave source are determined by triangulation and acoustic wave signals, and the spatial coordinates of the acoustic wave source are spatially mapped with the bubble contours in the optical image to obtain a 3D thermal map of the molten pool. Distribution uniformity module: extracts the position of each bubble in the optical bubble image, divides the molten pool surface into several small areas, and counts the position distribution of bubbles in each area. Calculates the variance of the bubble position in each area, and uses the inverse of each variance as the uniformity of the bubble distribution. Erosion energy density module: Divide the 3D heat map into several regions. For each region, obtain the corresponding regional acoustic signal, extract the frequency and amplitude of the acoustic signal through spectrum analysis, calculate the signal energy of the acoustic signal, and divide the signal energy by the regional volume to obtain the erosion energy density. High-risk area positioning module: If the bubble distribution uniformity is less than the preset bubble distribution uniformity threshold or the erosion energy is not less than the preset erosion energy density, the corresponding area is defined as a high-risk area for refractory spalling.

[0012] Optionally, the game decision module includes Associated spray gun module: Marks high-risk areas in the three-dimensional thermal map, identifies the high-risk areas, and determines the corresponding three-dimensional coordinates and associated spray guns of the high-risk areas; Coverage association rule module: Coverage association rule: If the center of the high-risk area is within the coverage of spray gun i, then mark spray gun i as an associated spray gun; Attacker module: Constructs a game decision model, defines the game participants and their attackers, and the attacker strategy is to increase the flow rate of the associated spray gun, increase the spray angle, and maximize the smelting efficiency; the profit function is: , where is the attacking party's profit value, is the mean value of the uniformity of bubble distribution in all regions, is the injection oxygen flow rate of the associated spray gun, is the overall maximum injection oxygen flow rate; and is the weight of the attacker’s payoff function; Defending module: The defending side of the game participant. The defending strategy is to reduce the flow rate of the associated spray gun, reduce the spray angle, and minimize the erosion risk. The payoff function is: , where is the profit value of the defending side, is the erosion energy density, is the maximum erosion energy density; is the weight of the defender’s payoff function; First constraint module: adjusted oxygen flow rate of the associated spray gun Within the restricted flow rate range, Second constraint condition module: the sum of the spray flow rate of the adjusted associated spray gun and the spray flow rates of the remaining non-associated spray guns is not greater than the preset maximum flow rate; The third constraint condition module: the adjustment angle of the spray angle of the associated spray gun is not greater than the maximum adjustable angle; Solving module: through The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun.

[0013] Optionally, the solution module includes: Initialization module: population initialization and constraint filtering, inputting the currently associated injection oxygen flow rate and injection angle, erosion energy density in high-risk areas, and bubble uniformity; Screening module: Based on the current parameters, it randomly generates the initial population, filters out the infeasible solutions that exceed the flow velocity angle, total flow rate and maximum adjustable angle constraints, and obtains the feasible solution set; Non-dominated sorting module: sorts the feasible solution set by non-dominated sorting and calculates the attacking benefit of the solution and defender's gain , hierarchical sorting by non-dominated relations, hierarchical sorting by non-dominated relations, and taking the first layer of the hierarchical solution set as the optimal frontier; New population module: process the optimal frontier solution and convert the attacking party's benefits and defender's benefits Perform weighted summation to obtain the target value, normalize the target value, generate a uniform reference point, perform crossover mutation to generate offspring, and output a new population; Control parameter module: eliminate the total flow exceeding limit solutions for the new population, retain the elite solutions, output the Pareto solution set, select the solutions according to the priority of the Pareto solution set, and output the injection oxygen flow rate and injection angle of the associated spray gun.

[0014] Optionally, the adjustment module further includes: Overall control effective value module: obtains the overall erosion energy of the oxygen-enriched side-blown smelting furnace after control and the erosion energy of the high-risk area, and subtracts the overall erosion energy before control from the overall erosion energy after control to obtain the overall erosion energy reduction value, and then divides the overall erosion energy reduction value by the overall erosion energy before control to obtain the overall control effective value; Regional control effective value module: subtract the erosion energy of the high-risk area before control from the erosion energy of the high-risk area after control to obtain the erosion energy reduction value of the high-risk area, and divide the erosion energy reduction value by the overall erosion energy before control to obtain the effective value of the high-risk area control; Erosion energy reduction value module: Add the overall control effective value and the high-risk area control effective value to obtain the erosion energy reduction value; Efficiency reduction ratio module: obtains the efficiency of the oxygen-enriched side-blowing melting furnace before and after regulation, and subtracts the efficiency of the oxygen-enriched side-blowing melting furnace after regulation from the efficiency of the oxygen-enriched side-blowing melting furnace before regulation to obtain the efficiency reduction value. The efficiency reduction value is divided by the efficiency of the oxygen-enriched side-blowing melting furnace before regulation to obtain the efficiency reduction ratio. Continuous control module: If the erosion energy reduction value is not less than the preset erosion energy reduction value threshold, and the efficiency reduction ratio is less than the preset efficiency reduction ratio threshold, it means that the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted according to the injection oxygen flow rate and injection angle of the regulated associated spray gun; Readjustment module: If the erosion energy decrease value is less than the preset erosion energy decrease value threshold or the efficiency decrease ratio is less than the preset efficiency decrease ratio threshold, it means that the control effect is low, and the dynamic game decision of the game decision module and the adjustment module is re-performed, and the new injection oxygen flow rate and injection angle of the associated spray gun are output until the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted.

[0015] Beneficial effects of the present invention: The present invention proposes a gas-liquid two-phase flow control method and system for an oxygen-enriched side-blown smelting furnace, which obtains status data of the oxygen-enriched side-blown smelting furnace and locates the risk of spalling of refractory materials in the oxygen-enriched side-blown smelting furnace according to the status data; performs dynamic game decision-making according to the risk of refractory spalling, and outputs a set of spray gun control parameters, wherein the control parameter set includes flow rate and angle; performs spray gun parameter control based on the spray gun control parameter set, adjusts the gas-liquid two-phase flow of the oxygen-enriched side-blown smelting furnace, and outputs the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio; if the control effect is effective, the gas-liquid two-phase flow of the smelting furnace is continuously adjusted; in this way, by adjusting the spray gun parameters (oxygen flow rate and blowing angle) of the oxygen-enriched side-blown smelting furnace, the risk of surface spalling and crack extension of the refractory material of the furnace lining is reduced, while reducing energy consumption and maintenance costs; while ensuring that the refractory material is not easy to spall during the gas-liquid two-phase flow control process in the molten pool, the smelting efficiency is improved as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1 This is a flow chart of a gas-liquid two-phase flow control method for an oxygen-enriched side-blown smelting furnace; Figure 2 This is a framework diagram of the gas-liquid two-phase flow control system of an oxygen-enriched side-blown smelting furnace. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0019] The embodiment of the present invention provides a method for controlling gas-liquid two-phase flow in an oxygen-enriched side-blown smelting furnace. Figure 1 , Figure 1 A flow chart of a method for controlling gas-liquid two-phase flow in an oxygen-enriched side-blown smelting furnace according to an embodiment of the present invention. The method comprises the following steps: S1: Acquire status data of the oxygen-enriched side-blown smelting furnace and locate high-risk areas of refractory spalling in the oxygen-enriched side-blown smelting furnace based on the status data; S2: Determine the associated spray guns according to the high-risk areas, perform dynamic game decision-making on the high-risk spalling areas, and output a set of control parameters for the associated spray guns, the set of control parameters including the injection oxygen flow rate and injection angle; S3: adjusting the control parameters of the associated lances based on the lance control parameter set, regulating the gas-liquid two-phase flow of the oxygen-enriched side-blowing smelting furnace, and outputting the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio; S4: Determine whether the control effect is effective. If effective, continue to adjust the gas-liquid two-phase flow of the smelting furnace.

[0020] Based on a gas-liquid two-phase flow control method for an oxygen-enriched side-blown smelting furnace provided in an embodiment of the present invention, the above-mentioned method adjusts the lance parameters (oxygen flow rate and blowing angle) of the oxygen-enriched side-blown smelting furnace to reduce the risk of surface peeling and crack propagation of the refractory material of the furnace lining, while reducing energy consumption and maintenance costs; while ensuring that the refractory material is not easy to peel off during the gas-liquid two-phase flow control process in the molten pool, the smelting efficiency is improved as much as possible.

[0021] In one embodiment, the steps of obtaining status data of an oxygen-enriched side-blowing smelting furnace and locating the risk of refractory spalling of the oxygen-enriched side-blowing smelting furnace according to the status data are as follows: Multiple acoustic emission sensors are arranged on the outer wall of the furnace. The frequency response range of the acoustic emission sensors is 200kHz to 2MHz to collect the acoustic wave signals in the molten pool; An industrial camera is installed outside the furnace to capture optical bubble images on the molten pool surface and obtain bubble characteristics; All data are timestamped based on the IEEE1588 protocol; The spatial coordinates of the acoustic wave source are determined by triangulation and acoustic wave signals, and the spatial coordinates of the acoustic wave source are spatially mapped with the bubble contours in the optical bubble image to obtain a three-dimensional thermal map of the molten pool. The position of each bubble in the optical bubble image is extracted through image processing technology, and the molten pool surface is divided into several small areas. The position distribution of the bubbles in each area is statistically analyzed, and the variance of the bubble position in each area is calculated. The inverse of each variance is used as the uniformity of the bubble distribution. The three-dimensional heat map is divided into several regions. For each region, the corresponding regional acoustic wave signal is obtained, and the frequency and amplitude of the acoustic wave signal are extracted through spectrum analysis. The signal energy of the acoustic wave signal is calculated and divided by the regional volume to obtain the erosion energy density. The bubble distribution uniformity is compared with the preset bubble distribution uniformity threshold, and the erosion energy is compared with the preset erosion energy density. If the bubble distribution uniformity is less than the preset bubble distribution uniformity threshold or the erosion energy is not less than the preset erosion energy density, the corresponding area is defined as a high-risk area for refractory spalling.

[0022] It is important to note that sound source localization (triangulation): The location of a sound source can be calculated using triangulation (or time-difference-based localization) based on the differences in signal strength received from multiple sensors. For example, geometric methods can be used to determine the coordinates of the sound source based on the time difference in sound waves reaching each sensor. Bubble Image Processing (OpenCV): The OpenCV library is used to process camera images and extract bubble outlines through threshold segmentation (for example, setting a threshold of 0.5). This allows the shape, size, and distribution density of bubbles within the melt pool to be identified. Time synchronization is performed using the IEEE 1588 protocol to ensure precise alignment of all data collected by the acoustic emission sensor and the industrial camera. This step is crucial because the flow field, bubble distribution, and acoustic wave signals within the melt pool change in real time, and only precise time synchronization ensures the accuracy of subsequent analysis. 3D Heat Map Generation: Once the 2D projection image of the bubble distribution and the 3D coordinate data of the sound source are obtained, they can be combined with spatial mapping methods to create a 3D heat map, showing the bubble density distribution within the melt pool and the corresponding locations of erosion events. The 3D heat map can highlight areas of high bubble density, indicating potential hotspots of lining erosion.

[0023] It should be noted that multiple acoustic emission sensors with a frequency response range of 200kHz to 2MHz are positioned on the outer wall of the furnace. These sensors capture acoustic signals generated by the intense collision or reaction between the high-temperature melt and the strong airflow in the molten pool. Variations in the intensity of the acoustic signals can reveal dramatic fluctuations in the airflow within the molten pool. Abnormally strong acoustic signals in a particular area typically indicate overly concentrated airflow, potentially causing significant mechanical impact on the refractory material and thus affecting its durability. Secondly, an industrial camera is installed outside the furnace to capture optical bubble images of the molten pool surface. The shape, size, and distribution density of the bubbles reflect the intensity and uniformity of the gas-liquid flow within the molten pool. Excessively concentrated or uneven bubble distribution indicates strong localized turbulence within the molten pool, which can easily impact the refractory surface. This intense airflow can cause spalling and crack propagation. To ensure data synchronization, all data acquisition is time-stamped using the IEEE 1588 protocol to ensure precise temporal correspondence between the acoustic signals and bubble images. Next, triangulation (or time-difference-based positioning) is used to infer the spatial coordinates of the acoustic source based on the intensity differences in the acoustic signals received by multiple sensors. This allows the identification of hotspots of gas-liquid flow and reaction within the melt pool. The spatial coordinates of the acoustic source are then spatially mapped to the bubble distribution in the optical image to generate a three-dimensional thermal map of the melt pool, showing the distribution of bubble density and thermal effects in different regions. These thermal maps clearly identify areas of high bubble density within the melt pool and highlight potential hotspots of lining erosion.

[0024] Image processing techniques are used to extract bubble positions from the optical bubble image. The uniformity of bubble distribution is assessed based on the variance of the bubble positions. A larger variance indicates a more uneven bubble distribution and a stronger localized airflow impact, which can exacerbate refractory wear. Spectral analysis extracts the frequency and amplitude of the acoustic signal within a region. The signal energy in that region is calculated and divided by the volume to obtain the erosion energy density. Regions with higher erosion energy density typically experience stronger airflow impact and reaction, resulting in greater thermal stress and more severe erosion of the furnace lining. The uniformity of bubble distribution and erosion energy density are compared with pre-set thresholds. Uneven bubble distribution or excessively high erosion energy density indicate a high risk of refractory spalling in that region, resulting in a high-risk state. Otherwise, a low-risk state is entered. Based on these risk assessment results, dynamic game decisions are made to adjust lance control parameters (such as flow rate and injection angle) to optimize gas-liquid flow within the melt pool, minimize impact on the furnace lining, and reduce the risk of spalling.

[0025] For example, if acoustic signal analysis reveals abnormal acoustic intensity in a certain area, with concentrated bubbles and high erosion energy density, this indicates excessive airflow in that area, which could impact the furnace lining significantly and accelerate refractory spalling. The system automatically adjusts the spray gun parameters to reduce airflow impact in that area, minimizing the risk of spalling and extending the furnace's service life.

[0026] In one implementation, this multidimensional data analysis based on acoustic signals, bubble distribution, thermal maps, and erosion energy density can assess the risk of refractory spalling in various areas of the molten pool in real time, and take effective measures to optimize and control it, thereby improving smelting efficiency and reducing maintenance costs.

[0027] In one embodiment, the steps of determining the associated spray guns according to the high-risk areas and making dynamic game decisions for the high-risk spalling areas are as follows: Mark the high-risk areas in the three-dimensional thermal map, identify the high-risk areas, and determine the corresponding three-dimensional coordinates and associated spray guns of the high-risk areas; Coverage association rule: If the center of the high-risk area is within the coverage of spray gun i, then mark spray gun i as an associated spray gun; Construct a game decision model, define the game participants, the attackers, and the attacker strategy: increase the flow rate of the associated spray gun, increase the spray angle, and maximize the smelting efficiency; the payoff function is: , where is the attacking party's profit value, is the mean value of the uniformity of bubble distribution in all regions, is the injection oxygen flow rate of the associated spray gun, is the overall maximum injection oxygen flow rate; and is the weight of the attacker’s payoff function; The defensive strategy of the game participants is to reduce the flow rate of the associated spray gun, reduce the spray angle, and minimize the erosion risk; the payoff function is: , where is the profit value of the defending side, is the erosion energy density, is the maximum erosion energy density; is the weight of the defender’s payoff function; Set constraints: Adjusted oxygen flow rate for the associated spray gun Within the restricted flow rate range, The sum of the spray flow rate of the adjusted associated spray gun and the spray flow rates of the remaining non-associated spray guns is not greater than the preset maximum flow rate; The spray angle of the associated spray gun is adjusted to not more than the maximum adjustable angle; pass The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun.

[0028] It should be noted that the weights of the attacking and defending side's reward functions are typically set based on the goals and priorities of the actual smelting process. The attacking side's reward function weights typically focus on improving smelting efficiency, and therefore may assign greater weight to efficiency-related indicators such as bubble distribution uniformity and injection flow rate. The defending side's reward function weights, on the other hand, focus more on reducing erosion risk, and therefore assign a higher weight to erosion energy density. The specific weight values depend on the equipment's operating needs and protection requirements. Typically, the attacking side's weight may be between 0.5 and 0.7, while the defending side's weight may be between 0.3 and 0.5. These weights can be adjusted through experiments or historical data to achieve a balance between smelting efficiency and equipment protection.

[0029] Reducing the flow rate and injection angle of the associated lance can reduce the risk of erosion because changes in the injection flow rate and angle directly affect the turbulence intensity of the gas-liquid two-phase flow and the thermodynamic environment of the melt pool. When the flow rate is too high or the injection angle is too large, bubbles within the melt pool will generate intense turbulence in the melt. This not only increases the mixing and reaction rates of the melt, but also causes greater mechanical shock and thermal stress on the furnace lining surface, further exacerbating spalling and damage to the lining material. If the flow rate is low and the injection angle is small, the bubbles injected by the lance will be distributed more slowly into the melt pool, reducing the turbulence intensity within the melt pool and ensuring smoother gas-liquid flow within the melt pool, which helps reduce erosion on the furnace lining and extend its service life. Conversely, increasing the flow rate and injection angle of the associated lance can improve melting efficiency because, in oxygen-enriched side-blown furnaces, higher flow rates and larger injection angles enhance the dispersion of bubbles in the melt pool and the contact area between the melt and the gas. This not only increases the reaction rate between oxygen and metal sulfides, accelerates the oxidation of sulfides and the enrichment of metal phases, but also promotes heat transfer within the molten pool, increasing the temperature and reaction efficiency of the melt. Through stronger stirring, the substances in the melt can be more evenly distributed, thereby improving the overall smelting efficiency. However, although this strategy of increasing the flow rate and injection angle can improve the smelting efficiency, it is also accompanied by the risk of increasing the turbulence intensity in the molten pool, thereby exacerbating the erosion of the furnace lining. Therefore, in the game decision model, it is necessary to weigh the relationship between smelting efficiency and furnace lining protection, and achieve optimization by reasonably adjusting the flow rate and injection angle.

[0030] It should be noted that in the dynamic game decision-making process based on refractory spalling risk, high-risk areas must first be identified by analyzing the 3D thermal map. High-risk areas are typically characterized by uneven bubble distribution and high erosion energy density within the melt pool. Therefore, the 3D coordinates of these areas must be precisely determined, and the associated spray guns must be labeled. For example, if a high-risk area falls within the coverage area of spray gun i, then spray gun i is labeled as an "associated spray gun." Next, a game decision-making model is constructed, defining the players. The attacker's goal is to increase the flow rate and spray angle of the associated spray gun to maximize melting efficiency. This means adjusting spray gun parameters to increase the oxygen supply to the melt pool, promote a full reaction between the metal and the gas, and increase the melting rate. The attacker's reward function is calculated based on the relationship between bubble distribution uniformity and the spray gun flow rate and the maximum flow rate. The defender, on the other hand, aims to reduce the risk of erosion on the refractory by reducing the flow rate and spray angle, thereby avoiding excessive erosion energy density and extending the service life of the furnace lining. Therefore, the defender's reward function focuses on reducing the erosion energy density and keeping it within a maximum value. The model also needs to set some constraints, such as the gun flow rate must be within a certain range, the sum of the injection flow and other non-related gun flows cannot exceed the preset maximum flow, and the adjustment of the injection angle cannot exceed the maximum adjustable angle. These conditions ensure that the adjustment of the gun parameters will not cause other risks. The algorithm solves the game decision model and can output the optimal injection flow rate and angle, thereby finding a balance between melting efficiency and refractory protection.

[0031] For example, suppose that during a certain smelting process, the three-dimensional thermal map shows that the bubbles in a certain area of the molten pool are unevenly distributed and the erosion energy density is high, and it is marked as a high-risk area. This area is within the coverage of spray gun i, so spray gun i is marked as an associated spray gun. In the game decision-making, the attacking strategy is to increase the spray gun flow rate and spray angle in an attempt to improve the smelting efficiency, but this may also increase the erosion risk in this area. The defending side tends to reduce the spray gun flow rate and angle to reduce the erosion of the furnace lining. The algorithm,the model finally outputs the optimal injection flow rate and injection angle,which should be adopted by the spray gun i to ensure that the erosion risk is minimized,and the refractory spalling in the high-risk area is avoided while the melting efficiency is improved.

[0032] In one implementation, dynamic game-based decision-making outputs link the spray gun's jet flow rate and spray angle, finding a balance between multiple objectives in the smelting process, ensuring both improved smelting efficiency and effective control of refractory erosion risks. The dynamic game model considers the opposing strategies of attack (optimizing smelting efficiency) and defense (reducing erosion risk). By exploring the strategy space, it helps find the optimal spray parameter settings, thus avoiding the side effects of simply optimizing a single objective. For example, simply increasing the flow rate and spray angle may improve efficiency, but it also increases the risk of refractory damage and spalling. However, through game-based decision-making, it is possible to maximize smelting efficiency while avoiding excessive erosion, extending equipment life, and reducing maintenance costs, thereby achieving a win-win situation in terms of economic benefits and safety.

[0033] In one embodiment, by The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun in the following steps: Population initialization and constraint filtering, input the injection oxygen flow rate and injection angle of the current associated spray gun, the erosion energy density and bubble uniformity of the high-risk area; Based on the current parameters, the initial population is randomly generated, and the infeasible solutions that exceed the flow velocity angle, total flow rate and maximum adjustable angle constraints are filtered out to obtain the feasible solution set; Calculate the attacking benefit of the solution by non-dominated sorting of the feasible solution set and defender's gain , hierarchical sorting by non-dominated relations, hierarchical sorting by non-dominated relations, and taking the first layer of the hierarchical solution set as the optimal frontier; And process the optimal frontier solution to obtain the attacking party's benefits and defender's benefits Perform weighted summation to obtain the target value, normalize the target value, generate a uniform reference point, perform crossover mutation to generate offspring, and output a new population; For the new population, the solutions with excessive total flow are eliminated, the elite solutions are retained, the Pareto solution set is output, and the solutions are selected according to the priority of the Pareto solution set, and the injection oxygen flow rate and injection angle of the associated spray gun are output.

[0034] It should be noted that through The algorithm solves the game decision model and outputs the injection flow rate and injection angle of the associated spray gun in a multi-step process. Each step involves detailed optimization and screening to ensure that the final injection parameters can maximize the melting efficiency and reduce the erosion of refractory materials while meeting the requirements of safety and efficiency.

[0035] First of all, population initialization and constraint filtering are the first steps of the algorithm. In order to start optimization, some key parameters need to be input, including the current flow rate and spray angle of the spray gun, as well as the erosion energy density and bubble uniformity in high-risk areas. These parameters are the core factors affecting the operation of the spray gun. In the initial state, the spray flow rate and spray angle of the spray gun are known. Next, the initial population is randomly generated, and each individual represents a possible combination of spray parameters. At this stage, solutions that exceed the constraints will be filtered out. Specifically, the flow rate and angle need to be within a certain range, and the total flow rate cannot exceed the maximum flow rate, and the adjustment range of the spray angle cannot exceed the preset maximum adjustable angle. This process ensures that only solutions that meet the actual operating constraints will enter the subsequent optimization process.

[0036] Then enter the non-dominated sorting stage. In the feasible solution set, Each solution will be sorted and the attacking and defending benefits of each solution will be calculated. The attacking benefit is mainly to optimize the melting efficiency by increasing the flow rate of the spray gun and increasing the injection angle, while the defending benefit is to reduce the risk of erosion of the refractory material by reducing the flow rate of the spray gun and reducing the injection angle. By calculating the attacking and defending benefits, These solutions are sorted by non-dominated order to determine which solutions are closer to the Pareto frontier and which solutions have greater potential optimization value. The Pareto frontier refers to the set of solutions in multi-objective optimization where no other solution is better than it in all objectives and represents the optimal solution.

[0037] After obtaining the non-dominated sorting, the next step is to generate uniform reference points and perform cross-mutation. In this stage, the gains of the attacker and defender are weighted and summed according to the preset weights to calculate the target value of each solution and normalize the target value. The purpose of normalization is to convert different target values to the same scale, making the optimization process more stable. On this basis, Uniform reference points are generated, which are used to guide the generation of the next generation of populations. Crossover and mutation are the core of genetic algorithms. Crossover combines information from two solutions to generate a new solution, while mutation introduces random changes to the current solution, increasing diversity and exploring more potential solution spaces.

[0038] Next, the algorithm performs elite retention and constraint elimination. During this phase, the algorithm removes from the new generation population solutions that violate constraints, particularly those with excessive flow rates or angle adjustments that fall outside the specified range, while retaining excellent solutions that meet the constraints. This phase results in an elite set of solutions, each of which achieves a balanced optimization across various objectives.

[0039] Finally, the algorithm outputs a Pareto solution set. These solutions represent the optimal injection flow rate and injection angle. No single solution in this set is superior to the others in terms of all objectives. Based on the preset priorities, the most appropriate solution is selected as the final injection flow rate and injection angle. These output parameters maximize melting efficiency while avoiding excessive refractory erosion, thereby extending the equipment's service life and reducing maintenance frequency.

[0040] For example, in a high-temperature smelting scenario, the initial injection flow rate is 500L / min and the injection angle is 15 degrees. After optimization, the final output may be that the injection flow rate is adjusted to 480L / min and the injection angle is adjusted to 17 degrees. Through this optimization, the working condition of the spray gun is improved, the melting efficiency is improved, and at the same time, due to the adjustment of the injection angle, excessive erosion energy density is avoided, the damage to the refractory material is reduced, and the service life of the equipment is extended. Through such optimization, the melting efficiency and equipment protection in the production process are balanced, thereby achieving long-term benefit maximization. In one implementation method, by The advantage of the algorithm solving the game decision model and outputting the injection velocity and injection angle of the associated spray gun is that it can simultaneously consider the melting efficiency and the erosion risk of refractory materials within the framework of multi-objective optimization. The algorithm can handle complex constraints and multiple objectives, ensuring that the final injection parameters not only effectively improve smelting efficiency but also minimize the risk of equipment damage and refractory spalling, thereby extending equipment life and improving production process stability. Furthermore, the algorithm provides operators with a more accurate and reliable basis for decision-making, avoiding the limitations of manual experience and helping to achieve automated and intelligent optimization of the production process.

[0041] In one embodiment, the steps of outputting the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio are as follows: Obtain the overall erosion energy of the oxygen-enriched side-blown smelting furnace after regulation and the erosion energy of the high-risk area, and subtract the overall erosion energy before regulation from the overall erosion energy after regulation to obtain an overall erosion energy reduction value, and divide the overall erosion energy reduction value by the overall erosion energy before regulation to obtain an overall regulation effective value; The erosion energy of the high-risk area before regulation is subtracted from the erosion energy of the high-risk area after regulation to obtain the erosion energy reduction value of the high-risk area, and the erosion energy reduction value is divided by the overall erosion energy before regulation to obtain the effective value of regulation in the high-risk area; Add the overall control effective value and the high-risk area control effective value to obtain the erosion energy reduction value; Obtaining the efficiency of the oxygen-enriched side-blowing smelting furnace before and after the regulation, subtracting the efficiency of the oxygen-enriched side-blowing smelting furnace after the regulation from the efficiency of the oxygen-enriched side-blowing smelting furnace before the regulation to obtain an efficiency decrease value, and dividing the efficiency decrease value by the efficiency of the oxygen-enriched side-blowing smelting furnace before the regulation to obtain an efficiency decrease ratio; Compare the erosion energy reduction value with a preset erosion energy reduction value threshold, and compare the efficiency reduction ratio with a preset efficiency reduction ratio threshold. If the erosion energy reduction value is not less than the preset erosion energy reduction value threshold, and the efficiency reduction ratio is less than the preset efficiency reduction ratio threshold, it indicates that the regulation effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously regulated according to the injection oxygen flow rate and injection angle of the associated lance that are regulated; If the erosion energy decrease value is less than the preset erosion energy decrease value threshold or the efficiency decrease ratio is less than the preset efficiency decrease ratio threshold, it means that the control effect is low, and the S2-S3 dynamic game decision is performed again, and the new injection oxygen flow rate and injection angle of the associated spray gun are output until the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted.

[0042] It should be noted that, for example, in the operation of an oxygen-enriched side-blown smelting furnace, assuming that during a certain batch of smelting, the flow rate and injection angle of the spray gun were not optimized, resulting in excessive erosion energy in certain high-risk areas of the furnace and low smelting efficiency. Through the dynamic game decision model, the flow rate and angle of the spray gun were adjusted to reduce the erosion energy. However, the results after preliminary regulation showed that the reduction in erosion energy has not reached the expected level, and the efficiency reduction ratio has exceeded the set threshold. At this time, re-through The algorithm optimized the spray gun parameters and adjusted them again to ensure that the reduction in erosion energy exceeded a preset threshold while keeping the efficiency drop within a reasonable range. After several adjustments, an effective control state was finally achieved, effectively reducing erosion energy in high-risk areas while ensuring that smelting efficiency was not excessively affected.

[0043] In one implementation, this method of continuously adjusting and optimizing spray gun parameters effectively reduces the risk of equipment damage, improves production efficiency, and avoids resource waste or equipment failure caused by excessive control. Furthermore, this control method offers great flexibility, allowing adjustments based on varying production conditions and real-time feedback, thereby optimizing every step of the production process and improving overall production stability and economic efficiency.

[0044] Based on the same inventive concept, the present invention also provides a gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace. Figure 2 , Figure 2 A framework diagram of a gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace provided in an embodiment of the present invention, the system comprising: Spalling risk module: obtains status data of the oxygen-enriched side-blown smelting furnace and locates high-risk areas of refractory spalling in the oxygen-enriched side-blown smelting furnace based on the status data; Game decision module: determines the associated spray gun according to the high-risk area, performs dynamic game decision on the high-risk spalling area, and outputs a set of control parameters for the associated spray gun, the control parameter set including the injection oxygen flow rate and injection angle; Regulation module: This module adjusts the control parameters of the associated lances based on the lance control parameter set, regulates the gas-liquid two-phase flow of the oxygen-enriched side-blown smelting furnace, and outputs the regulation effect based on the erosion energy reduction value and efficiency reduction ratio after adjustment; Continuous control module: determines whether the control effect is effective. If it is effective, it continuously adjusts the gas-liquid two-phase flow of the smelting furnace.

[0045] Based on a gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace provided in an embodiment of the present invention, the above-mentioned method adjusts the lance parameters (oxygen flow rate and blowing angle) of the oxygen-enriched side-blown smelting furnace to reduce the risk of surface spalling and crack propagation of the refractory material of the furnace lining, while reducing energy consumption and maintenance costs; while ensuring that the refractory material is not easy to spall during the gas-liquid two-phase flow control process in the molten pool, the smelting efficiency is improved as much as possible.

[0046] In one embodiment, the spalling risk module includes: Acoustic signal acquisition module: multiple acoustic emission sensors are arranged on the outer wall of the furnace to collect acoustic signals in the molten pool; Image acquisition module: An industrial camera is installed outside the furnace to capture optical bubble images on the molten pool surface and obtain bubble characteristics; Data alignment module: aligns timestamps of all data based on the IEEE1588 protocol; 3D thermal map module: The spatial coordinates of the acoustic wave source are determined by triangulation and acoustic wave signals, and the spatial coordinates of the acoustic wave source are spatially mapped with the bubble contours in the optical image to obtain a 3D thermal map of the molten pool. Distribution uniformity module: extracts the position of each bubble in the optical bubble image, divides the molten pool surface into several small areas, and counts the position distribution of bubbles in each area. Calculates the variance of the bubble position in each area, and uses the inverse of each variance as the uniformity of the bubble distribution. Erosion energy density module: Divide the 3D heat map into several regions. For each region, obtain the corresponding regional acoustic signal, extract the frequency and amplitude of the acoustic signal through spectrum analysis, calculate the signal energy of the acoustic signal, and divide the signal energy by the regional volume to obtain the erosion energy density. High-risk area positioning module: If the bubble distribution uniformity is less than the preset bubble distribution uniformity threshold or the erosion energy is not less than the preset erosion energy density, the corresponding area is defined as a high-risk area for refractory spalling.

[0047] In one embodiment, the game decision module includes Associated spray gun module: Marks high-risk areas in the three-dimensional thermal map, identifies the high-risk areas, and determines the corresponding three-dimensional coordinates and associated spray guns of the high-risk areas; Coverage association rule module: Coverage association rule: If the center of the high-risk area is within the coverage of spray gun i, then mark spray gun i as an associated spray gun; Attacker module: Constructs a game decision model, defines the game participants and their attackers, and the attacker strategy is to increase the flow rate of the associated spray gun, increase the spray angle, and maximize the smelting efficiency; the profit function is: , where is the attacking party's profit value, is the mean value of the uniformity of bubble distribution in all regions, is the injection oxygen flow rate of the associated spray gun, is the overall maximum injection oxygen flow rate; and is the weight of the attacker’s payoff function; Defending module: The defending side of the game participant. The defending strategy is to reduce the flow rate of the associated spray gun, reduce the spray angle, and minimize the erosion risk. The payoff function is: , where is the profit value of the defending side, is the erosion energy density, is the maximum erosion energy density; is the weight of the defender’s payoff function; First constraint module: adjusted oxygen flow rate of the associated spray gun Within the restricted flow rate range, Second constraint condition module: the sum of the spray flow rate of the adjusted associated spray gun and the spray flow rates of the remaining non-associated spray guns is not greater than the preset maximum flow rate; The third constraint condition module: the adjustment angle of the spray angle of the associated spray gun is not greater than the maximum adjustable angle; Solving module: through The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun.

[0048] In one embodiment, the solution module includes: Initialization module: population initialization and constraint filtering, inputting the currently associated injection oxygen flow rate and injection angle, erosion energy density in high-risk areas, and bubble uniformity; Screening module: Based on the current parameters, it randomly generates the initial population, filters out the infeasible solutions that exceed the flow velocity angle, total flow rate and maximum adjustable angle constraints, and obtains the feasible solution set; Non-dominated sorting module: sorts the feasible solution set by non-dominated sorting and calculates the attacking benefit of the solution and defender's gain , hierarchical sorting by non-dominated relations, hierarchical sorting by non-dominated relations, and taking the first layer of the hierarchical solution set as the optimal frontier; New population module: process the optimal frontier solution and convert the attacking party's benefits and defender's benefits Perform weighted summation to obtain the target value, normalize the target value, generate a uniform reference point, perform crossover mutation to generate offspring, and output a new population; Control parameter module: eliminate the total flow exceeding limit solutions for the new population, retain the elite solutions, output the Pareto solution set, select the solutions according to the priority of the Pareto solution set, and output the injection oxygen flow rate and injection angle of the associated spray gun.

[0049] In one embodiment, the adjustment module further includes: Overall control effective value module: obtains the overall erosion energy of the oxygen-enriched side-blown smelting furnace after control and the erosion energy of the high-risk area, and subtracts the overall erosion energy before control from the overall erosion energy after control to obtain the overall erosion energy reduction value, and then divides the overall erosion energy reduction value by the overall erosion energy before control to obtain the overall control effective value; Regional control effective value module: subtract the erosion energy of the high-risk area before control from the erosion energy of the high-risk area after control to obtain the erosion energy reduction value of the high-risk area, and divide the erosion energy reduction value by the overall erosion energy before control to obtain the effective value of the high-risk area control; Erosion energy reduction value module: Add the overall control effective value and the high-risk area control effective value to obtain the erosion energy reduction value; Efficiency reduction ratio module: obtains the efficiency of the oxygen-enriched side-blowing melting furnace before and after regulation, and subtracts the efficiency of the oxygen-enriched side-blowing melting furnace after regulation from the efficiency of the oxygen-enriched side-blowing melting furnace before regulation to obtain the efficiency reduction value. The efficiency reduction value is divided by the efficiency of the oxygen-enriched side-blowing melting furnace before regulation to obtain the efficiency reduction ratio. Continuous control module: If the erosion energy reduction value is not less than the preset erosion energy reduction value threshold, and the efficiency reduction ratio is less than the preset efficiency reduction ratio threshold, it means that the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted according to the injection oxygen flow rate and injection angle of the regulated associated spray gun; Readjustment module: If the erosion energy decrease value is less than the preset erosion energy decrease value threshold or the efficiency decrease ratio is less than the preset efficiency decrease ratio threshold, it means that the control effect is low, and the dynamic game decision of the game decision module and the adjustment module is re-performed, and the new injection oxygen flow rate and injection angle of the associated spray gun are output until the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted.

[0050] The above is a detailed description of an embodiment of the present invention. However, the content is only a preferred embodiment of the present invention and should not be used to artificially limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for controlling gas-liquid two-phase flow in an oxygen-enriched side-blown smelting furnace, characterized in that: The following steps are involved: S1: Acquire status data of the oxygen-enriched side-blown smelting furnace and locate high-risk areas of refractory spalling in the oxygen-enriched side-blown smelting furnace based on the status data; S2: Determine the associated spray guns according to the high-risk areas, perform dynamic game decision-making on the high-risk spalling areas, and output a set of control parameters for the associated spray guns, the set of control parameters including the injection oxygen flow rate and injection angle; S3: adjusting the control parameters of the associated lances based on the lance control parameter set, regulating the gas-liquid two-phase flow of the oxygen-enriched side-blowing smelting furnace, and outputting the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio; S4: Determine whether the control effect is effective. If effective, continue to adjust the gas-liquid two-phase flow of the smelting furnace.

2. The gas-liquid two-phase flow control method of an oxygen-enriched side-blown smelting furnace according to claim 1, characterized in that: The steps for obtaining status data of an oxygen-enriched side-blown smelting furnace and locating high-risk areas of refractory spalling of the oxygen-enriched side-blown smelting furnace based on the status data are as follows: Arrange multiple acoustic emission sensors on the outer wall of the furnace to collect acoustic wave signals in the molten pool; An industrial camera is installed outside the furnace to capture optical bubble images on the molten pool surface and obtain bubble characteristics; All data are timestamped based on the IEEE1588 protocol; The spatial coordinates of the acoustic wave source are determined by triangulation and acoustic wave signals, and the spatial coordinates of the acoustic wave source are spatially mapped with the bubble contours in the optical bubble image to obtain a three-dimensional thermal map of the molten pool. Extract the position of each bubble in the optical bubble image, divide the molten pool surface into several small areas, and count the position distribution of bubbles in each area. Calculate the variance of the bubble position in each area, and use the inverse of each variance as the uniformity of the bubble distribution. The three-dimensional heat map is divided into several regions. For each region, the corresponding regional acoustic signal is obtained, and the frequency and amplitude of the acoustic signal are extracted through spectrum analysis. The signal energy of the acoustic signal is calculated and divided by the regional volume to obtain the erosion energy density. If the bubble distribution uniformity is less than the preset bubble distribution uniformity threshold or the erosion energy is not less than the preset erosion energy density, the corresponding area is defined as a high-risk area for refractory material spalling.

3. The gas-liquid two-phase flow control method for an oxygen-enriched side-blown smelting furnace according to claim 2, characterized in that: The steps for determining the associated spray guns based on the high-risk areas and making dynamic game decisions for the high-risk spalling areas are as follows: Mark the high-risk areas in the three-dimensional thermal map, identify the high-risk areas, and determine the corresponding three-dimensional coordinates and associated spray guns of the high-risk areas; Coverage association rule: If the center of the high-risk area is within the coverage of spray gun i, then mark spray gun i as an associated spray gun; Construct a game decision model, define the game participants, the attackers, and the attacker strategy: increase the flow rate of the associated spray gun, increase the spray angle, and maximize the smelting efficiency; the payoff function is: , where is the attacking party's profit value, is the mean value of the uniformity of bubble distribution in all regions, is the injection oxygen flow rate of the associated spray gun, is the overall maximum injection oxygen flow rate; is the weight of the attacker’s payoff function; The defensive strategy of the game participants is to reduce the flow rate of the associated spray gun, reduce the spray angle, and minimize the erosion risk; the payoff function is: , where is the profit value of the defender, is the erosion energy density, is the maximum erosion energy density; is the weight of the defender’s payoff function; Set constraints: Adjusted oxygen flow rate for the associated spray gun Within the restricted flow rate range, The sum of the spray flow rate of the adjusted associated spray gun and the spray flow rates of the remaining non-associated spray guns is not greater than the preset maximum flow rate; The spray angle of the associated spray gun is adjusted to not more than the maximum adjustable angle; pass The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun.

4. The gas-liquid two-phase flow control method for an oxygen-enriched side-blown smelting furnace according to claim 3, characterized in that: pass The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun in the following steps: Population initialization and constraint filtering, input the injection oxygen flow rate and injection angle of the current associated spray gun, the erosion energy density and bubble uniformity of the high-risk area; Based on the current parameters, the initial population is randomly generated, and the infeasible solutions that exceed the flow velocity angle, total flow rate and maximum adjustable angle constraints are filtered out to obtain the feasible solution set; Calculate the attacking benefit of the solution by non-dominated sorting of the feasible solution set and defender's gain , hierarchical sorting by non-dominated relations, hierarchical sorting by non-dominated relations, and taking the first layer of the hierarchical solution set as the optimal frontier; And process the optimal frontier solution to obtain the attacking party's benefits and defender's benefits Perform weighted summation to obtain the target value, normalize the target value, generate a uniform reference point, perform crossover mutation to generate offspring, and output a new population; For the new population, the solutions with excessive total flow are eliminated, the elite solutions are retained, the Pareto solution set is output, and the solutions are selected according to the priority of the Pareto solution set, and the injection oxygen flow rate and injection angle of the associated spray gun are output.

5. The gas-liquid two-phase flow control method for an oxygen-enriched side-blown smelting furnace according to claim 1, characterized in that: The steps for outputting the control effect according to the adjusted erosion energy reduction value and efficiency reduction ratio are as follows: Obtain the overall erosion energy of the oxygen-enriched side-blown smelting furnace after regulation and the erosion energy of the high-risk area, and subtract the overall erosion energy before regulation from the overall erosion energy after regulation to obtain an overall erosion energy reduction value, and divide the overall erosion energy reduction value by the overall erosion energy before regulation to obtain an overall regulation effective value; The erosion energy of the high-risk area before regulation is subtracted from the erosion energy of the high-risk area after regulation to obtain the erosion energy reduction value of the high-risk area, and the erosion energy reduction value is divided by the overall erosion energy before regulation to obtain the effective value of regulation in the high-risk area; Add the overall control effective value and the high-risk area control effective value to obtain the erosion energy reduction value; Obtaining the efficiency of the oxygen-enriched side-blowing smelting furnace before and after the regulation, subtracting the efficiency of the oxygen-enriched side-blowing smelting furnace after the regulation from the efficiency of the oxygen-enriched side-blowing smelting furnace before the regulation to obtain an efficiency decrease value, and dividing the efficiency decrease value by the efficiency of the oxygen-enriched side-blowing smelting furnace before the regulation to obtain an efficiency decrease ratio; If the erosion energy reduction value is not less than the preset erosion energy reduction value threshold, and the efficiency reduction ratio is less than the preset efficiency reduction ratio threshold, it indicates that the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously regulated according to the injection oxygen flow rate and injection angle of the regulated associated spray gun; If the erosion energy decrease value is less than the preset erosion energy decrease value threshold or the efficiency decrease ratio is less than the preset efficiency decrease ratio threshold, it means that the control effect is low, and the S2-S3 dynamic game decision is performed again, and the new injection oxygen flow rate and injection angle of the associated spray gun are output until the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted.

6. A gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace, characterized in that: The system comprises: Spalling risk module: obtains status data of the oxygen-enriched side-blown smelting furnace and locates high-risk areas of refractory spalling in the oxygen-enriched side-blown smelting furnace based on the status data; Game decision module: determines the associated spray gun according to the high-risk area, performs dynamic game decision on the high-risk spalling area, and outputs a set of control parameters for the associated spray gun, the control parameter set including the injection oxygen flow rate and injection angle; Regulation module: This module adjusts the control parameters of the associated lances based on the lance control parameter set, regulates the gas-liquid two-phase flow of the oxygen-enriched side-blown smelting furnace, and outputs the regulation effect based on the erosion energy reduction value and efficiency reduction ratio after adjustment; Continuous control module: determines whether the control effect is effective. If it is effective, it continuously adjusts the gas-liquid two-phase flow of the smelting furnace.

7. The gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace according to claim 6, characterized in that: The Spalling Risk Module includes: Acoustic signal acquisition module: multiple acoustic emission sensors are arranged on the outer wall of the furnace to collect acoustic signals in the molten pool; Image acquisition module: An industrial camera is installed outside the furnace to capture optical bubble images on the molten pool surface and obtain bubble characteristics; Data alignment module: aligns timestamps of all data based on the IEEE1588 protocol; 3D thermal map module: The spatial coordinates of the acoustic wave source are determined by triangulation and acoustic wave signals, and the spatial coordinates of the acoustic wave source are spatially mapped with the bubble contours in the optical image to obtain a 3D thermal map of the molten pool. Distribution uniformity module: extracts the position of each bubble in the optical bubble image, divides the molten pool surface into several small areas, and counts the position distribution of bubbles in each area. Calculates the variance of the bubble position in each area, and uses the inverse of each variance as the uniformity of the bubble distribution. Erosion energy density module: Divide the 3D heat map into several regions. For each region, obtain the corresponding regional acoustic signal, extract the frequency and amplitude of the acoustic signal through spectrum analysis, calculate the signal energy of the acoustic signal, and divide the signal energy by the regional volume to obtain the erosion energy density. High-risk area positioning module: If the bubble distribution uniformity is less than the preset bubble distribution uniformity threshold or the erosion energy is not less than the preset erosion energy density, the corresponding area is defined as a high-risk area for refractory spalling.

8. The gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace according to claim 6, characterized in that: The game decision module includes Associated spray gun module: Marks high-risk areas in the three-dimensional thermal map, identifies the high-risk areas, determines the corresponding three-dimensional coordinates of the high-risk areas and associates the spray guns; Coverage association rule module: Coverage association rule: If the center of the high-risk area is within the coverage of spray gun i, then mark spray gun i as an associated spray gun; Attacker module: Constructs a game decision model, defines the game participants and their attackers, and the attacker strategy is to increase the flow rate of the associated spray gun, increase the spray angle, and maximize the smelting efficiency; the profit function is: , where is the attacking party's profit value, is the mean value of the uniformity of bubble distribution in all regions, is the injection oxygen flow rate of the associated spray gun, is the overall maximum injection oxygen flow rate; and is the weight of the attacker’s payoff function; Defending module: The defending side of the game participant. The defending strategy is to reduce the flow rate of the associated spray gun, reduce the spray angle, and minimize the erosion risk. The payoff function is: , where is the profit value of the defender, is the erosion energy density, is the maximum erosion energy density; is the weight of the defender’s payoff function; First constraint module: adjusted oxygen flow rate of the associated spray gun Within the restricted flow rate range, Second constraint condition module: the sum of the spray flow rate of the adjusted associated spray gun and the spray flow rates of the remaining non-associated spray guns is not greater than the preset maximum flow rate; The third constraint condition module: the adjustment angle of the spray angle of the associated spray gun is not greater than the maximum adjustable angle; Solving module: through The algorithm solves the game decision model and outputs the injection oxygen flow rate and injection angle of the associated spray gun.

9. The gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace according to claim 8, characterized in that: The solution module includes: Initialization module: population initialization and constraint filtering, inputting the currently associated injection oxygen flow rate and injection angle, erosion energy density in high-risk areas, and bubble uniformity; Screening module: Based on the current parameters, it randomly generates the initial population, filters out the infeasible solutions that exceed the flow velocity angle, total flow rate and maximum adjustable angle constraints, and obtains the feasible solution set; Non-dominated sorting module: sorts the feasible solution set by non-dominated sorting and calculates the attacking benefit of the solution and defender's gain , hierarchical sorting by non-dominated relations, hierarchical sorting by non-dominated relations, and taking the first layer of the hierarchical solution set as the optimal frontier; New population module: process the optimal frontier solution and convert the attacking party's benefits and defender's benefits Perform weighted summation to obtain the target value, normalize the target value, generate a uniform reference point, perform crossover mutation to generate offspring, and output a new population; Control parameter module: eliminate the total flow exceeding limit solutions for the new population, retain the elite solutions, output the Pareto solution set, select the solutions according to the priority of the Pareto solution set, and output the injection oxygen flow rate and injection angle of the associated spray gun.

10. The gas-liquid two-phase flow control system for an oxygen-enriched side-blown smelting furnace according to claim 6, characterized in that: The adjustment module also includes: Overall control effective value module: obtains the overall erosion energy of the oxygen-enriched side-blown smelting furnace after control and the erosion energy of the high-risk area, and subtracts the overall erosion energy before control from the overall erosion energy after control to obtain the overall erosion energy reduction value, and then divides the overall erosion energy reduction value by the overall erosion energy before control to obtain the overall control effective value; Regional control effective value module: subtract the erosion energy of the high-risk area before control from the erosion energy of the high-risk area after control to obtain the erosion energy reduction value of the high-risk area, and divide the erosion energy reduction value by the overall erosion energy before control to obtain the effective value of the high-risk area control; Erosion energy reduction value module: Add the overall control effective value and the high-risk area control effective value to obtain the erosion energy reduction value; Efficiency reduction ratio module: obtains the efficiency of the oxygen-enriched side-blowing melting furnace before and after regulation, and subtracts the efficiency of the oxygen-enriched side-blowing melting furnace after regulation from the efficiency of the oxygen-enriched side-blowing melting furnace before regulation to obtain the efficiency reduction value. The efficiency reduction value is divided by the efficiency of the oxygen-enriched side-blowing melting furnace before regulation to obtain the efficiency reduction ratio. Continuous control module: If the erosion energy reduction value is not less than the preset erosion energy reduction value threshold, and the efficiency reduction ratio is less than the preset efficiency reduction ratio threshold, it means that the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted according to the injection oxygen flow rate and injection angle of the regulated associated spray gun; Readjustment module: If the erosion energy decrease value is less than the preset erosion energy decrease value threshold or the efficiency decrease ratio is less than the preset efficiency decrease ratio threshold, it means that the control effect is low, and the dynamic game decision of the game decision module and the adjustment module is re-performed, and the new injection oxygen flow rate and injection angle of the associated spray gun are output until the control effect is effective, and the gas-liquid two-phase flow of the smelting furnace is continuously adjusted.

Images