Non-ferrous metal smelting method and system with intelligent regulation and control of sulfidic carbon and slag type

Abstract

The application discloses a smelting method and system for non-ferrous metal with intelligent regulation and control of sulfide carbon and slag type, which comprises the following steps: initial setting and target issuing processing of production plan and initial data of furnace condition to obtain current control target and starting condition, forming a preparation state of closed-loop control by the starting condition; model calculation and setting generation processing of the last cycle furnace condition and current input raw materials to obtain each operating parameter setting value, forming an operation scheme of the current cycle by the setting value; closed-loop execution and process monitoring processing of the setting value in the operation scheme to obtain real-time process data and furnace temperature control curve, forming a process monitoring state by the real-time process data; deviation feedback and dynamic correction processing of the real-time process data and setting deviation to obtain corrected operation instructions, forming a corrected furnace condition of the next cycle by the corrected operation instructions. The application improves the stability of sulfide carbon and slag type, enhances the process adaptability and production efficiency.

Description

Technical Field

[0001] This invention relates to the field of low-carbon heating and process control technology in non-ferrous metal smelting processes, and particularly to a smelting method and system for non-ferrous metals with sulfur substitution for carbon and synergistic intelligent control of slag type. Background Technology

[0002] Traditional non-ferrous metal smelting mainly relies on carbon fuels such as coke, pulverized coal, heavy oil, and natural gas for heating, resulting in high energy consumption and carbon dioxide emission intensity per unit product. Although sulfur in sulfide ores oxidizes exothermically under high-temperature and oxygen-rich conditions, and the sulfur dioxide in the tail gas can be used to produce acid, there are two key contradictions in actual production that have long been unresolved through engineering closed-loop systems: First, the sulfur content and oxidation degree of the ore fluctuate strongly depending on the batch, particle size, moisture, oxygen distribution, and reaction zone of the raw materials, making sulfur heat an uncontrollable heat source. This often results in a mismatch between the exothermic and heat-demanding phases. When sulfur heat is insufficient, the only way to compensate is to increase carbon fuel, which directly increases carbon dioxide emissions. Second, the concentration and total amount of sulfur dioxide are affected by the capacity of the acid production system and environmental constraints. When sulfur heat is passively too strong or oxygen enrichment is too high, it may cause fluctuations or exceedances in sulfur dioxide. Traditional ex-post adjustments based on experience are difficult to balance heat, emissions, and stability. Meanwhile, the regulation of slag profile (Fe / SiO2, viscosity, liquidus temperature, etc.) often lags behind thermal control, lacking a closed-loop linkage with heating / oxygen supply. This leads to higher slag viscosity, slagging, erosion, metal loss, and increased risk of shutdown. Therefore, a solution that can maintain controllability under fluctuating feedstocks is urgently needed: introducing elemental sulfur as a measurable and rapidly dispatchable carbon-free exothermic supplementary heat source into the process, and constructing a constrained closed-loop optimization control with heat, oxygen, emission windows, and slag profile windows as hard constraints, fundamentally achieving synchronous and optimal synergy between heat, oxygen, slag, and emission. Summary of the Invention

[0003] The main objective of this invention is to provide a smelting method and system for non-ferrous metals that utilizes sulfur substitution for carbon and intelligent slag shape control. This addresses the problem in existing technologies where sulfur is introduced as a measurable and rapidly dispatchable carbon-free exothermic heat source, and a constrained closed-loop optimization control is constructed using heat, oxygen, emission windows, and slag shape windows as hard constraints. This fundamentally achieves the problem of synchronous and optimal heat-oxygen-slag-emission coordination. Specifically, it relates to a smelting method and supporting system that achieves carbon-free or low-carbon exothermic heat supplementation through sulfur addition; it can be applied to flash furnaces, bottom-blown furnaces, side-blown furnaces, top-blown furnaces, and other smelting furnace types, enabling intelligent control of heat supply and slag shape during the smelting process.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A smelting method for non-ferrous metals using sulfur-for-carbon substitution and intelligent slag shape control is disclosed. This method explicitly decomposes the heat supply into exothermic reactions of ore sulfur oxidation, sulfur elemental replenishment, and effective carbon fuel release, making the sulfur elemental replenishment amount a key controllable variable that can be solved independently. Simultaneously, an oxygen balance model and a tail gas composition prediction model are established, incorporating the metered oxygen demand for sulfur-carbon reaction, sulfur dioxide, carbon monoxide, and oxygen windows in the tail gas, as well as the acid production load capacity, into the constraints to achieve coupled balancing of oxygen supply, sulfur replenishment, and fuel. Furthermore, a slag shape target window model is constructed, using the iron-to-silica ratio, slag viscosity, and liquid phase temperature as hard constraints or penalties, and linking the flux ratio and slag shape deviation for calculation. This forms a unified constraint-type closed-loop optimizer encompassing matter, energy, emissions, and slag shape.

[0005] As a further improvement of the present invention, the smelting method for non-ferrous metals with sulfide-carbon synergistic slag intelligent control includes the following steps: The initial data of production plan and furnace condition are processed through initial setting and target issuance to obtain the current control target and start-up conditions. The start-up conditions form the preparation state for closed-loop control. The furnace condition at the end of the previous cycle and the current raw material input are used to calculate and generate the set values ​​of each operating parameter, and the control plan for the current cycle is formed from the set values. The setpoints in the control scheme are processed through closed-loop execution and process monitoring to obtain real-time process data and control curves such as furnace temperature. The real-time process data forms the process monitoring status. The real-time process data and the set deviation are processed through deviation feedback and dynamic correction to obtain the corrected operation instructions. The corrected operation instructions form the corrected furnace condition for the next cycle and automatically enter the next round of model calculation.

[0006] As a further improvement of the present invention, the process of forming the control scheme for the current cycle from the set value includes the following steps: The furnace condition at the end of the previous cycle and the current raw materials were processed by the heat balance equation and the sulfur-carbon heat substitution model to obtain the required total heat and sulfur-carbon distribution. The sulfur-carbon distribution formed the initial estimates of the sulfur replenishment and fuel quantity. The initial estimates are processed by the oxygen balance model and exhaust gas window constraints to obtain the oxygen supply setpoint and exhaust gas composition prediction values. The oxygen supply setpoint and exhaust gas composition prediction values ​​form the executable controllable range within the emission window. The control variables within the executable control range are processed by a slag-type window model and a constrained closed-loop optimizer to obtain the optimal setpoint vector, which forms the control scheme for the current period.

[0007] As a further improvement of the present invention, a constrained closed-loop optimizer is constructed to solve for the executable setpoint and perform closed-loop correction in each control cycle; the constrained closed-loop optimizer uses the current measured or estimated state. Furnace temperature, exhaust gas, feed rate, slag composition, and disturbance prediction Fluctuations in raw material sulfur content and changes in production volume are used as inputs, with manipulated variables as the primary inputs. As the decision vector, under the condition of satisfying hard constraints, it minimizes carbon fuel consumption and CO2 emissions and suppresses temperature and slag shape deviations; the optimization solution also includes feasibility judgment and backoff strategy: when constraints cannot be satisfied at the same time, it automatically switches to the strategies of reducing feed or load, fuel backup, and window priority.

[0008] As a further improvement of the present invention, the heat balance equation is as follows: based on the composition of the furnace charge and the target smelting temperature, the total heat required per unit time or per unit output is calculated. This includes the latent heat required for material melting, the heat effect of chemical reactions, and heat loss from the furnace wall; calculations can be performed by substituting online collected furnace temperature and flow rate data, and can be updated in real time. To reflect current operating conditions and requirements; Sulfur-Carbon Heat Substitution Model: This model introduces heat supply terms from two heat sources, sulfur and carbon, into the heat balance equation, and incorporates a sulfur-carbon substitution factor. A model for the distribution of sulfur heat and carbon heat was established; the model transforms the heat balance problem into a problem of... and The solution is derived by combining practical constraints. , The optimization sub-problem; the various calorific values ​​required for model calculation. From standard thermochemical data; in the model As key manipulated variables, their upper and lower limits, ramp rate, and total SO2 or concentration window will serve as the constraint set. Write to the optimizer; Oxygen balance model: Based on the stoichiometric oxygen demand of sulfur oxidation and carbon combustion, a model is established to determine the relationship between oxygen demand and feed rate. The control system calculates the supply of combustion oxygen through a model, which serves as the setpoint for the smelting blower or oxygen valve; the oxygen balance and the exhaust gas window together constitute a set of constraints. ; Slag type target window model: Based on historical production data, target windows for key slag parameters are predefined, including: iron-silicon ratio; viscosity; liquidus temperature; Constrained closed-loop optimization model: Integrates heat conservation, oxygen balance, exhaust gas window, and slag type window to construct a real-time constrained optimization problem; defines decision variables. ,in Sulfur elemental replenishment mass flow rate, carbon fuel mass flow rate, Oxygen supply mass flow rate, Flux flow rate; the objective function is chosen as a weighted form that balances low carbon emissions and steady-state performance: The weight parameters Each term in the corresponding objective function is uniformly represented as unit time cost. Assuming a fuel cost weight, the price per unit mass of fuel is used; The carbon emission cost weight depends on the internal carbon price / quota price / assessment price. Carbon asset / financial weight; To obtain the stable temperature weight, the unit time loss caused by temperature deviation is calculated; This indicates the penalty weight for exceeding the slag / exhaust gas limits, including historical cost conversions such as losses from slagging shutdowns, penalties for exceeding limits, and losses from fluctuations in acid production. It is the weight of the comprehensive cost item. The values ​​are calculated from the unit prices and corresponding flow rates of elemental sulfur, oxygen, flux, and fuel; the weights of the temperature deviation term and the window overrun term are determined by the unit time loss when the upper limit of the allowable limit is reached. , Penalty for poor quality The exhaust gas window is designed to penalize exceeding the boundary, and its boundary is determined by the process specifications, environmental protection limits, and the capacity of the acid production system.

[0009] As a further improvement of the present invention Volume fraction of exhaust gas components: Energy constraints: ; Oxygen balance constraints: ; Hard constraints on exhaust window: And meet the upper limit of acid production load. ;in, To meet the requirements for stable operation of the acid production system, environmental emission constraints, and flue gas corrosion or safety limitations; Characterizes burnout and safety requirements, procedures, or safety standards; Excess oxygen control bandwidth, procedures, experience, or trade-offs between thermal efficiency and erosion; Source: flue gas flow meter; The maximum permissible throughput from the acid production system; Slag-type window hard constraint: iron-silicon viscosity Liquidus temperature ;in Source: Process specifications / historical stable low slagging range statistics and enterprise standard slagging type window; The requirement is given by whether the object is flowable or non-slagging; The minimum superheat required to indicate the solid fraction or fluidity; Climbing speed constraint: Slope constraint parameters Determined jointly by the actuator's capability and the allowable rate of change in the process: ;in The results are obtained from valve stroke time, frequency converter acceleration / deceleration slope, and flow calibration curve or commissioning ramp test. The sensitivity coefficients are derived by inversely calculating the allowable change rates of the exhaust gas window, furnace temperature change rate, and slag shape window, and obtained through regression analysis of historical operating data or online identification.

[0010] As a further improvement of the present invention, the process of forming process monitoring status from real-time process data includes the following steps: The setpoints in the control scheme are processed by fast variable constrained model predictive control or proportional integral derivative control to obtain the adjustment signal of the actuator, and the adjustment signal forms the actual action of the actuator; The actual actions are processed by the process equipment response and sensor data acquisition to obtain real-time process data and furnace temperature control curves, and the real-time process data forms a sequence of instantaneous values ​​of process variables; The instantaneous value sequence of process variables is processed by window boundary comparison and dead zone discrimination to obtain the boundary crossing indicator and steady-state deviation. The boundary crossing indicator and steady-state deviation form the process monitoring status of the current cycle.

[0011] As a further improvement of the present invention, the process of obtaining a corrected operation command by means of deviation feedback and dynamic correction processing of real-time process data and set deviation includes the following steps: Real-time process data and set deviation are compared and dead zone discrimination is processed to obtain out-of-bounds deviation signal, which forms the activation condition of dynamic corrector; The start-up conditions and the disturbance terms predicted by the long short-term memory network are processed by deviation feedback and dynamic correction to obtain the corrected operation command, which forms the adjustment amount of the actuator. The adjustment amount is processed through furnace condition response and status update to obtain the corrected furnace condition for the next cycle, and the corrected furnace condition for the next cycle is automatically used to enter the next round of model calculation.

[0012] To achieve the above objectives, the present invention also provides the following technical solution: A non-ferrous metal smelting system with sulfur-substituted carbon and slag-type intelligent control is applied to the aforementioned non-ferrous metal smelting method with sulfur-substituted carbon and slag-type intelligent control. The non-ferrous metal smelting system with sulfur-substituted carbon and slag-type intelligent control includes: The start-up condition setting module is used to process the initial data of production plan and furnace condition after initial setting and target issuance to obtain the current control target and start-up conditions, and to form a preparation state for closed-loop control based on the start-up conditions; The model calculation module is used to calculate and generate the set values ​​of each operating parameter based on the furnace condition at the end of the previous cycle and the current raw material input, and to form the control plan for the current cycle from the set values. The monitoring status generation module is used to obtain real-time process data and control curves such as furnace temperature by processing the set values ​​in the control scheme through closed-loop execution and process monitoring. The process monitoring status is formed from the real-time process data. The calibration command generation module is used to obtain the calibrated operation command by processing the deviation feedback and dynamic correction between real-time process data and set deviation. The calibrated operation command forms the calibrated furnace condition for the next cycle and automatically enters the next round of model calculation.

[0013] As a further improvement of the present invention, it also includes: a data acquisition layer, a control and execution layer, an optimization and modeling layer, and a safety redundancy mechanism; the data acquisition layer includes integrated sensors, analyzers, and flow meters; the control and execution layer includes a PLC / DCS control core, a feeder, a blower, and valves; the optimization and modeling layer includes a sulfur and carbon thermal control model, a slag type target window model, and their learning modules; the safety redundancy mechanism includes dual redundant sensors and a safety interlock strategy.

[0014] The data acquisition layer consists of sensors and acquisition devices deployed at key measuring points in the smelting system, forming a data acquisition unit for real-time monitoring of process parameters. The control and execution layer, consisting of a distributed control system or programmable logic controller and actuators, is used to manipulate field devices according to optimized instructions. The optimization and modeling layer, located on the system's host computer, is hosted by an industrial computer or a dedicated server. It runs the sulfur elemental heating control model, slag window model, tail gas window prediction model, constrained closed-loop optimizer, and feasibility judgment module. The safety redundancy mechanism includes redundant configuration, fault handling and safety interlocks. Key measurement parameters adopt dual-sensor redundancy configuration and abnormal readings are eliminated through signal verification. Key actuators are configured with parallel backup or manual / automatic dual loops. The control system is embedded with a safety interlock strategy: when the furnace temperature or pressure exceeds the safety threshold, emergency cooling or air reduction protection actions are automatically triggered.

[0015] 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 above-described intelligent control method for smelting non-ferrous metals using sulfur-substituted carbon slag.

[0016] 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 a smelting method for non-ferrous metals with intelligent control of slag type in the form of sulfur-substituted carbon.

[0017] This invention achieves significant energy conservation, emission reduction, and emission reduction effects: by using sulfur to replace part of the carbon for heating, the demand for external fuel and carbon dioxide emissions are greatly reduced; at the same time, by optimizing the supply and utilization of oxygen, combustion is more complete and efficient, resulting in a decrease in the amount of flue gas per unit product and a reduction in the total emissions of harmful gases such as sulfur dioxide, making it more environmentally friendly. Since the addition of elemental sulfur does not introduce carbon, this embodiment can prioritize the use of sulfur supplementation and heat supplementation when the heat load increases or the sulfur content of the raw materials decreases, thereby reducing the lower limit of carbon fuel from the source; and through the closed-loop solution of sulfur dioxide window constraints and acid production load constraints, the fluctuation of sulfur dioxide concentration in the tail gas is significantly reduced, reducing the load impact and emission risks of the acid production system. This embodiment improves process stability, productivity, and product quality: due to the synergy of heat control and slag type regulation, the smelting thermal process and metallurgical reaction are always kept in an optimized state; the furnace temperature is kept constant within the ideal range through closed-loop control, avoiding reaction imbalance caused by overheating or overcooling; the sulfur dioxide concentration in the flue gas is stable, and this stability is not the result of empirical adjustment, but rather determined by the constraint optimizer. The window, acting as a hard constraint and using sulfur replenishment as a controllable variable, provides greater robustness to raw material fluctuations and facilitates stable operation of the acid production system. Slag viscosity and composition are strictly controlled within the target window, ensuring smooth melt flow and thorough metal-slag separation. Consequently, the smelting process is more stable and continuous, reducing unplanned shutdowns and malfunctions, and increasing operational efficiency. Metals in the raw materials are recovered more fully, and optimized slag profiles reduce metal loss, resulting in more stable product quality, such as matte grade and crude lead purity. This embodiment enhances smelting intensification and process stability, increasing processing efficiency per unit time. Attached Figure Description

[0018] Figure 1 This is a schematic flowchart of one embodiment of the smelting method for non-ferrous metals using sulfur-substituted carbon and synergistic intelligent control of slag type according to the present invention. Figure 2 This is a schematic diagram of an embodiment of the smelting method for non-ferrous metals using sulfur-substituted carbon and synergistic intelligent control of slag type according to the present invention. Figure 3This is a schematic flowchart illustrating the steps of a non-ferrous metal smelting method based on sulfur substitution for carbon and synergistic slag intelligent control, according to an embodiment of the present invention, which generates a control scheme for the current cycle from set values. Figure 4 This is a schematic diagram of the sulfur-carbon thermal balance substitution mechanism and thermal self-sufficiency point in an embodiment of the non-ferrous metal smelting method of the present invention, which uses sulfur-carbon substitution and slag-type intelligent control. Figure 5 This is a schematic flowchart illustrating the steps of forming process monitoring status from real-time process data in an embodiment of the non-ferrous metal smelting method of using sulfur-substituted carbon and synergistic slag type intelligent control according to the present invention. Figure 6 This is a schematic diagram of the slow and fast variable collaborative control logic of an embodiment of the smelting method for non-ferrous metals with slag type intelligent control using sulfur-substituted carbon. Figure 7 This is a schematic diagram of the steps in an embodiment of the smelting method for non-ferrous metals using sulfur-substituted carbon synergistic slag intelligent control, where real-time process data and set deviations are processed through deviation feedback and dynamic correction to obtain corrected operation instructions. Figure 8 This is a schematic diagram of machine learning model integration and decision support for an embodiment of the non-ferrous metal smelting method with sulfide-carbon synergistic slag type intelligent control according to the present invention. Figure 9 This is a schematic diagram of the functional modules of an embodiment of the non-ferrous metal smelting system with sulfur-substituted carbon and synergistic slag intelligent control according to the present invention. Figure 10 This is a schematic diagram of an embodiment of the non-ferrous metal smelting system with sulfide-carbon synergistic slag type intelligent control 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

[0019] 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.

[0020] 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 accompanying drawings). 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.

[0021] 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.

[0022] The non-ferrous metal smelting method of this invention, which uses sulfur-to-carbon synergistic intelligent control of slag shape, explicitly decomposes the heat supply into the exothermic reaction of sulfur in the ore, the exothermic addition of elemental sulfur, and the effective exothermic reaction of carbon fuel, making the amount of elemental sulfur added a key controllable quantity that can be solved independently. At the same time, an oxygen balance model and a tail gas composition prediction model are established, and the metered oxygen demand for the reaction of sulfur and carbon, the sulfur dioxide, carbon monoxide, and oxygen windows in the tail gas, as well as the acid production load capacity are written into the constraints to achieve the coupled balancing of oxygen supply, sulfur replenishment, and fuel. Furthermore, a slag shape target window model is constructed, using the iron to silica ratio, slag viscosity, and liquid phase temperature as hard constraints or penalty terms, and linking the flux ratio and slag shape deviation for calculation. Finally, a constraint-type closed-loop optimizer that unifies material, energy, emissions, and slag shape is formed.

[0023] like Figure 1 The present embodiment provides an example of a smelting method for non-ferrous metals using sulfur-substituted carbon and slag-type intelligent control. In this embodiment, the smelting method for non-ferrous metals using sulfur-substituted carbon and slag-type intelligent control specifically includes the following steps: Step S1: After initial setting and target issuance processing of the production plan and initial furnace condition data, the current control target and start-up conditions are obtained, and the start-up conditions form the preparation state for closed-loop control. Step S2: The furnace condition at the end of the previous cycle and the current raw material input are processed by model calculation and setting to obtain the set values ​​of each operating parameter, and the control plan for the current cycle is formed from the set values; Step S3: The set values ​​in the control scheme are processed through closed-loop execution and process monitoring to obtain real-time process data and control curves such as furnace temperature. The process monitoring status is formed from the real-time process data. Step S4: The real-time process data and the set deviation are processed through deviation feedback and dynamic correction to obtain the corrected operation command. The corrected operation command forms the corrected furnace condition for the next cycle and automatically enters the next round of model calculation.

[0024] The initial settings and target issuance are determined by operators or the upper-level scheduling system based on the production plan, setting the production targets and process constraints for the shift. These include the planned ore processing volume, product quality indicators such as copper matte grade and lead ingot purity, slag indicators within the Fe / SiO2 ratio range, and energy consumption indicators. Target parameters are input into the control system via the human-machine interface as optimization solutions and control criteria. Simultaneously, the initial data of the current furnace condition is read to complete the preparation for the start of closed-loop control. Model calculation and setting generation: At the beginning of each control cycle, the optimization and modeling layer acquires the most recent data and runs the model to solve the problem, generating the setpoints for the main operating parameters of the cycle. In each control cycle, the optimizer first calculates the current heat gap based on the required net heat load within the control cycle, and determines the feasible region for sulfur replenishment under the constraints of the tail gas window and acid production capacity. Subsequently, it solves the constrained optimization problem and outputs the setpoint manipulation variables, with the mass flow rate of elemental sulfur replenishment given as a priority variable. If the sulfur replenishment region is limited, fuel is automatically increased or a feed reduction / load reduction scheme is triggered to ensure that the solution always falls within the executable region. The model solution comprehensively considers the furnace conditions at the end of the previous cycle (such as slag composition and furnace temperature) and the current raw material input. The control settings provided include: the acceleration rate of each raw material, fuel, and flux, the oxygen flow rate for combustion, the target furnace temperature, and the expected range of SO2 concentration in the flue gas. These settings are equivalent to formulating a control plan for the current cycle. Closed-loop execution and process monitoring: The control execution layer controls the field equipment according to the model setpoints, and the system enters an automatic operation state. The fast variable controller continuously collects measured values ​​at a high frequency and compares them with the setpoints, outputting adjustment signals to the actuators to achieve stable control of temperature, pressure, gas composition, etc. The slow variable is adjusted at a low frequency according to the cumulative deviation at the end of the cycle. During the control process, the data acquisition layer continuously uploads real-time process data for monitoring. Deviation Feedback and Dynamic Correction: While the system operates automatically, the optimization and modeling layer and machine learning module continuously correct the model and parameters based on the latest collected data, forming a feedback loop. If a significant deviation in a key indicator is detected and it exceeds the dead zone, the dynamic regulator will immediately correct it: for example, when the furnace temperature is lower than the set value, it automatically increases fuel and oxygen supply to raise the temperature; if real-time monitoring shows that sulfur dioxide is approaching the upper limit or the acid production load has reached the upper limit, the system will use this state as a hard constraint and feed it back to the optimizer: automatically tightening the upper limit of sulfur replenishment mass flow rate and resolving the set value for the next cycle, while the fast loop prioritizes oxygen and small-scale fuel adjustments to suppress window overflows; if this is still not feasible, a preset backoff strategy (reducing feed / load, switching to fuel backup, limiting sulfur replenishment) will be executed to ensure safety and compliance are prioritized; if the slag Fe / SiO2 ratio is detected to deviate from the target, the flux charging valve opening will be adjusted or the flux ratio will be changed in the next batch to gradually correct the slag shape. These feedback control actions are usually executed within seconds to compensate for raw material fluctuations, environmental changes, and model errors, ensuring that each indicator remains stable within the set range. With the effect of closed-loop feedback, the actual process and the model prediction will continue to converge, and the deviation will approach zero after entering a steady state.After running in this closed loop for a period of time, the task of one control cycle is completed; the system will automatically enter the model calculation for the next cycle, and so on. The entire process in this embodiment realizes fully automatic adjustment and closed-loop control from raw material input to product output.

[0025] Preferably, Figure 2 This diagram illustrates the principle of a non-ferrous metal smelting method using sulfonated carbon synergistic slag type intelligent control. It includes initial settings and target issuance, model calculation and setting generation, closed-loop execution and process monitoring, and deviation feedback and dynamic correction. This embodiment integrates sulfonated carbon synergistic thermal control with slag type adjustment to achieve intelligent closed-loop control of the smelting process. First, by integrating production plan and initial furnace condition data, clear control targets and start-up conditions are established, ensuring the system is in a preparatory state for closed-loop control from the initial stage, laying the foundation for dynamic optimization of the entire process. Second, based on model calculations of the previous cycle's furnace condition and current raw materials, set values ​​for various operating parameters are generated, forming a precise control scheme for the current cycle, achieving dynamic matching of raw material ratio, thermal parameters, and slag type requirements. Through closed-loop execution and process monitoring, key parameter curves such as furnace temperature are collected in real time, forming continuous process status monitoring and providing a data basis for timely identification of process fluctuations. Finally, using a feedback mechanism of real-time data and set deviations, operating commands are dynamically corrected, achieving cycle-by-cycle adaptive adjustment of the furnace condition and automatically connecting to the next round of model calculation, forming a continuously optimized intelligent control cycle.

[0026] In summary, this embodiment achieves closed-loop intelligent control of the entire smelting process, from target setting, parameter generation, process monitoring to dynamic correction, improving the stability of sulfonated carbon synergistic thermal control and the accuracy of slag type formulation, and enhancing process adaptability and production efficiency.

[0027] In summary, this embodiment addresses four pain points in non-ferrous smelting: uncontrollable heat sources, high carbon consumption, slag shape lag, and fluctuating sulfur dioxide emissions. It proposes a constrained closed-loop optimization control method and system that uses the metered addition of elemental sulfur as a carbon-free exothermic control quantity and multiple constraints of heat, oxygen, emissions, and slag shape as boundaries. This transforms sulfur heat from a passive source to a controllable heat supply, minimizing carbon fuel and carbon dioxide emissions while meeting acid production and environmental constraints. Simultaneously, it stabilizes furnace temperature, tail gas window, and slag shape indicators, achieving a balance between low carbon emissions, high efficiency, and steady-state operation.

[0028] Furthermore, such as Figure 3 The process of forming the control scheme for the current cycle from the set value in step S2 specifically includes the following steps: Step S21: The furnace condition at the end of the previous cycle and the current raw materials are processed by the heat balance equation and the sulfur-carbon heat substitution model to obtain the required total heat and sulfur-carbon distribution. The sulfur-carbon distribution forms the initial estimated values ​​of sulfur replenishment and fuel quantity. Step S22: The initial estimated values ​​are processed by the oxygen balance model and exhaust gas window constraints to obtain the oxygen supply setpoint and exhaust gas composition prediction values. The oxygen supply setpoint and exhaust gas composition prediction values ​​form the executable controllable range within the emission window. Step S23: The control variables within the executable control range are processed by the slag-type window model and the constrained closed-loop optimizer to obtain the optimal setpoint vector, and the control scheme for the current period is formed by the optimal setpoint vector.

[0029] The thermal control principle of this embodiment lies in using sulfur to replace carbon for heating, and supplementing with elemental sulfur as a key means to achieve controllable heat: on the one hand, utilizing the exothermic oxidation of sulfur in the ore; on the other hand, when the sulfur content of the raw material is low or there is a sulfur-heat / demand heat mismatch, elemental sulfur (solid sulfur / particulate sulfur / molten sulfur are all acceptable) is metered and added as a carbon-free exothermic source to participate in the oxidation reaction (S + O2 → SO2) to supplement the heat load, thereby stabilizing the furnace temperature and reducing CO2 emissions without increasing carbon fuel. To make sulfur-carbon substitution calculable in engineering, this embodiment defines an effective sulfur-carbon substitution factor. This is used to measure the proportion of heat released by sulfur oxidation per unit mass to the effective heat released per unit mass of carbon fuel under a given operating condition. The heat released by sulfur oxidation per unit mass: Enthalpy of reaction (given by standard thermochemical database). The molar mass of sulfur; Effective heat release per unit mass of carbon fuel: ,in From thermochemical database; The effective oxidation degree of carbon in the furnace / afterburner can be inferred from the molar fractions of CO and CO2 measured by the online exhaust gas analyzer. If necessary, normalization correction can be performed based on oxygen content and dilution. Considering reaction efficiency and heat transfer efficiency, It can be calibrated from historical operating conditions and updated online.

[0030] Under this definition, the amount of elemental sulfur added can be manipulated by a closed-loop solution based on the online tail gas composition, furnace temperature, and energy demand: when sulfur content is insufficient or heat load increases, the system prioritizes increasing elemental sulfur for supplemental heating, minimizing carbon fuel input while meeting SO2 window and acid production capacity constraints, thereby achieving low-carbon and stable temperature. To describe the supply ratio of each heat source under heat self-sufficiency conditions, a heat distribution model is established based on energy conservation: ,in For the mass of sulfur to be oxidized, The mass of carbon to be burned. By optimizing the supply combination of sulfur heat and carbon heat, partial or complete self-sufficiency of smelting heat can be achieved. Based on the conservation of energy per unit time, the required net heat load within the control period Δt is defined. for: .in, (Sensible heat) (Latent heat of phase transition) This is the main endothermic reaction term (which can be calculated as reaction progress × reaction enthalpy or given by the calibration model). Alternatively, it can be obtained from the cooling water heat balance (both can be calculated from online flow / temperature measurements). Controllable heating is defined as the sum of the heat released by sulfur oxidation and the effective heat released by carbon fuel, then: in Calculated from raw material sulfur grade × feed rate. The mass flow rate for replenishing elemental sulfur is measured and controlled by a screw balance / mass flow meter. The mass flow rate of carbon fuel (measured and controllable by a scale / flow meter). For other auxiliary combustion / electric heating, etc. (if any). The constraints are: For safety margin (set by operating procedures). This formula allows the system to directly operate within a given range. Find feasible combinations of sulfur replenishment and fuel quantity under the given conditions. This is especially important when the ore's inherent sulfur content is insufficient or the sulfur heat release is inadequate to cover the required amounts. In this embodiment, the increase in carbon fuel is not simply a matter of increasing carbon fuel, but rather follows a strategy of prioritizing sulfur supplementation and using carbon fuel as a safety net: under the premise of meeting constraints such as SO2 concentration / total amount, acid production capacity, and environmental emissions, priority is given to increasing... To compensate for the heat load; only when the sulfur supplementation reaches its upper limit (e.g., triggered by acid production capacity constraints, SO2 window upper boundary constraints, or sulfur supplementation equipment capacity constraints) and still cannot meet the requirements. Only then did it gradually improve. As a compensation for excess heat load, it minimizes carbon consumption and CO2 emissions within the feasible region; the sulfur-carbon heat balance substitution mechanism and heat self-sufficiency point, such as Figure 4 As stated above.

[0031] Regarding oxygen supply, this embodiment establishes an oxygen balance and exhaust gas window constraint model to accurately match the combustion oxygen quantity, and solves the problem in conjunction with the sulfur supplementation quantity. The metered oxygen demand (kg / s) within the control cycle is defined as: ,in (S→SO2 measurement) kgO2 / kgC, covering the effective oxidation degree of C→CO and C→CO2. It can be inferred from the online measurement of CO / CO2 in the exhaust gas; For other oxidizable components, such as The oxygen demand term can be given by a calibration model or a thermodynamic / empirical model. Based on this, the oxygen supply is set as follows: , For oxygen utilization efficiency, This provides an operational margin. More importantly, this embodiment incorporates the exhaust gas window as a hard constraint into the optimization: The window boundary is jointly set by environmental limits, the steady-state requirements of the acid production system, and safety regulations. Therefore, when the sulfur content of the raw material increases or the amount of sulfur supplemented increases, the system does not simply increase oxygen, but rather ensures that the aforementioned window and... Under the common constraints, synchronous scheduling is optimized through constraint optimization. and This ensures stable furnace temperature, consistent SO2 concentration, and minimal CO2 emissions.

[0032] Applying the above mechanism model to industrial processes, this embodiment constructs a constrained closed-loop optimizer based on online monitoring data. Within each control cycle, it solves for the executable setpoint and performs closed-loop correction. The optimizer uses the current measured / estimated state... Furnace temperature, tail gas SO2 / CO / O2, feed rate, slag composition, and disturbance prediction. Fluctuations in raw material sulfur content and changes in production volume are used as inputs, with manipulated variables as the main inputs. Using the decision vector, the system minimizes carbon fuel consumption and CO2 emissions while suppressing temperature and slag shape deviations under hard constraints such as heat balance, oxygen balance, emission window, and slag shape window. The optimization solution also includes feasibility assessment and a fallback strategy: when constraints cannot be simultaneously satisfied, such as when the upper limit of acid production capacity or sulfur supplementation equipment is triggered, the system automatically switches to a feed reduction / load reduction strategy, fuel backup, and window priority strategy to ensure safe and stable operation.

[0033] (1) Heat balance equation: Calculate the total heat required per unit time or per unit output based on the composition of the furnace charge and the target smelting temperature. This includes the latent heat required for material melting, the heat effect of chemical reactions, and heat loss from the furnace wall. The heat balance equation has been previously given; it can be updated in real time by substituting online collected data such as furnace temperature and flow rate. To reflect the current operating conditions and requirements.

[0034] (2) Sulfur-carbon heat substitution model: The heat supply terms of two heat sources, sulfur and carbon, are introduced into the heat balance equation, combined with the sulfur-carbon substitution factor. A model for the distribution of sulfur heat and carbon heat was established; the model transforms the heat balance problem into a problem of... and The solution can be derived by combining practical constraints, such as maximum fuel flow rate and allowable SO2 concentration in flue gas. , The optimization sub-problem; the various calorific values ​​required for model calculation. From standard thermochemical data; in this model The key manipulated variables in this embodiment, including their upper and lower limits, ramp rate, and SO2 total / concentration window, will serve as the constraint set. Write to the optimizer.

[0035] (3) Oxygen balance model: Based on the stoichiometric oxygen demand of sulfur oxidation and carbon combustion, a model is established to establish the relationship between oxygen demand and feed rate. The control system uses this model to accurately calculate the supply of combustion oxygen, which serves as the setpoint for the smelting blower or oxygen valve. The oxygen consumption coefficients involved in the model, such as 2.67 kg O2 / kg C, are constants determined by stoichiometry. The oxygen balance and the exhaust gas window together constitute a set of constraints. It is used to ensure the stability of SO2 / CO / O2 and to meet acid production and environmental protection constraints.

[0036] (4) Slag Type Target Window Model: To ensure good slag performance, target windows for key slag parameters are predefined based on historical production data, including: ① Iron-Silicon Ratio (Fe / SiO2 ratio), used to measure the relative content of iron oxides and silicon dioxide in the slag; ② Viscosity μ ③ Liquid phase temperature, melting point or liquidus temperature, indicates the temperature at which the slag begins to solidify. The ideal slag type range is determined based on literature and production experience, for example, an Fe / SiO2 ratio within a certain range, viscosity below a certain upper limit, and... The temperature should be kept below the actual furnace temperature by a certain margin. The iron-silicon ratio determines the acidity and equilibrium distribution of the slag: if the ratio is too high, the slag will have more iron and less silicon, increasing slag viscosity and melting point, and enhancing its corrosiveness to the furnace lining; if the ratio is too low, the slag will be too acidic, easily causing metal loss; it needs to be controlled within an appropriate range. Slag viscosity needs to be kept at a low level, for example, 0.5–1.5 Pa·s, depending on the process, to ensure good slag flowability and metal-slag separation efficiency. Liquidation temperature. The target slag temperature should be a certain range lower than the actual slag temperature inside the furnace, and should not exceed the current furnace temperature by more than 50 °C, to avoid premature crystallization and precipitation of the slag. The above target window parameters can all be obtained through experimental calibration curves or thermodynamic calculations, showing their functional relationship with composition, and stored in a database for real-time querying. For example, in an iron silicate slag system, increasing the Fe / SiO2 ratio from 0.8 to 1.2 can significantly reduce slag viscosity and lower the slag crystallization temperature, which is beneficial for maintaining low viscosity and a fully molten state of the slag. This embodiment selects a reasonable target range for the Fe / SiO2 ratio based on this, supplemented by viscosity and melting point constraints, to form a triple constraint window for the slag composition; Fe / SiO2, μ, ... The defined slag-type window constitutes a constraint set. Together with the thermal / oxygen constraint, it determines the feasible region for optimization.

[0037] (5) Constrained Closed-Loop Optimization Model: This embodiment integrates heat conservation, oxygen balance, tail gas window, and slag type window to construct a real-time constrained optimization problem. Taking the control period k as an example, the decision variables are defined. ,in Sulfur elemental replenishment mass flow rate, Carbon fuel mass flow rate (converted to pulverized coal / coke / natural gas) Oxygen supply mass flow rate, Flux flow rate. The objective function can be a weighted form that balances low carbon emissions and steady-state performance: The weight parameters Each term in the corresponding objective function is uniformly represented as unit time cost. Assuming a fuel cost weight, the price per unit mass of fuel is used; The carbon emission cost weight depends on the internal carbon price / quota price / assessment price. Carbon asset / financial weight; To obtain the stable temperature weight, the unit time loss caused by temperature deviation is calculated. This indicates the penalty weight for exceeding the slag / exhaust gas limits, which includes historical cost conversions such as losses from slagging shutdowns, penalties for exceeding limits, and losses from fluctuations in acid production. It is the weight of the comprehensive cost item. The values ​​are calculated from the unit prices and corresponding flow rates of elemental sulfur, oxygen, flux, and fuel; the weights of the temperature deviation term and the window overrun term are determined by the unit time loss when the upper limit of the allowable limit is reached. , Penalty for poor quality The exhaust window is designed to penalize exceeding the boundary, and its boundary is determined by the process specifications, environmental protection limits and the capacity of the acid production system. Volume fraction of exhaust gas components: ① Energy constraint: ; ②Oxygen balance constraint: ; ③ Hard constraint on exhaust window: And meet the upper limit of acid production load. .in, The lower limit is used to ensure stable acid production, while the upper limit is subject to the acid production absorption capacity, tail emission limits, and equipment corrosion constraints, in accordance with the requirements for stable operation of the acid production system, environmental emission constraints, and flue gas corrosion / safety restrictions. Characterizing burnout and safety requirements, procedures / safety standards; Excess oxygen control bandwidth, trade-offs between procedures / experience / thermal efficiency and erosion; Source: flue gas flow meter; The maximum allowable throughput is given by the acid production system nameplate / design documents / operating procedures. ④ Slag type window hard constraint: iron-silicon ratio. viscosity Liquidus temperature .in Source: Process specifications / historical stable low slagging range statistics and enterprise standard slagging type window. Given by the requirements for flowability / non-slagging. This indicates the minimum superheat required for solids content / fluidity, determined by the company's established procedures and experience, representing the minimum superheat.

[0038] ⑤ Climbing speed constraint: Slope constraint parameters Determined jointly by the actuator's capability and the allowable rate of change in the process: .in The results are obtained from valve stroke time, frequency converter acceleration / deceleration slope, and flow calibration curve or commissioning ramp test. The sensitivity coefficients are derived by inversely from the allowable change rates of the exhaust gas window (SO2 / CO / O2), furnace temperature change rate, and slag type window, and can be obtained through regression analysis of historical operating data or online identification. The optimizer solves for the results in each cycle. Then, it is sent to the PLC / DCS as a set value for execution, and recalculated based on online feedback in the next cycle to achieve constrained closed-loop self-optimization control of sulfur replenishment-oxygen supply-fuel-flux.

[0039] In this embodiment, the window boundary parameters involved in the optimization model are determined by the process specifications, environmental protection limits, and the processing capacity of the acid production system; the actuator upper limit and ramp constraint are determined by the rated capacity of the equipment, valve stroke time, frequency converter acceleration / deceleration slope or commissioning ramp test, and the more stringent one is chosen under the premise of meeting the exhaust gas window and temperature change rate constraints; the energy and oxygen balance parameters are calculated by combining online flow, temperature, exhaust gas analysis data with thermochemical database and historical calibration efficiency coefficients; the objective function weight parameters are determined by the purchase price, internal carbon price or enterprise KPI weight, or are normalized to 1 after each item according to the allowable deviation to avoid manual parameter adjustment.

[0040] Preferably, this embodiment achieves systematization and refinement of the furnace condition control process through the integrated application of heat balance equations, sulfur-carbon heat substitution models, oxygen balance models, tail gas window constraints, slag type window models, and constrained closed-loop optimizers. First, based on heat balance and sulfur-carbon substitution relationships, the furnace condition and raw material input are transformed into heat demand and elemental allocation, thereby establishing initial benchmarks for fuel and sulfur supplementation to ensure the basic stability of the in-furnace thermochemical process. Second, through oxygen balance models and tail gas composition constraints, the initial estimates are mapped to oxygen supply setpoints and tail gas predictions, forming an operable range that complies with emission limits, maintaining the controllability of the reactant gas composition while meeting environmental protection requirements. Finally, within the operable range, slag type window conditions are introduced, and a closed-loop optimization algorithm is used to perform multi-objective optimization of the controllable variables, generating an optimal setting scheme that balances slag phase properties, operating costs, and process stability, thereby improving the overall efficiency and economy of the smelting process.

[0041] In summary, this embodiment realizes model-based decision-making throughout the entire process from raw material input to final control scheme generation, enhances the system's adaptability to operating condition fluctuations, and optimizes heat distribution and element utilization efficiency while ensuring emission compliance and slag stability.

[0042] Furthermore, such as Figure 5 The process of forming the process monitoring status from real-time process data in step S3 specifically includes the following steps: Step S31: The setpoint in the control scheme is processed by fast variable constrained model predictive control or proportional integral derivative control to obtain the adjustment signal of the actuator, and the actual action of the actuator is formed by the adjustment signal. Step S32: The actual action is processed by the process equipment response and sensor data acquisition to obtain real-time process data and furnace temperature control curve, and the instantaneous value sequence of process variables is formed from the real-time process data; Step S33: The instantaneous value sequence of the process variable is processed by window boundary comparison and dead zone discrimination to obtain the boundary crossing indicator and steady-state deviation. The boundary crossing indicator and steady-state deviation form the process monitoring status of the current cycle.

[0043] Among them, such as Figure 6 The control strategy, driven by sulfur replenishment, employs a constrained closed-loop optimization: slow variable RTO and fast variable MPC. A hierarchical, rate-based control strategy is designed for variables with different dynamic characteristics in the smelting process to achieve coordinated regulation of slow and fast variables. Slow variables mainly refer to process parameters that change slowly and have high inertia, such as batching ratio and slag composition; fast variables refer to parameters that respond quickly and fluctuate frequently, such as furnace temperature and flue gas composition. The two types of variables are controlled using different periods and algorithms, but coordination is achieved through information sharing and hierarchical constraints. (1) Slow variable control, constrained optimization of sulfur supplementation-feeding-slag type: Slow variables operate on a minute-level / furnace secondary cycle, and the core calculation is solved by the constrained optimizer. ;in The preferred amount is used to achieve heat shaping when the sulfur content of the raw material fluctuates or the heat load changes; the flux and feedstock are used to combine Fe / SiO2, μ, Maintain within the target window; the slow loop solution strictly satisfies the exhaust gas sulfur dioxide window and acid production capacity constraints to ensure that sulfur supplementation will not cause emissions to exceed the limit; (2) Fast variable control, furnace temperature, oxygen content, flue gas composition, etc.: Fast variable control, furnace temperature / exhaust gas window constraint tracking: Fast variables are executed at the second level, and constrained MPC / PID is used to realize rapid correction of furnace temperature and exhaust gas windows. The optimization goal of the fast loop is to minimize temperature deviation and window overrun risk under actuator limit and ramp constraints. Typical manipulated variables are oxygen valve and a small amount of fuel fine adjustment; the sulfur supplement is adjusted by speed limit or updated by the slow loop to avoid rapid and large changes that cause sulfur dioxide shock.

[0044] (3) Coordination and decoupling: The slow loop provides the "optimal setpoint and constraint boundary", while the fast loop is responsible for steady-state tracking and maintaining the window under disturbances and model errors. By combining the slow and fast variable control strategies, when the fast loop continuously triggers the window boundary, such as when sulfur dioxide approaches the upper limit, the signal is sent back to the slow loop, which automatically tightens the sulfur replenishment upper limit or triggers the retreat strategy to achieve consistency between the upper and lower closed loops. The two cycles repeatedly, so that the smelting maintains the optimal stable state under various disturbances.

[0045] Preferably, in this embodiment, the setpoint is processed by fast variable constraint model predictive control or proportional-integral-derivative control to generate a precise adjustment signal for the actuator. The adjustment signal enables the actuator to complete the corresponding action within a finite time, thereby establishing a controllable channel for the physical process. After system delay or nonlinear attenuation, the response mapping coefficient will be reflected back to the measuring instrument through stable output data. The corresponding fluctuation of the process variable in the numerical record library is imaged at precise intervals of a specific scanning cycle to form a dataset for later use. After this process, the signal oscillation rate that cannot be skipped is transmitted to the clock boundary evaluation, dead zone interval identification, and even lag judgment. The variable value is deducted layer by layer and then compared layer by layer. The structure is delivered to the discrete alarm condition and stability error deviation library. The state analysis is then performed. At this time, the negative wide interval is collected again to continuously feed the state fine-tuning framework. The deviation information is automatically recorded. The result is then identified by the variable value. The next phase of the machine's layout system demonstrates a two-tiered, closed-loop characteristic. First, it features rapid inspection and feedback to correct overt instability, edge jump warnings, or abnormal loosening intervals, accurately recording abnormal deviations and movement fields. Second, the current measurement transfer vector generated by each time series is directly coupled with the measurement volume fork evaluation, timing sequence cut-off judgment, and data over-control throughout the entire jump time. It can achieve single-cycle quantization and continuous stable coupling at all times, reflecting the above coordinated and consistent successful binding of signals from the initial element to the completed measurement description package, the coverage capability of the operation evaluation output, the stable response in continuous state, and the absence of noise, dynamic ambiguity, disconnection, or dead zones, maintaining process consistency.

[0046] Furthermore, such as Figure 7 The process of obtaining the corrected operation command by processing the deviation between the real-time process data and the set deviation through deviation feedback and dynamic correction in step S4 specifically includes the following steps: Step S41: The real-time process data and the set deviation are compared and the dead zone is judged to obtain the out-of-bounds deviation signal, which forms the start condition of the dynamic corrector. Step S42: The start-up conditions and the disturbance terms predicted by the long short-term memory network are processed by deviation feedback and dynamic correction to obtain the corrected operation command, which forms the adjustment amount of the actuator. Step S43: The adjustment amount is processed by furnace condition response and status update to obtain the corrected furnace condition for the next cycle, and the corrected furnace condition for the next cycle is automatically used to enter the next round of model calculation.

[0047] Among them, the machine learning model is integrated to enhance the predictability of the system to disturbances and the adaptive updating of model parameters. The machine learning model is introduced as an auxiliary module of the constrained closed-loop optimizer. Its output does not directly replace the optimization decision, but is used to provide prediction of disturbances, soft measurement and initial value of solution, thereby improving the solution speed and robustness. (1) LSTM time series prediction model: used to predict the trend of furnace temperature, SO2 / CO / O2 and raw material fluctuations in the short term, such as the equivalent disturbance term of sulfur content change. The prediction results are used as Input into the constrained MPC / RTO to tighten the feasible domain of sulfur replenishment and oxygen supply in advance, and avoid SO2 shock or temperature overshoot.

[0048] (2) XGBoost ingredient recommendation model: used to establish raw material composition-slag type index, Fe / SiO2, μ, T L - A fast mapping between flux proportions provides soft measurements / initial values ​​for the slag model and enables constrained optimizers. u 0 Initial guesses and parameter correction suggestions. The final sulfur replenishment, oxygen supply, fuel and flux setpoints will be based on the results of the constrained optimizer solution, ensuring that hard constraints are met and that the system is feasible and interpretable.

[0049] By integrating these two types of models, a certain degree of self-learning and adaptive capabilities are achieved: the LSTM prediction module improves the insight into future trends, making control more forward-looking; the XGBoost recommendation module, essentially condensing expert experience, can provide a second set of decision suggestions outside the model, enabling the system to obtain a reasonable control scheme even under unknown operating conditions. Both models can be continuously updated and trained online through the collection of new data, ensuring that their applicability and accuracy continuously improve with changes in production. It should be noted that the introduction of machine learning models is to enhance, not replace, traditional mechanistic model control: in practical applications, when the output of the ML model is inconsistent with the mechanistic optimization results, the scheduler can evaluate the decision or take a weighted compromise to ensure safety and robustness.

[0050] Preferably, in this embodiment, the deviation between real-time data and the setpoint during the process forms a signal that enters the first step of processing. After passing through a dead-zone discrimination subsystem with configurable threshold attributes, the signal completes the filtering process, filtering out invalid fluctuations within the threshold range. The valid deviations after dead-zone discrimination are assigned specific Boolean probability parameters, which then correct the input relationship of the related action interval, including the upper and lower consecutive states, generated by the long short-term memory prediction model for subsequent processing.

[0051] like Figure 9 The present embodiment also provides an embodiment of a smelting system for non-ferrous metals with sulfur-substituted carbon and slag-type intelligent control. In this embodiment, the smelting system for non-ferrous metals with sulfur-substituted carbon and slag-type intelligent control is applied to the smelting method for non-ferrous metals with sulfur-substituted carbon and slag-type intelligent control as described in the above embodiment. The smelting system for non-ferrous metals with sulfur-substituted carbon and slag-type intelligent control includes a start-up condition setting module 1, a model calculation module 2, a monitoring status generation module 3, and a correction instruction generation module 4 that are connected in sequence. The system comprises the following modules: Start-up Condition Setting Module 1, which processes the initial data of production plan and furnace condition into initial settings and target issuance to obtain the current control target and start-up conditions, forming a preparation state for closed-loop control; Model Calculation Module 2, which processes the furnace condition at the end of the previous cycle and the current raw material input into model calculation and setting to obtain the set values ​​of each operating parameter, forming the control scheme for the current cycle; Monitoring Status Generation Module 3, which processes the set values ​​in the control scheme into real-time process data and control curves such as furnace temperature through closed-loop execution and process monitoring, forming the process monitoring status; and Correction Command Generation Module 4, which processes the real-time process data and setting deviation into a corrected operation command through deviation feedback and dynamic correction, forming the corrected furnace condition for the next cycle and automatically entering the next round of model calculation.

[0052] Preferably, this embodiment realizes a complete closed-loop control process from target setting, model calculation, process execution to dynamic correction; enabling the smelting process to significantly reduce carbon fuel consumption and carbon dioxide emissions while ensuring acid production and environmental protection constraints, while stabilizing furnace temperature, flue gas composition and slag type indicators, ultimately achieving the comprehensive goal of high efficiency, low carbon and stable operation.

[0053] This embodiment achieves low-carbon, high-efficiency, and stable operation of the smelting process by constructing a smelting control method that combines elemental sulfur replenishment with constrained closed-loop optimization. Its core logic is as follows: First, the amount of elemental sulfur replenishment is used as an independently adjustable manipulated quantity for heat compensation and steady-state adjustment when the raw material sulfur content is insufficient or when there is a mismatch between sulfur heat and demand heat, thereby reducing dependence on carbon fuels and reducing carbon dioxide emissions. Second, a multi-constrained closed-loop optimization framework is established, covering heat balance, oxygen balance, SO2 / CO / O2 emission windows, furnace temperature window, and slag type window (Fe / SiO2, viscosity, liquidus temperature). This framework solves for key parameters such as elemental sulfur replenishment, carbon fuel input, combustion oxygen flow rate, and flux ratio in real time during each control cycle, and executes and dynamically corrects these parameters through a DCS / PLC system in a closed loop. To achieve the aforementioned closed-loop optimization, a mechanism model system based on online data-driven approaches was constructed: the heat balance model explicitly decomposes the net heat load into the exothermic reaction of sulfur oxidation in the ore, the exothermic reaction of elemental sulfur addition, and the effective exothermic reaction of carbon fuel, making the amount of elemental sulfur addition an independently solvable manipulated variable; the oxygen balance model is coupled with the tail gas composition prediction model, using the sulfur / carbon reaction metering oxygen demand, tail gas composition window, and acid production load constraints to achieve coordinated balancing of oxygen supply, sulfur addition, and fuel; the slag shape target window model uses indicators such as Fe / SiO2, slag viscosity, and liquid phase temperature as constraints to calculate the flux ratio and evaluate the slag shape state. Finally, the above models are integrated to form a unified constraint-based closed-loop optimizer encompassing matter, energy, emissions, and slag shape; this optimizer outputs executable setpoints in each control cycle and continuously corrects model parameters and constraint boundaries based on online feedback, achieving a shift from empirical adjustment to calculable and verifiable closed-loop self-optimization control. This significantly reduces carbon fuel consumption and CO2 emissions while stabilizing furnace temperature, flue gas composition, and slag shape indicators, while ensuring acid production and environmental protection requirements.

[0054] Furthermore, such as Figure 10 The non-ferrous metal smelting system with sulfur-substituted carbon and slag type intelligent control provided in this embodiment also includes: a data acquisition layer, a control and execution layer, an optimization and modeling layer, and a safety redundancy mechanism; the data acquisition layer includes integrated sensors, analyzers, and flow meters; the control and execution layer includes a PLC / DCS control core, feeders, fans, and valves; the optimization and modeling layer includes a sulfur-carbon thermal control model, a slag type target window model, and its learning module; the safety redundancy mechanism includes dual redundant sensors and a safety interlock strategy.

[0055] The data acquisition layer comprises sensors and acquisition devices deployed at key measuring points throughout the smelting system, forming data acquisition units for real-time monitoring of process parameters. This layer includes at least: an online X-ray fluorescence analyzer or batching scale for determining the sulfur, iron, and silicon content in raw materials and fuels; thermocouples and infrared thermometers arranged in different areas of the furnace; a flue gas analyzer for online measurement of sulfur dioxide concentration and oxygen content in flue gas; an online molten material analysis device or rapid sampling experimental device for obtaining the iron-to-silica ratio and the grade of matte or crude lead in the slag; and flow meters for measuring the speed and flow rate of each feeder. Furthermore, the data acquisition layer also collects environmental and operating parameters such as furnace pressure, cooling water flow rate, and temperature, providing a basis for heat balance and material balance calculations. All sensor signals are connected to a field-programmable logic controller (FPGA) or distributed control system and uploaded to a host computer database via an industrial communication network, achieving unified high-frequency acquisition and storage. The acquisition frequency for fast variables such as furnace temperature and gas composition is on the order of seconds, while the acquisition frequency for slow variables such as slag composition is updated per furnace cycle or per hour.

[0056] The control and execution layer, composed of a distributed control system or programmable logic controller (PLC) and actuators, is used to manipulate field equipment according to optimized instructions. Actuators include: feeders for raw materials such as concentrate and electric furnace slag; fuel feeders; flux addition devices; combustion fans and oxygen valves; top-blown or side-blown oxygen lances; pulverized coal injection lances; and cooling and regulating devices. When the optimization and modeling layer provides new control setpoints, the control and execution layer adjusts each actuator through closed-loop regulation: adjusting the speed of the batching and weighing feeder or the opening of the feed valve to add ore, fuel, and flux according to the new proportion; adjusting the oxygen valve opening or blower frequency to change the oxygen supply flow rate; adjusting the oxygen-coal injection rate or insertion depth of the injection lances to affect the temperature distribution in the furnace reaction zone; and adjusting the cooling water flow rate or furnace tilt angle to assist in heat management. The control and execution layer integrates various operations into the distributed control system or PLC program to achieve automatic continuous control, with pre-set manual intervention interfaces. Each execution loop is equipped with limit and interlock protection.

[0057] The optimization and modeling layer, located on the system's host computer and hosted by an industrial computer or dedicated server, runs the sulfur elemental supplementation heating control model, slag type window model, tail gas window prediction model, constrained closed-loop optimizer, and feasibility assessment module. The constrained closed-loop optimizer, using real-time data and predicted disturbances as input, solves for the optimal setpoint under constraints of heat balance, oxygen balance, sulfur dioxide / carbon monoxide / oxygen window, slag type window, and actuator boundary constraints. When the solution proves infeasible, the feasibility assessment module automatically triggers fallback strategies such as feed reduction or load reduction, fuel redundancy, and limited sulfur supplementation, and issues an alarm prompt awaiting manual confirmation.

[0058] The safety redundancy mechanism includes redundant configuration, fault handling, and safety interlocks. Key measurement parameters such as furnace temperature, pressure, and gas composition employ dual-sensor redundancy configurations, and abnormal readings are eliminated through signal verification. Key actuators such as fans and feeders are configured with parallel backups or manual / automatic dual-loop systems. The control system incorporates safety interlock strategies: when furnace temperature or pressure exceeds a safety threshold, emergency cooling or reduced ventilation is automatically triggered; if the model calculation results show an infeasible solution or deviate from common sense, execution is refused and an alarm prompts manual intervention. The system performs regular online self-diagnostics; when sensor drift or communication interruption is detected, it automatically switches to a simplified control mode based on empirical fixed values ​​to maintain production stability and notifies maintenance personnel for repair.

[0059] like Figure 11 The present embodiment provides an example of an electronic device 4, which includes a processor 41 and a memory 42 coupled to the processor 41.

[0060] The memory 42 stores program instructions for implementing the smelting method of non-ferrous metals with sulfide-carbon synergistic slag intelligent control in any of the above embodiments.

[0061] The processor 41 is used to execute program instructions stored in the memory 42 to perform intelligent control of non-ferrous metal smelting with sulfur-substituted carbon and slag-type synergistic process.

[0062] The processor 41 can also be referred to as a CPU (Central Processing Unit). The processor 41 may be an integrated circuit chip with signal processing capabilities. The processor 41 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.

[0063] Furthermore, Figure 12This is a schematic diagram of the structure of a storage medium according to an embodiment of this application. The storage medium 5 of this embodiment stores program instructions 51 capable of implementing all the methods described above. These program instructions 51 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.

[0064] 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.

[0065] 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.

[0066] 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 smelting method for non-ferrous metals using sulfur-substituted carbon and synergistic slag-type intelligent control, characterized in that, The non-ferrous metal smelting method, which uses sulfur-to-carbon synergistic slag type intelligent control, explicitly decomposes the heat supply into the exothermic reaction of sulfur in the ore, the exothermic reaction of elemental sulfur, and the effective exothermic reaction of carbon fuel, making the amount of elemental sulfur added a key controllable quantity that can be solved independently. At the same time, an oxygen balance model and a tail gas composition prediction model are established, and the oxygen demand for sulfur-carbon reaction, the sulfur dioxide, carbon monoxide, and oxygen windows in the tail gas, as well as the acid production load capacity are written into the constraints to achieve the coupled balancing of oxygen supply, sulfur replenishment, and fuel. Furthermore, a slag type target window model is constructed, using the iron to silica ratio, slag viscosity, and liquid phase temperature as hard constraints or penalty terms, and linking the flux ratio and slag type deviation to calculate the results. This forms a unified constraint-type closed-loop optimizer that integrates material, energy, emissions, and slag type.

2. The smelting method for non-ferrous metals with sulphur-substituted carbon and synergistic slag-type intelligent control according to claim 1, characterized in that, The smelting method for non-ferrous metals using sulfur-substituted carbon synergistic slag intelligent control includes the following steps: The initial data of production plan and furnace condition are processed through initial setting and target issuance to obtain the current control target and start-up conditions. The start-up conditions form the preparation state for closed-loop control. The furnace condition at the end of the previous cycle and the current raw material input are used to calculate and generate the set values ​​of each operating parameter, and the control plan for the current cycle is formed from the set values. The setpoints in the control scheme are processed through closed-loop execution and process monitoring to obtain real-time process data and control curves such as furnace temperature. The real-time process data forms the process monitoring status. The real-time process data and the set deviation are processed through deviation feedback and dynamic correction to obtain the corrected operation instructions. The corrected operation instructions form the corrected furnace condition for the next cycle and automatically enter the next round of model calculation.

3. The smelting method for non-ferrous metals with sulphur-substituted carbon and synergistic slag type intelligent control according to claim 2, characterized in that, The process of generating the control scheme for the current cycle from the set values ​​includes the following steps: The furnace condition at the end of the previous cycle and the current raw materials were processed by the heat balance equation and the sulfur-carbon heat substitution model to obtain the required total heat and sulfur-carbon distribution. The sulfur-carbon distribution formed the initial estimates of the sulfur replenishment and fuel quantity. The initial estimates are processed by the oxygen balance model and exhaust gas window constraints to obtain the oxygen supply setpoint and exhaust gas composition prediction values. The oxygen supply setpoint and exhaust gas composition prediction values ​​form the executable controllable range within the emission window. The control variables within the executable control range are processed by a slag-type window model and a constrained closed-loop optimizer to obtain the optimal setpoint vector, which forms the control scheme for the current cycle.

4. The smelting method for non-ferrous metals with sulphur-substituted carbon synergistic slag intelligent control according to claim 3, characterized in that, Construct a constrained closed-loop optimizer to solve for the executable setpoint and perform closed-loop corrections in each control cycle; the constrained closed-loop optimizer uses the current measured or estimated state. Furnace temperature, exhaust gas, feed rate, slag composition, and disturbance prediction Fluctuations in raw material sulfur content and changes in production volume are used as inputs, with manipulated variables as the primary inputs. As the decision vector, under the condition of satisfying hard constraints, it minimizes carbon fuel consumption and CO2 emissions and suppresses temperature and slag shape deviations; the optimization solution also includes feasibility judgment and backoff strategy: when constraints cannot be satisfied at the same time, it automatically switches to the strategies of reducing feed or load, fuel backup, and window priority.

5. The smelting method for non-ferrous metals with sulphur-substituted carbon and synergistic slag-type intelligent control according to claim 3, characterized in that, Heat balance equation: Calculates the total heat required per unit time or per unit output based on the composition of the furnace charge and the target smelting temperature. This includes the latent heat required for material melting, the heat effect of chemical reactions, and heat loss from the furnace wall; calculations can be performed by substituting online collected furnace temperature and flow rate data, and can be updated in real time. To reflect current operating conditions and requirements; The sulfur-carbon heat substitution model introduces heat supply terms from two heat sources, sulfur and carbon, into the heat balance equation, and incorporates a sulfur-carbon substitution factor. Establish a model for the distribution of sulfur heat and carbotherm; The model transforms the thermal equilibrium problem into a problem of... and The solution is derived by combining practical constraints. , The optimization sub-problem; the various calorific values ​​required for model calculation. From standard thermochemical data; in the model As key manipulated variables, their upper and lower limits, ramp rate, and total SO2 or concentration window will serve as the constraint set. Write to the optimizer; Oxygen balance model: Based on the stoichiometric oxygen demand of sulfur oxidation and carbon combustion, a model is established to determine the relationship between oxygen demand and feed rate. The control system calculates the supply of combustion oxygen through a model, which serves as the setpoint for the smelting blower or oxygen valve; the oxygen balance and the exhaust gas window together constitute a set of constraints. ; Slag type target window model: Based on historical production data, target windows for key slag parameters are predefined, including: iron-silicon ratio; viscosity; liquidus temperature; Constrained closed-loop optimization model: Integrates heat conservation, oxygen balance, exhaust gas window, and slag type window to construct a real-time constrained optimization problem; defines decision variables. ,in Sulfur elemental replenishment mass flow rate, carbon fuel mass flow rate, Oxygen supply mass flow rate, Flux flow rate; the objective function is chosen as a weighted form that balances low carbon emissions and steady-state performance: The weight parameters Each term in the corresponding objective function is uniformly represented as unit time cost. Assuming a fuel cost weight, the price per unit mass of fuel is used; The carbon emission cost weight depends on the internal carbon price / quota price / assessment price. Carbon asset / financial weight; To obtain the stable temperature weight, the unit time loss caused by temperature deviation is calculated; This indicates the penalty weight for exceeding the slag / exhaust gas limits, including historical cost conversions such as losses from slagging shutdowns, penalties for exceeding limits, and losses from acid production fluctuations. It is the weight of the comprehensive cost item. The values ​​are calculated from the unit prices and corresponding flow rates of elemental sulfur, oxygen, flux, and fuel; the weights of the temperature deviation term and the window overrun term are determined by the unit time loss when the upper limit of the allowable limit is reached. , Penalty for poor quality The exhaust gas window is designed to penalize exceeding the boundary, and its boundary is determined by the process specifications, environmental protection limits, and the capacity of the acid production system.

6. The smelting method for non-ferrous metals with sulphur-substituted carbon and synergistic slag type intelligent control according to claim 5, characterized in that, Volume fraction of exhaust gas components: Energy constraints: ; Oxygen balance constraints: ; Hard constraints on exhaust window: And meet the upper limit of acid production load. ;in, To meet the requirements for stable operation of the acid production system, environmental emission constraints, and flue gas corrosion or safety limitations; Characterizes burnout and safety requirements, procedures, or safety standards; Excess oxygen control bandwidth, procedures, experience, or trade-offs between thermal efficiency and erosion; Source: flue gas flow meter; Maximum permissible throughput from the acid production system; Slag-type window hard constraint: iron-silicon viscosity Liquidus temperature ;in Source: Process specifications / historical stable low slagging range statistics and enterprise standard slagging type window; The requirement is given by whether the object is flowable or non-slagging; The minimum superheat required to indicate the solid fraction or fluidity; Climbing speed constraint: Slope constraint parameters Determined jointly by the actuator's capability and the allowable rate of change in the process: ;in The results are obtained from valve stroke time, frequency converter acceleration / deceleration slope, and flow calibration curve or commissioning ramp test. The sensitivity coefficients are derived by inversely calculating the allowable change rates of the exhaust gas window, furnace temperature change rate, and slag shape window, and obtained through regression analysis of historical operating data or online identification.

7. The smelting method for non-ferrous metals with sulphur-substituted carbon and synergistic slag-type intelligent control according to claim 2, characterized in that, The process of generating process monitoring status from real-time process data includes the following steps: The setpoints in the control scheme are processed by fast variable constrained model predictive control or proportional integral derivative control to obtain the adjustment signal of the actuator, and the adjustment signal forms the actual action of the actuator; The actual actions are processed by the process equipment response and sensor data acquisition to obtain real-time process data and furnace temperature control curves, and the real-time process data forms a sequence of instantaneous values ​​of process variables; The instantaneous value sequence of process variables is processed by window boundary comparison and dead zone discrimination to obtain the boundary crossing indicator and steady-state deviation. The boundary crossing indicator and steady-state deviation form the process monitoring status of the current cycle.

8. The smelting method for non-ferrous metals with sulphur-substituted carbon and synergistic slag-type intelligent control according to claim 2, characterized in that, The process of obtaining corrected operating instructions from real-time process data and set deviations through deviation feedback and dynamic correction includes the following steps: Real-time process data and set deviation are compared and dead zone discrimination is processed to obtain out-of-bounds deviation signal, which forms the activation condition of dynamic corrector; The start-up conditions and the disturbance terms predicted by the long short-term memory network are processed by deviation feedback and dynamic correction to obtain the corrected operation command, which forms the adjustment amount of the actuator. The adjustment amount is processed through furnace condition response and status update to obtain the corrected furnace condition for the next cycle, and the corrected furnace condition for the next cycle is automatically used to enter the next round of model calculation.

9. A smelting system for non-ferrous metals with sulfur-substituted carbon and slag-type intelligent control, applied to the smelting method for non-ferrous metals with sulfur-substituted carbon and slag-type intelligent control as described in any one of claims 1 to 8, characterized in that, The non-ferrous metal smelting system with sulfide-carbon synergistic slag-type intelligent control includes: The start-up condition setting module is used to process the initial data of production plan and furnace condition after initial setting and target issuance to obtain the current control target and start-up conditions, and to form a preparation state for closed-loop control based on the start-up conditions; The model calculation module is used to calculate and generate the set values ​​of each operating parameter based on the furnace condition at the end of the previous cycle and the current raw material input, and to form the control plan for the current cycle from the set values. The monitoring status generation module is used to obtain real-time process data and control curves such as furnace temperature by processing the set values ​​in the control scheme through closed-loop execution and process monitoring. The process monitoring status is formed from the real-time process data. The calibration command generation module is used to obtain the calibrated operation command by processing the deviation feedback and dynamic correction between real-time process data and set deviation. The calibrated operation command forms the calibrated furnace condition for the next cycle and automatically enters the next round of model calculation.

10. The non-ferrous metal smelting system with sulfur-substituted carbon synergistic slag-type intelligent control as described in claim 9, characterized in that, It also includes: a data acquisition layer, a control and execution layer, an optimization and modeling layer, and a safety redundancy mechanism; the data acquisition layer includes integrated sensors, analyzers, and flow meters; the control and execution layer includes a PLC / DCS control core, feeders, fans, and valves; the optimization and modeling layer includes a sulfur and carbon thermal control model, a slag type target window model, and its learning module; the safety redundancy mechanism includes dual redundant sensors and a safety interlock strategy. The data acquisition layer consists of sensors and acquisition devices deployed at key measuring points in the smelting system, forming a data acquisition unit for real-time monitoring of process parameters. The control and execution layer, consisting of a distributed control system or programmable logic controller and actuators, is used to manipulate field devices according to optimized instructions. The optimization and modeling layer is located on the system's host computer and is hosted by an industrial computer or server. It runs the sulfur elemental heating control model, slag window model, tail gas window prediction model, constrained closed-loop optimizer, and feasibility judgment module. The safety redundancy mechanism includes redundant configuration, fault handling and safety interlocks. Key measurement parameters adopt dual-sensor redundancy configuration and abnormal readings are eliminated through signal verification. Key actuators are configured with parallel backup or manual / automatic dual loops. The control system is embedded with a safety interlock strategy: when the furnace temperature or pressure exceeds the safety threshold, emergency cooling or air reduction protection actions are automatically triggered.

Images