Four-property collaborative digital design method and system for waste heat recovery in non-ferrous metal metallurgy

Abstract

The invention relates to the technical field of flue gas waste heat utilization and acid making system optimization in the non-ferrous metal smelting process, in particular to a non-ferrous metal smelting waste heat recovery four-property collaborative digital design method and system, and the method comprises the steps: arranging various sensors at key positions of a smelting furnace, a flue, a waste heat boiler, an acid making system, a steam pipe network and the like; collecting process parameters and waste heat boiler data in real time; building a digital model based on the process parameter data and the waste heat boiler data, and calculating four optimization indexes such as waste heat volatility, flow field uniformity, acid making stability and energy level matching in real time; constructing a time sequence analysis and prediction model; inputting the real-time calculated four-property optimization index into the time sequence analysis and prediction model for prediction; and the optimal parameter combination is obtained through the real-time process parameter data and the waste heat boiler data. According to the method, the optimal economic benefits and environmental benefits can be achieved, and a feasible solution is provided for green and low-carbon transformation of the non-ferrous metal smelting industry.

Description

Technical Field

[0001] This invention relates to the field of flue gas waste heat utilization and acid production system optimization technology in non-ferrous metal smelting processes, and particularly to a digital design method and system for the synergistic optimization of four properties (waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching) of flue gas waste heat recovery systems in non-ferrous metal smelting processes such as molten pool smelting, side-blown furnaces, flash furnaces, and oxygen bottom-blown furnaces. Specifically, this invention proposes a multi-objective optimization framework to address the complex flow characteristics and variable heat load fluctuations of smelting flue gas, aiming to achieve efficient energy recovery, stable operation, and safe operation of the flue gas waste heat recovery system and its supporting waste heat boiler and acid production unit. Background Technology

[0002] Non-ferrous metal smelting processes (such as pool smelting and continuous smelting processes for copper, lead, zinc, nickel, etc.) generally generate high temperatures, high dust levels, and high temperatures. Concentration of flue gas. To improve energy efficiency, reduce carbon emissions, and meet the requirements for comprehensive utilization of sulfur resources, smelting enterprises typically install flue gas waste heat boilers, waste heat steam systems, and corresponding acid production devices to achieve flue gas waste heat recovery and... Comprehensive utilization of acid production. However, in existing technologies and engineering practices, flue gas waste heat recovery and acid production systems have the following prominent problems: strong fluctuations in waste heat, and a lack of systematic quantitative characterization and control methods. Smelting processes such as molten pool smelting inherently possess strong unsteady-state characteristics; changes in furnace charge ratio, oxygen / oxygen-enriched flow rate, and charge surface condition directly affect flue gas flow rate, temperature, and... Significant fluctuations in concentration. Existing systems often rely on empirical methods to roughly adjust spray water volume, bypass opening, or blower volume, lacking dynamic mathematical models and quantitative indicators for "waste heat fluctuations," thus failing to effectively buffer and optimize these fluctuations at their source. Such fluctuations not only affect the system's energy recovery efficiency but may also negatively impact the stability of downstream equipment and processes.

[0003] Uneven flue gas flow distribution easily leads to localized overheating and slagging / ash accumulation. Inside waste heat boilers and subsequent heat exchange equipment, due to the complex flue structure, unstable inflow, and lack of controllable distribution mechanisms, the flue gas velocity and temperature fields exhibit significant non-uniformity. This easily creates localized high-temperature zones, low-velocity stagnant zones, and high-dust deposition zones, resulting in problems such as localized slagging, ash blockage, and tube erosion on the heating surfaces. These problems directly limit the intensity of waste heat recovery and affect the long-term stable operation of the system, reducing the effective waste heat utilization rate of the flue gas. The acid production unit is highly sensitive to flue gas fluctuations and lacks operational stability. The acid production section is highly sensitive to the inflow of flue gas... Flue gas is extremely sensitive to fluctuations in concentration, temperature, and flow rate. Fluctuations in flue gas parameters cause fluctuations in conversion rate, changes in absorber efficiency, and frequent adjustments to system load, resulting in unstable acid production and quality, frequent start-ups and shutdowns, and increased thermal stress fluctuations in equipment. Current control strategies mostly focus on the internal aspects of the acid production system (e.g., temperature regulation of individual conversion towers, local flow control), lacking a synergistic control method based on an integrated "smelting-waste heat-acid production" perspective. This fails to effectively mitigate the impact of upstream fluctuations on the acid production process, leading to overall operational instability. Waste heat energy level matching is inefficient, and energy cascade utilization is insufficient. High-temperature flue gas contains multi-grade thermal energy, theoretically capable of driving high-grade steam turbines for power generation in the high-temperature section, driving acid production blowers or performing process heating in the medium-temperature section, and being used for heating or providing industrial hot water in the low-temperature section. However, most existing engineering designs only use steam at a single pressure level or simply for heating hot water, lacking a system model and quantitative evaluation method for matching "waste heat energy level - energy user". This results in high-grade heat energy being consumed by low-grade operating conditions, or even directly emitted, leading to low efficiency in the cascade utilization of energy. This not only wastes valuable energy resources but also affects the overall energy efficiency of the smelting process.

[0004] Existing optimization methods are limited to single indicators and lack a comprehensive optimization approach that considers all four aspects ("four properties synergistically"). Current retrofitting schemes typically focus on optimizing a specific local indicator (e.g., improving waste heat boiler thermal efficiency, refining flue gas distribution structure, or locally reducing...). The system lacks a unified modeling and joint optimization approach for "waste heat fluctuations, flow field uniformity, acid production stability, and energy level matching." Such local optimization often fails to take into account the overall performance of the system, making it impossible to fundamentally solve complex process fluctuations and heat recovery efficiency problems, which is not conducive to the coordinated improvement of the entire process towards high efficiency, stability, and low carbon emissions.

[0005] The aforementioned problems are often intertwined and contradictory: optimization measures for one aspect may adversely affect another. For example, increasing spray cooling water or opening bypass flues to alleviate fluctuations in flue gas heat and composition can, to some extent, smooth out fluctuations entering subsequent systems, but at the same time, it reduces flue gas temperature and thermal energy grade, affecting the temperature conditions of downstream acid production reactions and the efficiency of high-grade energy recovery. Similarly, excessively introducing structures such as guide vanes to improve flow field uniformity increases system resistance and flue gas residence time, which may exacerbate the accumulation of fluctuations and pose new challenges to system stability.

[0006] Therefore, there is an urgent need for a new "four-in-one" synergistic method for waste heat recovery from flue gas throughout the entire non-ferrous metal smelting process. This method organically couples waste heat fluctuation control, flow field homogenization, acid production process stability control, and energy level matching optimization by establishing corresponding mathematical models and evaluation indicators. This achieves comprehensive optimization and intelligent control at the system level, thereby significantly improving energy utilization efficiency, ensuring the stability of the smelting process, and reducing carbon emissions and energy waste.

[0007] Prior art 1, Chinese Patent Application No.: 202510345526.1, relates to a waste heat gradient recovery system and a non-ferrous metal smelting system. The waste heat gradient recovery system includes a heat exchanger, a deaerator, an economizer, a pyrometallurgical waste heat boiler, a superheater, and a high-pressure pump. The heat exchanger's absorber side, the deaerator, the economizer, the pyrometallurgical waste heat boiler's absorber side piping, and the superheater are sequentially connected to allow demineralized water to sequentially exchange and recover waste heat from the low-temperature zone of the sulfuric acid dry absorption section, the low-temperature zone of the sulfuric acid conversion section, and the medium-temperature zone of the pyrometallurgical smelting section and the sulfuric acid conversion section. The high-pressure pump is located between the deaerator and the economizer. Although the waste heat gradient recovery system recovers low-grade waste heat from non-ferrous metal smelting systems, improving waste heat recovery efficiency and increasing steam production, and by recovering waste heat of different grades and ultimately producing high-grade superheated steam, it not only benefits subsequent steam utilization but also reduces energy consumption within the waste heat gradient recovery system; however, the waste heat energy level matching is crude and the energy cascade utilization is insufficient.

[0008] Prior art two, Chinese patent application number: 202410027360.4, relates to the field of waste heat recovery devices, and discloses a waste heat recovery device and method for a non-ferrous metal smelting furnace. The waste heat recovery device for the non-ferrous metal smelting furnace includes a support frame, a furnace body is arranged below the support frame, a heat-conducting ring is fixedly installed on the side wall of the furnace body, a waste heat collection tank is fixedly installed inside the heat-conducting ring, a connecting plate is fixedly installed on the side wall of the heat-conducting ring, and an arc-shaped heat-conducting plate is fixedly connected to the top of the two connecting plates. A through groove is opened at the top of the waste heat collection tank. Although the scale on the heating tube is scraped off by the scale-scraping mechanism, increasing the heat conduction efficiency and working efficiency of the heating tube, and the scale-scraping mechanism is further cleaned by the transmission mechanism and cleaning component, the transmission mechanism enables one start-up source to control multiple effects. Compared with traditional electric control devices, the above operation of the mechanical transmission mechanism has a higher fault tolerance rate; however, the acid production device is highly sensitive to flue gas fluctuations and has insufficient operational stability.

[0009] Prior art three, Chinese patent application number: 202510079802.4, relates to the field of waste heat recovery technology, specifically a waste heat recovery device for a non-ferrous metal smelting furnace, including a shell, heat-conducting pipes, a hot water tank and a cold water tank, and further including a support plate, a fixed plate, a cylinder, a transmission assembly, a transmission rod, a positioning scraper, a movable scraper and an adjustment assembly. The support plate is connected to the upper side of the inner cavity of the shell, the fixed plate array is connected to the upper side of the support plate, the cylinder is connected to the front end of the left side of the fixed plate, and the transmission assembly is connected to... At the rear of the cylinder, the transmission rod is connected to the transmission assembly, the positioning scraper array is connected to the circumferential surface of the transmission rod, the movable scraper is connected to the front side of the positioning scraper, and the adjustment assembly is connected to the inner cavity of the positioning scraper. Although by retaining a thin layer of impurities on the windward side of the heat pipe, the impurities can form a protective layer on the heat pipe when the airflow is flowing, thereby avoiding the wear of the heat pipe by the flow of impurities and achieving the purpose of ensuring stable use of the equipment; however, the uneven distribution of the flue gas flow field can easily cause local overheating and slag and ash accumulation.

[0010] Currently, existing technologies 1, 2, and 3 suffer from problems such as crude matching of waste heat energy levels, insufficient energy cascade utilization, high sensitivity of acid production units to flue gas fluctuations, insufficient operational stability, and uneven distribution of flue gas flow field, which can easily lead to local overheating and slagging and ash accumulation. To solve the above problems, this invention provides a four-property synergistic digital design method and system for waste heat recovery in non-ferrous metal smelting. Summary of the Invention

[0011] The main objective of this invention is to provide a digital design method and system for the coordinated operation of four properties in the recovery of waste heat from non-ferrous metal smelting, in order to solve the problems in the existing technology, such as crude matching of waste heat energy levels, insufficient energy cascade utilization, high sensitivity of acid production units to flue gas fluctuations, insufficient operational stability, and uneven distribution of flue gas flow field, which easily leads to local overheating and slagging and ash accumulation.

[0012] To achieve the above objectives, the present invention provides the following technical solution: A digital design method for the synergistic development of four properties in waste heat recovery from non-ferrous metal smelting, comprising: A digital model is constructed based on process parameter data and waste heat boiler data to calculate in real time the four optimization indicators of waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching; and a time series analysis and prediction model is constructed by retrieving process parameter data and waste heat boiler data. The key indicators for real-time computational optimization of the four properties are input into the time series analysis and prediction model for prediction; an optimization objective function is constructed based on the prediction results; constraints are set and weights are configured; and the optimal parameter combination is obtained through real-time process parameter data and waste heat boiler data.

[0013] As a further improvement of the present invention, the process of real-time calculation of key indicators for optimizing four properties, including waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching, includes the following steps: A digital model is constructed based on process parameters and waste heat boiler data to calculate the instantaneous sensible heat power of smelting flue gas, and to obtain its average value and standard deviation, thus obtaining the waste heat fluctuation index; by measuring the flue gas velocity and temperature of the waste heat boiler, the flow field uniformity index is calculated, thus obtaining the flow field uniformity. Based on flue gas Concentration, temperature, and flow rate data are used to calculate the stability index of the acid production system and obtain the acid production stability. The waste heat of the flue gas is divided into energy levels, the recovered heat of different energy levels is calculated, and the proportion of each energy level to the total waste heat is calculated. The energy level matching is obtained through the proportion of the total waste heat. Data on process parameters and waste heat boilers were retrieved to construct time series analysis and prediction models; historical time series data of the four optimization indicators were retrieved to train the time series analysis and prediction models.

[0014] As a further improvement of the present invention, the process of training a time series analysis and prediction model includes the following steps: Based on process parameter data and waste heat boiler data, a time series analysis and prediction model is constructed; the process parameter data is used as an external feature of the time series analysis and prediction model; in the historical four-indicator time series data, the past preset time series data of the four-indicator are retrieved according to a unified timestamp to form a dataset; The dataset is divided into training, validation, and test sets; the parameters of the time series analysis and prediction model are initialized; the training set data is input into the time series analysis and prediction model, and predictions are made based on the current parameters; Calculate the loss between the predicted result and the true value, and adjust the parameters of the time series analysis and prediction model according to the loss value; iterate sequentially until the preset number of iterations is reached; calculate various error indicators of the prediction results of the time series analysis and prediction model on the training set that has not participated in the training.

[0015] As a further improvement of the present invention, the process of adjusting the parameters of the time series analysis and prediction model according to the loss value includes the following steps: Construct a mechanism for calculating parameter adjustments. Take the loss value as input and calculate a set of adjustment data corresponding to the parameter dimensions of the time series analysis and prediction model. The adjustment data represents the value that each parameter should be modified in the current iteration to reduce the loss value. The obtained adjustment data is combined with the current parameter values ​​of the time series analysis and forecasting model to generate a new set of parameter values; After the parameter update is completed, the next iteration begins; the training set data is reprocessed using the updated time series analysis and prediction model parameters to obtain new prediction results; the loss value is recalculated based on the new prediction results, and the adjustment calculation and parameter update steps are repeated; the iteration process is repeated until the preset number of iterations is reached; after the iteration terminates, the final parameters of the time series analysis and prediction model are fixed.

[0016] As a further improvement of the present invention, the process of obtaining a set of adjustment data corresponding to the parameter dimensions of the time series analysis and prediction model through calculation includes the following steps: Using the obtained loss value as input, a direction indicator is generated for each parameter of the time series analysis and prediction model through inverse calculation associated with the internal computational structure of the time series analysis and prediction model; the direction indicator indicates whether the parameter value should be modified in the positive or negative direction in order to reduce the loss value. Simultaneously, it receives two inputs: one is the training sample data in the formed dataset, and the other is the intermediate processing results of the samples under the current parameters of the time series analysis and prediction model; it matches and quantifies the features of the input data with the intermediate processing results, and calculates a non-negative intensity coefficient for each parameter with a determined direction; the intensity coefficient represents the urgency or influence weight of adjusting the parameter in the iteration. The directional identifier of each parameter is paired with the corresponding intensity coefficient, and the directional information and intensity information are merged into a single scalar value; the scalar value is the adjustment amount of the parameter.

[0017] As a further improvement of the present invention, the process of matching and quantifying the features of the input data with the intermediate processing results includes the following steps: Using the obtained parameter direction identifiers and the provided intermediate processing results; based on the adjustment direction specified by the direction identifiers, filter out the data segments related to the current role of each parameter from the intermediate processing results; perform aggregation operations on the data segments to generate an initial contribution factor for each parameter; Using the provided training sample data, calculate the statistical dispersion of each external feature sequence data and target sequence data in the current training batch to obtain a dispersion sequence corresponding to the feature dimension. The contribution factor of each parameter is multiplied by the dispersion value of the feature dimension of the input data associated with it to obtain a weighted contribution metric. A scaling operation is performed on the weighted contribution metrics of all parameters. All metrics are scaled proportionally so that the maximum value is mapped to a preset baseline intensity value, and the remaining values ​​are mapped proportionally. The final value of each parameter obtained after this mapping is the intensity coefficient of the parameter.

[0018] As a further improvement of the present invention, the process of calculating its statistical dispersion within the current training batch includes the following steps: Take the training sample data used, which includes the external feature sequence and the target sequence; for each dimension of the data in the sequence; calculate the maximum and minimum values ​​of that dimension of the data in all samples of the current training batch. Subtract the minimum value from the maximum value to obtain a difference; calculate the average value of this dimension of data across all samples in the current training batch. Divide the difference by the mean to obtain a relative fluctuation, which is used as the statistical dispersion of the data in that dimension; repeat this process for all dimensions to obtain the dispersion sequence corresponding to the feature dimension.

[0019] As a further improvement of the present invention, the process of constructing an optimized objective function based on the prediction results includes the following steps: Based on the specific needs of different production stages, a comprehensive optimization objective function is set; when setting the function, reasonable constraints or target value ranges are set for each indicator, and the weights are adjusted; during the optimization process, adjustable structures and operating condition variables are determined. By using real-time operating data for rolling optimization, the optimal combination of control parameters that minimizes the comprehensive objective function is found; the optimization process is dynamically corrected based on real-time feedback process data; each optimization is recalculated based on the current operating conditions; The optimized control strategy is sent to the underlying DCS / PLC control system to drive each actuator to adjust according to the optimized instructions. After entering closed-loop operation, the deviation between the prediction and the actual situation is compared based on the real-time feedback process data, and the strategy parameters are corrected through feedback.

[0020] As a further improvement of the present invention, it also includes deploying various sensors at key locations in the smelting furnace, flue, waste heat boiler, acid production system, and steam pipeline network to collect real-time data on the temperature and flow rate of the smelting furnace flue gas. Concentration and Concentration process parameters and steam / hot water pressure, temperature and flow rate data of waste heat boiler.

[0021] To achieve the above objectives, the present invention also provides the following technical solution: A four-property synergistic digital design system for waste heat recovery in non-ferrous metal smelting is applied to the aforementioned four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting. The four-property synergistic digital design system for waste heat recovery in non-ferrous metal smelting includes: The waste heat data acquisition module is used to deploy various sensors at key locations in the smelting furnace, flue, waste heat boiler, acid production system, and steam pipeline network to collect real-time data on the temperature and flow rate of the flue gas from the smelting furnace. Concentration and Concentration process parameters and steam / hot water pressure, temperature, and flow rate data for waste heat boilers; The four-property optimization index module is used to build digital models based on process parameter data and waste heat boiler data, and to calculate the four-property optimization indexes of waste heat fluctuation, flow field uniformity, acid production stability and energy level matching in real time; it also retrieves process parameter data and waste heat boiler data to build time series analysis and prediction models. The optimal adjustment process combination module is used to input the key indicators of the four-property optimization in real time into the time series analysis and prediction model for prediction; construct the optimization objective function based on the prediction results; set constraints and weights; and obtain the optimal parameter combination through real-time process parameter data and waste heat boiler data.

[0022] 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 four-property collaborative digital design method for waste heat recovery in non-ferrous metal smelting as described above.

[0023] To achieve the above objectives, the present invention also provides the following technical solution: A storage medium storing program instructions, which, when executed by a processor, implement the four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting as described above.

[0024] This invention aims to provide a "four-property synergistic optimization method" based on mathematical models and evaluation indicators to overcome the defects in non-ferrous metal smelting flue gas waste heat recovery systems, such as large waste heat fluctuations, uneven flow fields, large fluctuations in acid production loads, and extensive energy utilization. The purpose of this invention is to achieve the following technical effects through synergistic optimization: Quantitative characterization and dynamic buffering control of flue gas waste heat fluctuations: This invention provides a set of quantitative indicators for the fluctuations of smelting flue gas waste heat, and achieves active buffering and dynamic adjustment of waste heat fluctuations through corresponding devices and control strategies. By controlling and optimizing system parameters, the impact of fluctuations on subsequent process systems is reduced from the source, maintaining stable operating conditions. Online evaluation of flue gas flow field uniformity and structure-operating condition linkage optimization: This invention establishes an index for evaluating the uniformity of flue gas velocity and temperature fields, and improves the flow field distribution inside equipment such as waste heat boilers in real time through a combination of adjustable structural design and operating condition parameter optimization. By precisely controlling the flow field distribution, the flow of flue gas within the boiler is ensured to meet predetermined uniformity requirements, thereby improving heat exchange efficiency and reducing slagging and ash accumulation problems. This invention provides a systematic evaluation of the stability of the acid production process and an integrated control system for smelting, waste heat recovery, and acid production. From a holistic system perspective, it evaluates the sensitivity of the acid production unit to fluctuations in flue gas parameters and integrates the smelting furnace, waste heat recovery system, and acid production unit into a unified control framework. By coordinating control measures across all stages, the impact of flue gas fluctuations on the acid production process is reduced, improving the stability of the acid production process and enhancing the flexibility and efficiency of load adjustments. Furthermore, this invention establishes a matching model between different temperature levels of waste heat from the flue gas and the applied equipment (such as steam turbines and heat exchangers). It quantitatively assesses the matching degree of heat energy at different temperature ranges and rationally allocates the utilization of heat energy at each level through optimization methods. By optimizing energy level matching, it avoids the consumption or direct emission of high-grade heat energy by low-grade operating conditions, thereby improving the cascade utilization efficiency of energy and maximizing the recovery and utilization of waste heat. Attached Figure Description

[0025] Figure 1 This is a schematic flowchart of one embodiment of the four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to the present invention. Figure 2 This is a schematic diagram of the steps involved in the real-time calculation of key indicators for optimizing four properties—waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching—in an embodiment of the present invention's synergistic digital design method for waste heat recovery in non-ferrous metal smelting. Figure 3 This is a schematic diagram illustrating the principle of real-time calculation of key indicators for optimizing four properties—waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching—in an embodiment of the present invention's four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting. Figure 4This is a schematic diagram illustrating the steps of constructing an optimized objective function based on prediction results in an embodiment of the four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to the present invention. Figure 5 This is a schematic diagram of the functional modules of an embodiment of the four-property synergistic digital design system for waste heat recovery in non-ferrous metal smelting according to the present invention. Figure 6 This is a schematic diagram of an embodiment of the four-property synergistic digital design system for waste heat recovery in non-ferrous metal smelting according to the present invention. Figure 7 This is a schematic diagram of the structure of an embodiment of the electronic device of the present invention; Figure 8 This is a schematic diagram of the structure of one embodiment of the storage medium of the present invention. Detailed Implementation

[0026] 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.

[0027] The terms "first," "second," and "third" used 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 that feature. In the description of this invention, "a plurality of" 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 indications also change accordingly. Furthermore, the terms "comprising" 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.

[0028] In this document, the term "embodiment" means 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.

[0029] like Figure 1 As shown, this embodiment provides an example of a four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting. In this embodiment, the four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting specifically includes the following steps: Step S1: Install various sensors at key locations such as the smelting furnace, flue, waste heat boiler, acid production system, and steam pipeline network to collect real-time data on the temperature and flow rate of the smelting furnace flue gas. Concentration and Process parameters such as concentration, as well as waste heat boiler data such as steam / hot water pressure, temperature, and flow rate; Step S2: Construct a digital model based on process parameter data and waste heat boiler data, and calculate in real time the four optimization indicators of waste heat fluctuation, flow field uniformity, acid production stability and energy level matching; retrieve process parameter data and waste heat boiler data to construct a time series analysis and prediction model; Step S3: Input the key indicators for real-time calculation of the four properties optimization into the time series analysis and prediction model for prediction; construct the optimization objective function based on the prediction results; set constraints and weights; and obtain the optimal parameter combination through real-time process parameter data and waste heat boiler data.

[0030] Preferably, data acquisition is a fundamental step in the optimization method of this embodiment; in order to understand the operating status of the smelting flue gas waste heat recovery system in real time, sensors need to be deployed at various key locations in the system to collect data on the temperature, flow rate, and other parameters of the smelting furnace flue gas in real time. concentration, Process parameters such as concentration; these parameters comprehensively reflect the main physical properties and chemical composition of the flue gas during the smelting process, serving as the basis for subsequent optimization control; in the waste heat boiler section, it is necessary to monitor the flow velocity and temperature fields of the flue gas in real time; through online temperature and flow velocity sensors, flow velocity and temperature data of key flue sections are collected. This data can be used to assess the uniformity of the flow field, thereby providing data support for subsequent flow field optimization. The inlet parameters of the acid production system also need to be monitored in real time, including... Concentration, flue gas flow rate, and temperature. These parameters directly affect... The conversion efficiency is crucial, therefore monitoring its fluctuations helps achieve precise control of the acid production process; in addition, data such as pressure, flow rate, and temperature of the steam and hot water networks should also be collected in real time; these data not only provide a basis for the rational allocation of energy, but also enable model prediction during system optimization to ensure the cascade utilization of thermal energy and energy recovery efficiency; for key parameters that are not directly measured (such as certain intermediate parameters, trends, etc.), by deploying multiple types of sensors at key production nodes (smelting furnace, flue, waste heat boiler, etc.), monitoring of flue gas process parameters (temperature, flow rate, etc.) is achieved. , The system synchronously and in real-time collects parameters of the waste heat medium (pressure, temperature, flow rate), forming a data sensing network covering the entire chain of "smelting—waste heat recovery—acid production—steam utilization." Based on real-time data, a digital model of "four optimization indicators" (waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching) is constructed, and dynamic prediction is performed in conjunction with a time series analysis model. This two-layer modeling mechanism of "real-time assessment + trend prediction" enables the system not only to depict the current state but also to predict the trend of operating conditions. The prediction results are transformed into an optimization objective function with weighted constraints, and the optimal parameter combination is searched through real-time data. This closed-loop optimization mechanism realizes an autonomous decision-making cycle from "monitoring—prediction—optimization—adjustment," enabling the system to have online adaptive tuning capabilities.

[0031] In summary, this embodiment employs soft measurement technology and state estimation methods. Online estimation of variables not directly monitored is performed through a system model, and data fusion technology enhances the system's responsiveness to complex operating conditions, acquiring complete system state information and providing accurate data support for subsequent optimization decisions. Through the coupled optimization of the "four properties indicators," the impact of waste heat fluctuations on the acid production process is effectively mitigated, flow field uniformity is enhanced, and dynamic synergy is achieved between the smelting, waste heat recovery, and acid production systems, significantly improving overall operational stability. Real-time calculation and prediction based on energy level matching allows for dynamic adjustment of steam / hot water parameters, ensuring precise matching between waste heat output and subsequent utilization stages, reducing energy waste, and improving overall system energy efficiency. The time series prediction model enables the system to identify abnormal trends in process parameters in advance, allowing for parameter adjustments before problems occur, transforming passive response into proactive intervention, and reducing the risk of unplanned downtime. The closed-loop optimization mechanism continuously accumulates operational data and optimization results, providing data support for process improvement and model iteration, driving the system towards continuous evolution towards greater efficiency and intelligence.

[0032] Furthermore, such as Figure 2 As shown, step S2 involves real-time calculation of key performance indicators for optimizing four properties: waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching. The specific steps include: Step S21: Construct a digital model based on process parameters and waste heat boiler data, calculate the instantaneous sensible heat power of smelting flue gas, and obtain its average value and standard deviation to obtain the waste heat fluctuation index; calculate the flow field uniformity index by measuring the flue gas velocity and temperature of the waste heat boiler to obtain the flow field uniformity. Step S22: Based on flue gas Concentration, temperature, and flow rate data are used to calculate the stability index of the acid production system and obtain the acid production stability. The waste heat of the flue gas is divided into energy levels, the recovered heat of different energy levels is calculated, and the proportion of each energy level to the total waste heat is calculated. The energy level matching is obtained through the proportion of the total waste heat. Step S23: Retrieve process parameter data and waste heat boiler data to construct a time series analysis and prediction model; retrieve historical time series data of the four optimization indicators to train the time series analysis and prediction model.

[0033] Preferably, this embodiment establishes mathematical models and evaluation index systems for four aspects: waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching. These models are used to quantitatively characterize each performance aspect and serve as the basis for optimized control. Waste heat fluctuation model and evaluation index: The instantaneous sensible heat power that can be recovered from smelting flue gas is defined as: in, For flue gas mass flow rate, The specific heat capacity of flue gas at constant pressure. For flue gas temperature, Use ambient temperature as a reference. Within a given time window. Inside, calculation average and standard deviation Based on this, the waste heat fluctuation index is defined. The larger the value, the stronger the waste heat fluctuation. For systems equipped with a flue gas buffer chamber (volume of...), the value indicates a more pronounced waste heat fluctuation. In cases where the buffer chamber is in a certain state, a dynamic energy balance model and a mass balance model can be established, for example: Dynamic energy balance equation:

[0034] mass conservation equation:

[0035] By appropriately designing the buffer chamber volume and adjusting the flue gas outlet flow rate This ensures that the temperature and flow rate fluctuations of the flue gas at the outlet of the buffer chamber are significantly smaller than those at the inlet, thereby achieving active buffering control of "waste heat fluctuations".

[0036] Flow field homogeneity model and evaluation index: Layout of key flue sections in equipment such as waste heat boilers At each measuring point, the local flue gas velocity is obtained. and temperature Calculate the average flow velocity at this cross section. and average temperature Define the velocity field uniformity index. and temperature field uniformity index For example, it is defined as: as well as The definition is similar.

[0037] index and The closer the value is to 1, the more uniform the flow field distribution. By optimizing the design of adjustable guide vanes, multi-layer flow equalization plates with variable orifice ratios, and adjustable swirl guide vanes, and combining this with adjusting operating parameters such as flue gas flow distribution, the flow field distribution can be improved during operation. , All values ​​remain above the set target value, ensuring the uniformity of the flue gas flow field from both structural and operational perspectives.

[0038] Acid production stability model and evaluation indicators: Flue gas parameters at key inlets of the acid production system (e.g., converter inlet) are selected as the main factors for evaluating the stability of the acid production process, including... Volume fraction flue gas temperature and flue gas flow rate Calculate the mean and standard deviation of these parameters within a given time window, and define the corresponding dimensionless volatility index. Taking into account the influence of various factors, a comprehensive index for acid production stability is constructed as follows: in, These are weighting coefficients (which can be determined based on the relative impact of each factor on stability). The smaller the value, the more stable the acid production process. Through the coordinated action of controlling the output and composition of flue gas at the smelting end, adjusting the intensity of waste heat recovery, and rationally distributing the load within the acid production system, the conversion rate is ensured. Maintaining a high and stable level throughout enhances the acid production process's resilience to upstream fluctuations.

[0039] Energy level matching model and evaluation index: The waste heat from the flue gas is divided into three energy level regions based on temperature: high-grade, medium-grade, and low-grade. For example, a high-grade region is defined. medium grade range Low-grade range ( (A threshold set according to operating conditions). Within the selected time window, the total heat recovered at each energy level is calculated and denoted as... Total residual heat Based on this, the proportion of actual recovered heat at each of the high, medium, and low energy levels can be calculated. Based on the downstream energy-consuming equipment's demand for heat energy of different grades, the optimal energy level allocation ratio is set for the target. And define an energy level matching deviation index: The smaller the value, the higher the degree of matching between the actual utilization of waste heat energy and the ideal target. By adjusting the production and distribution ratio of steam at each pressure level, controlling the heat exchanger outlet temperature, and adjusting the valve opening of each energy-consuming branch, the destination of heat energy at each grade is optimized, thus... Minimize as much as possible to achieve cascaded optimization of waste heat utilization.

[0040] Four-attribute synergistic comprehensive objective function: To comprehensively consider waste heat fluctuations, flow field uniformity, acid production stability, and energy level matching, this invention weights and integrates the above four types of indicators to construct a comprehensive evaluation objective function of "four-property synergy": in The weighting coefficients for each indicator item (which can be determined through experience or offline optimization based on production and operation priorities). The objective function is then optimized using online optimization algorithms or model predictive control methods. Perform multi-objective optimization to achieve the desired result while satisfying all operational constraints. Minimize these parameters to achieve an optimal state of synergistic optimization in waste heat fluctuation control, flow field uniformity, acid production process stability, and energy level matching utilization. Based on data acquisition, this invention establishes a mathematical model to calculate key performance indicators for the four optimization aspects in real time, including waste heat fluctuation. Flow field uniformity and Acid stability Energy level matching .

[0041] Waste heat fluctuation calculation, calculating the instantaneous sensible heat power of smelting flue gas. The average value and standard deviation of the waste heat fluctuation index are calculated to obtain the waste heat fluctuation index. : in, For flue gas mass flow rate, The specific heat capacity of flue gas at constant pressure. For flue gas temperature, Using ambient temperature as a reference, the volatility index is calculated through a dynamic model. In conjunction with the adjustment of the flue gas buffer chamber, active control of fluctuations can be achieved.

[0042] The flow field uniformity is calculated by measuring the flue gas velocity and temperature of the waste heat boiler and then calculating the flow field uniformity index. and These indicators reflect the uniformity of the flue gas flow field, providing target values ​​for flow field optimization. The optimized flow field indicators need to remain close to the set target values ​​to ensure maximum waste heat recovery efficiency.

[0043] Sulfuric acid production stability calculation based on flue gas Using concentration, temperature, and flow rate data, calculate the stability index of the acid production system. This indicator can reflect The impact of concentration fluctuations on the acid production process can be addressed by adjusting the operating conditions of the smelting furnace flue gas and waste heat boiler to achieve stable operation of the acid production process.

[0044] Energy level matching calculations divide flue gas waste heat into energy levels and calculate the recoverable heat at different energy levels. , and And calculate its proportion of the total waste heat. By optimizing the matching degree, high-grade thermal energy is ensured to be prioritized for steam turbine power generation and other high-efficiency energy-consuming equipment, thus avoiding the waste of low-grade thermal energy.

[0045] In this process, trend predictions are made for various indicators based on time series analysis and prediction models. By predicting key parameters such as flue gas volatility and flow field change trends, potential system fluctuations can be identified in advance, allowing for timely adjustments to control strategies and providing advanced information support for optimization decisions (see appendix for details). Figure 3 ).

[0046] In summary, this embodiment utilizes a four-property synergistic optimization method to organically combine four key performance indicators—waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching—providing a comprehensive optimized solution for smelting flue gas waste heat recovery systems. By dynamically adjusting the weights of each indicator and implementing real-time optimization control strategies, this invention effectively reduces the impact of flue gas fluctuations on subsequent process systems, improves flow field uniformity, ensures the stability of the acid production process, and maximizes the recovery of high-grade waste heat. The optimized system can maintain efficient and stable operation under different operating conditions, significantly improving overall energy recovery efficiency and production stability. The stability of the acid production system directly affects… The conversion efficiency and acid quality are improved. Through the acid production stability optimization method of this invention, the system can calculate in real time... The fluctuations in concentration, flue gas temperature, and flow rate are assessed, and the stability of the acid production process is comprehensively evaluated. By combining the regulation of flue gas output and composition at the smelting end, it is possible to effectively reduce... Fluctuations in concentration and temperature are managed to ensure that the conversion rate remains at a high and stable level. This optimization not only improves the yield and quality of acid but also enhances the adaptability of the acid production process to fluctuations in upstream flue gas, significantly improving the system's production stability. By establishing a flue gas waste heat energy level matching model, high, medium, and low-grade waste heat energy is rationally allocated to ensure optimal utilization of waste heat at different temperature ranges. By adjusting the destination of heat energy of each grade, high-grade heat energy is prioritized for use in high-efficiency equipment (such as steam turbine power generation and process heating), avoiding the waste of low-grade heat energy and maximizing the cascade utilization efficiency of energy. Energy level matching optimization not only improves the overall energy recovery rate of the system but also reduces energy waste, ensuring the green and low-carbon operation of the smelting process.

[0047] Furthermore, the process of training the time series analysis and prediction model in step S23 specifically includes the following steps: Step S231: Based on process parameter data and waste heat boiler data, construct a time series analysis and prediction model; use process parameter data as an external feature of the time series analysis and prediction model; retrieve the time series data of the four indicators from the historical time series data according to a unified timestamp to form a dataset; The time series analysis and prediction model is a data-driven machine learning / deep learning prediction framework, a supervised learning model used for multivariate time series prediction. It belongs to an encoder-predictor architecture: the encoder learns the temporal dependencies and patterns in historical sequences (including four performance indicators and process parameters); the predictor predicts the four performance indicators for future times based on the learned patterns. It follows a standard supervised learning process, including dataset partitioning (training set, validation set, test set), model parameter initialization, forward propagation, loss calculation, backpropagation optimization, iterative training, and final validation. The input is a multi-dimensional time series sample, consisting of two parts: the target sequence and historical time series data of the four performance optimization indicators over a past period (e.g., the past N hours); waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching. These indicators are a comprehensive representation of the system state, and the model predicts by learning its own changing patterns. The external feature sequence (auxiliary input) consists of the original process parameters and waste heat boiler data within the same time interval as the target sequence, including the temperature and flow rate of the smelting furnace flue gas. concentration, Concentration; pressure, temperature, flow rate, etc. of waste heat boilers and steam pipe networks can help models understand the process drivers behind index fluctuations, thereby improving the accuracy and interpretability of predictions.

[0048] Step S232: Divide the dataset into training set, validation set and test set; initialize the parameters of the time series analysis and prediction model; input the training set data into the time series analysis and prediction model, and make predictions based on the current parameters; Step S233: Calculate the loss between the predicted result and the true value, and adjust the parameters of the time series analysis and prediction model according to the loss value; iterate sequentially until the preset number of iterations is reached; calculate the various error indicators of the prediction results of the time series analysis and prediction model on the training set that has not participated in the training.

[0049] Preferably, this embodiment uses time series analysis and prediction models to predict the trends of various indicators. By predicting key parameters such as flue gas volatility and flow field change trends, potential system fluctuations can be identified in advance, allowing for timely adjustments to control strategies and providing advanced information support for subsequent optimization decisions.

[0050] In summary, this embodiment achieves proactive control of waste heat fluctuations by establishing a mathematical model of waste heat fluctuations in smelting flue gas and combining it with the design and adjustment of the flue gas buffer chamber. By monitoring parameters such as the mass flow rate and temperature of the flue gas in real time, the system can dynamically calculate and predict waste heat fluctuations. Adjustable buffer devices smooth short-cycle flue gas fluctuations, reducing the impact of fluctuations on downstream equipment (such as waste heat boilers and acid production systems) at the source, and avoiding the instability caused by experience-based adjustments in traditional systems. This optimization significantly improves the system's responsiveness, ensuring the efficient and stable operation of the entire system. This invention introduces a multi-objective optimization algorithm, using intelligent optimization methods to comprehensively optimize multiple objectives of the smelting flue gas waste heat recovery system. Utilizing intelligent optimization techniques such as genetic algorithms, it automatically explores a vast design space, finding the optimal solution under multiple design constraints. Through this optimized design, multiple performance indicators such as maximum furnace wall heat flux density, maximum stress, temperature uniformity, and metal recovery rate can be significantly improved while ensuring safety constraints and process requirements. Simultaneously, the optimized design reduces the workload of repeated trial calculations, improves design efficiency, and lowers design costs.

[0051] Furthermore, the process of adjusting the parameters of the time series analysis and prediction model based on the loss value in step S233 specifically includes the following steps: Step S2331: Construct a mechanism for calculating parameter adjustment amounts. Using the loss value as input, the mechanism calculates a set of adjustment amount data corresponding to the parameter dimensions of the time series analysis and prediction model. The adjustment amount data represents the value that each parameter should be modified in the current iteration to reduce the loss value. Step S2332: Combine the obtained adjustment data with the current parameter values ​​of the time series analysis and prediction model to generate a new set of parameter values; Step S2333: After completing the parameter update, start the next iteration; use the updated time series analysis and prediction model parameters to reprocess the training set data and obtain new prediction results; recalculate the loss value based on the new prediction results, and repeat the adjustment calculation and parameter update steps; the iteration process is repeated until the preset number of iterations is reached; after the iteration ends, fix the final parameters of the time series analysis and prediction model.

[0052] Preferably, this embodiment achieves automatic optimization of model parameters through loss value-driven parameter adjustment, thereby improving the performance of time series analysis and prediction models; by using loss value to quantify model error and continuously adjusting parameters through iterative methods, the model gradually converges to the optimal solution, thus obtaining more accurate prediction results.

[0053] Furthermore, step S2331, which involves obtaining a set of adjustment data corresponding to the parameter dimensions of the time series analysis and prediction model through calculation, specifically includes the following steps: Step S23311: Using the obtained loss value as input, generate a direction indicator for each time series analysis and prediction model parameter through inverse calculation associated with the internal computational structure of the time series analysis and prediction model; the direction indicator indicates whether the parameter value should be modified in the positive or negative direction in order to reduce the loss value. Step S23312: Simultaneously receive two inputs: one is the training sample data in the formed dataset, and the other is the intermediate processing result of the time series analysis and prediction model on the samples under the current parameters; match and quantify the features of the input data with the intermediate processing result, and calculate a non-negative intensity coefficient for each parameter with a determined direction; the intensity coefficient represents the urgency or influence weight of adjusting the parameter in the iteration. Step S23313: Pair the direction identifier of each parameter with the corresponding intensity coefficient, and merge the direction information and intensity information into a single scalar value; the scalar value is the adjustment amount data of the parameter.

[0054] Preferably, this embodiment can effectively improve the performance of time series analysis and prediction models. By precisely controlling the direction and intensity of parameter adjustment, the model can converge to a lower loss value more quickly during the iteration process, thereby improving the accuracy of prediction.

[0055] Furthermore, the process of matching and quantifying the features of the input data with the intermediate processing results in step S23312 specifically includes the following steps: Step S233121: Using the obtained parameter direction identifier and the provided intermediate processing results; according to the adjustment direction specified by the direction identifier, filter out the data segments related to the current role of each parameter from the intermediate processing results; perform aggregation operations on the data segments to generate an initial contribution factor for each parameter; The process involves several steps: First, data segment filtering is performed. Based on the parameter direction identifiers used, the computational nodes associated with the current parameter during model computation are identified. Then, based on the location of the computational nodes, all values ​​flowing through the nodes are extracted from the provided intermediate processing results to form a set of associated data segments for the parameter. Next, aggregation operations are performed to generate an initial contribution factor. For the values ​​in the associated data segment set, selective accumulation is performed based on their corresponding parameter direction identifiers. If the direction identifier is positive, all positive numbers in the associated data segment set are accumulated, ignoring negative numbers. If the direction identifier is negative, the absolute values ​​of all negative numbers in the associated data segment set are accumulated, ignoring positive numbers. Finally, the accumulated sum is divided by the total number of values ​​in the associated data segment set to obtain an average contribution, which serves as the initial contribution factor for the parameter. Step S233122: Using the provided training sample data, calculate the statistical dispersion of each dimension of external feature sequence data and target sequence data in the current training batch to obtain a dispersion sequence corresponding to the feature dimension. Step S233123: Multiply the contribution factor of each parameter by the dispersion value of the feature dimension of the input data associated with it to obtain a weighted contribution metric; perform a scaling operation on the weighted contribution metric set of all parameters; scale all metrics proportionally so that the maximum value is mapped to a preset baseline intensity value, and the remaining values ​​are mapped proportionally; the final value of each parameter obtained after this mapping is the intensity coefficient of the parameter.

[0056] Preferably, this embodiment enhances the model's ability to identify key information in the data by refining the matching degree between features and parameters; it generates initial contribution factors by aggregating parameter-related data fragments, thus preliminarily assessing the impact of parameters on the results; it quantifies the degree of feature variation in the sample data by calculating the statistical dispersion of the feature sequence data, providing a basis for subsequent weight adjustment; and it obtains a weighted contribution metric by multiplying the parameter contribution factor by the feature dispersion, and maps the metric to a preset baseline intensity value through scaling operations, making the parameter intensity coefficient more accurately reflect its importance in the model. Ultimately, this improves the model's sensitivity to input data features, optimizes parameter weight allocation, and thus enhances the model's prediction accuracy and generalization ability.

[0057] Furthermore, the process of calculating its statistical dispersion within the current training batch in step S233122 specifically includes the following steps: Step S2331221: Obtain the training sample data used, including the external feature sequence and the target sequence. For each dimension of the data in the sequence, calculate the maximum and minimum values ​​of that dimension in all samples of the current training batch; Step S2331222: Subtract the minimum value from the maximum value to obtain a difference; calculate the average value of this dimension of data in all samples of the current training batch; Step S2331223: Divide the aforementioned difference by the average value to obtain a relative fluctuation amount, which is used as the statistical dispersion of the data in this dimension; repeat this process for all dimensions to obtain the dispersion sequence corresponding to the feature dimension.

[0058] Preferably, in this embodiment Furthermore, such as Figure 4 As shown, step S3, which involves constructing the objective function based on the prediction results, specifically includes the following steps: Step S31: Based on the specific needs of different production stages, set a comprehensive optimization objective function; when setting the function, set reasonable constraints or target value ranges for each indicator and adjust the weights; during the optimization process, determine the adjustable structure and operating condition variables. Step S32: Through rolling optimization using real-time operating data, find the optimal combination of control parameters that minimizes the comprehensive objective function; the optimization process is dynamically corrected based on real-time feedback process data; each optimization is recalculated based on the current operating conditions; Step S33: The optimized control strategy is sent to the underlying DCS / PLC control system to drive each actuator to adjust according to the optimized instructions; after entering closed-loop operation, the deviation between the prediction and the actual is compared based on the real-time feedback process data, and the strategy parameters are corrected through feedback.

[0059] Preferably, in this embodiment, based on the specific needs of different production stages (e.g., start-up stage, steady-state production stage, high-load production stage, energy-saving operation stage, etc.), the present invention sets a comprehensive optimization objective function. The objective function consists of a weighted sum of four main performance metrics: in, The weighting coefficients for each indicator are determined empirically or through offline optimization methods, depending on the production stage. Furnace start-up stage: This stage requires high control over temperature fluctuations and flow field uniformity; therefore, the weighting coefficients are appropriately increased. and , The weights of [the factors]. Steady-state production stage: This stage requires focusing on ensuring the stability of acid production and the cascade utilization efficiency of waste heat, optimizing the objective function [of the objective function]. and The weight of the parameters. High-load production stage: During this stage, the smelting furnace is under heavy load, and the indicators related to waste heat recovery in the objective function (such as...) are optimized. and The weight of ) can be appropriately increased. Energy-saving operation stage: This stage emphasizes energy level matching and energy utilization efficiency, optimizing the objective function. To maximize the weight of waste heat recovery and ensure efficient utilization of low-grade heat energy, a series of adjustable structural and operational variables need to be determined during the optimization process. These variables include, but are not limited to: Smelting end variables: such as oxygen supply and furnace load, which directly affect the composition and temperature of flue gas, thus impacting waste heat recovery efficiency. Flue gas distribution structure: such as the opening of guide vanes, baffles, and bypass flue openings, which are used to control flue gas flow rate and flow field uniformity. Flue gas buffer chamber: adjusting the volume and outlet flow rate of the buffer chamber to achieve a smooth transition of flue gas fluctuations. Waste heat boiler variables: including the commissioning and switching of various heating surfaces of the waste heat boiler to ensure efficient recovery of heat energy of different grades. Sulfuric acid system variables: such as the load distribution of each absorption tower / converter to ensure efficient utilization of low-grade heat energy. Concentration stability control. Steam system variables: such as the extraction ratio of high-pressure, medium-pressure, and low-pressure steam and hot water output, optimizing energy level matching and energy utilization efficiency. These decision variables are used as control variables in the optimization calculation and dynamically adjusted in the objective function to ensure that the system can achieve optimal performance under different operating stages and conditions. When setting the objective function, reasonable constraints or target value ranges need to be set for each indicator to ensure that the system meets the requirements of safety, environmental protection, and efficiency during operation.

[0060] After setting constraints and weighting coefficients, this invention employs a multi-objective optimization algorithm or model predictive control method to solve for the decision variables. Through real-time operating data-driven rolling optimization, it seeks to find the comprehensive objective function that satisfies the condition. The optimal combination of control parameters is minimized. This optimization process dynamically corrects itself based on real-time feedback process data. Each optimization is recalculated based on the current operating conditions to ensure the control system has adaptive capabilities in the face of dynamic changes and maintains system stability and efficiency during long-term operation. The control strategy obtained from the optimization solution is issued to the underlying DCS / PLC control system, driving various actuators (such as guide vane drive devices, flue damper valves, water spray regulating valves, high and low pressure steam valves, blower frequency converters, etc.) to adjust according to the optimization instructions. After the system enters closed-loop operation, it adjusts based on real-time feedback process data (such as flue gas flow rate, temperature, etc.). By comparing the predicted and actual values ​​(such as concentration), the system further corrects the strategy parameters through feedback. This adaptive correction mechanism ensures that the system can continuously optimize during long-term operation, keeping all performance indicators within the optimal range, and achieving stable, efficient, and sustainable operation of the smelting flue gas waste heat recovery system.

[0061] In summary, in this embodiment, the non-uniformity of the flow field in key equipment such as waste heat boilers is one of the main reasons for low heat recovery efficiency and equipment failure. This invention uses a flow field uniformity evaluation model to calculate and optimize the uniformity index of the flue gas flow field in real time. By employing adjustable guide vanes and swirl vanes, combined with the adjustment of flue gas flow rate, uniform distribution of the velocity and temperature fields is achieved. The optimized flow field distribution ensures uniform heating of the heating surface, avoiding local overheating, ash accumulation, and slagging problems, significantly improving waste heat recovery efficiency and reducing equipment maintenance requirements.

[0062] like Figure 4 As shown, this embodiment also provides an embodiment of a four-property collaborative digital design system for waste heat recovery in non-ferrous metal smelting. In this embodiment, the four-property collaborative digital design system for waste heat recovery in non-ferrous metal smelting is applied to the four-property collaborative digital design method for waste heat recovery in non-ferrous metal smelting as described in the above embodiment. The four-property collaborative digital design system for waste heat recovery in non-ferrous metal smelting includes: The waste heat data acquisition module 1 is used to deploy various sensors at key locations such as the smelting furnace, flue, waste heat boiler, acid production system, and steam pipeline network to collect real-time data on the temperature and flow rate of the smelting furnace flue gas. Concentration and Process parameters such as concentration, as well as waste heat boiler data such as steam / hot water pressure, temperature, and flow rate; The Four-Property Optimization Index Module 2 is used to build a digital model based on process parameter data and waste heat boiler data, and to calculate in real time the four-property optimization indexes such as waste heat fluctuation, flow field uniformity, acid production stability and energy level matching; it also retrieves process parameter data and waste heat boiler data to build time series analysis and prediction models. The optimal adjustment process combination module 3 is used to input the key indicators of the four-property optimization in real time into the time series analysis and prediction model for prediction; construct the optimization objective function based on the prediction results; set constraints and weights; and obtain the optimal parameter combination through real-time process parameter data and waste heat boiler data.

[0063] Preferably, the process requirements and physical property data input module in this embodiment is responsible for collecting and inputting process requirement parameters for smelting flue gas, including basic data such as smelting furnace type (e.g., molten pool furnace, side-blown furnace, flash furnace, etc.), flue gas composition, temperature, and flow rate. In addition, it includes design parameters for the waste heat boiler and required heat recovery data, operating conditions of the acid production system, and energy requirement parameters for various energy-consuming equipment (e.g., steam header, hot water network, etc.). This input data provides a foundation for subsequent modeling and optimization.

[0064] The waste heat fluctuation analysis and control module is responsible for constructing a mathematical model of waste heat fluctuations in smelting flue gas and for real-time monitoring and analysis of flue gas temperature, flow rate, and... Fluctuations in concentration. By establishing a dynamic energy balance model and mass conservation equation, the impact of flue gas fluctuations on the waste heat recovery system is evaluated, and active buffering and optimized regulation of fluctuations are achieved through control strategies (such as adjustable flue gas buffer chamber volume and flow rate).

[0065] The flow field uniformity assessment and optimization module evaluates the uniformity of the flue gas flow field by real-time acquisition and analysis of flow field data (including flow velocity and temperature distribution) inside the waste heat boiler. Combined with adjustable guide vanes, swirl vanes, and other structural designs that can adjust the flow field distribution, the module optimizes the flue gas flow path, making the temperature and velocity distribution of the flue gas more uniform, thereby improving waste heat recovery efficiency and reducing slagging and ash accumulation problems.

[0066] Sulfuric Acid Production Stability Optimization Module. This module focuses on the stability of the sulfuric acid production process, especially... The impact of fluctuations in concentration, flue gas temperature, and flow rate on acid production efficiency. This is achieved by analyzing flue gas inlet parameters (such as...). Real-time monitoring of concentration and flue gas flow rate is used to calculate and optimize the control strategy of the acid production system, ensuring stable flue gas treatment and consistent acid production and quality.

[0067] The energy level matching and optimization configuration module is responsible for establishing a matching model for flue gas waste heat energy levels. Based on the flue gas energy (high-grade, medium-grade, low-grade) at different temperature ranges, it allocates the energy demand of different equipment (such as steam headers, heat exchangers, and steam turbines). By optimizing the allocation of heat at each energy level, it ensures that high-grade heat energy is fully utilized, avoids the waste of low-grade heat energy, and improves overall energy efficiency.

[0068] The Four-Performance Synergistic Optimization Module integrates four evaluation indicators—waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching—with weighted fusion to construct a synergistic comprehensive objective function. Based on the weight coefficients of each indicator, a multi-objective optimization algorithm is used to solve the problem. The goal is to achieve the optimal state of synergistic optimization for the four key performance indicators, thereby ensuring the overall efficiency and stability of the smelting, waste heat recovery, and acid production processes.

[0069] The data management and optimization decision-making module integrates real-time data acquisition, processing, and management functions. It acquires key data on smelting flue gas through an online monitoring system and compares this data with model calculations to provide a basis for subsequent optimization decisions. The optimization decision-making module combines advanced process control methods (such as model predictive control) to adjust various control parameters in real time, ensuring continuous optimization of the system under dynamic operating conditions.

[0070] The visualization, interaction, and feedback module provides a graphical user interface (GUI) to enhance the visibility and interactivity of the system. This allows for the intuitive presentation of various performance indicators, control parameters, and optimization results of the smelting flue gas waste heat recovery system. Users can adjust the control parameters according to actual needs and optimize the operating parameters through the feedback mechanism, forming a closed-loop optimization system.

[0071] The modules of the above method are implemented in an integrated digital optimization control system via computer programs. This system can collect and process various data from the smelting, waste heat recovery, and acid production processes in real time, and automatically perform multi-objective optimization and decision adjustments while ensuring safety and process requirements. Through this digital optimization control platform, the "four-property synergistic optimization method" of this invention can effectively improve the overall performance of the smelting flue gas waste heat recovery system, achieving efficient energy recovery and green, low-carbon operation.

[0072] System Structure: The method of this invention is applicable to typical non-ferrous smelting flue gas waste heat recovery and acid production system structures (taking copper molten pool smelting process as an example), including the following parts: Smelting flue gas sources: Smelting furnaces such as molten pool furnaces, side-blown furnaces, flash furnaces, and bottom-blown furnaces, which generate high-temperature flue gas containing... The system includes: flue gas; primary flue and adjustable distribution device: a high-temperature flue drawn from the smelting furnace, and adjustable guide vanes, throttling valves or swirl vanes configured therein, bypass flue and flue gas mixing chamber, etc., used to distribute and regulate flue gas flow; waste heat boiler and heating surface: a multi-stage heating surface structure including a high-temperature radiation section, convection section and economizer / water saver of the waste heat boiler, used to recover heat energy of different grades in the flue gas; flue gas buffer and mixing device: a flue gas mixing chamber or buffer section of a certain volume is set at an appropriate position in the flue before entering the waste heat boiler, used to reduce short-cycle flue gas parameter fluctuations and achieve preliminary mixing equilibrium; flue gas purification, cooling and acid production system: including flue gas cooling tower, electrostatic precipitator, dry / wet suction system, converter, absorption tower, etc., to cool and purify the flue gas before sending it to the acid production process, and to... Converted and absorbed to produce sulfuric acid; Multi-pressure level steam and hot water network: High-pressure, medium-pressure, and low-pressure steam headers and low-temperature hot water network from waste heat boilers, used for energy utilization for various purposes (such as steam turbine power generation, process heating, heating, etc.); Measurement and control system: Temperature, pressure, flow rate, etc., located in key positions. , Online monitoring sensors for concentration and other parameters, along with a DCS / PLC control system, combined with advanced process control (APC) and optimization decision-making modules, enable monitoring and coordinated control of the aforementioned components. Four-dimensional synergistic mathematical model and evaluation indicators: This invention establishes a mathematical model and evaluation indicator system for four aspects: waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching. These are used to quantitatively characterize each performance aspect and serve as the basis for optimized control. Waste heat fluctuation model and evaluation index: Define the instantaneous sensible heat power that can be recovered from smelting flue gas as: ,in, For flue gas mass flow rate, The specific heat capacity of flue gas at constant pressure. For flue gas temperature, Use ambient temperature as a reference. Within a given time window. Inside, calculation average and standard deviation Based on this, the waste heat fluctuation index is defined. The larger the value, the stronger the waste heat fluctuation. For systems equipped with a flue gas buffer chamber (volume of...), the value indicates a more pronounced waste heat fluctuation. In cases where the buffer chamber is in a certain state, a dynamic energy balance model and a mass balance model can be established, for example: Dynamic energy balance equation:

[0073] mass conservation equation:

[0074] By appropriately designing the buffer chamber volume and adjusting the flue gas outlet flow rate This ensures that the temperature and flow rate fluctuations of the flue gas at the outlet of the buffer chamber are significantly smaller than those at the inlet, thereby achieving active buffering control of "waste heat fluctuations".

[0075] Flow field uniformity model and evaluation index: Layout of key flue sections in equipment such as waste heat boilers At each measuring point, the local flue gas velocity is obtained. and temperature Calculate the average flow velocity at this cross section. and average temperature Define the velocity field uniformity index. and temperature field uniformity index For example, it is defined as: as well as The definition is similar.

[0076] index and The closer the value is to 1, the more uniform the flow field distribution. By optimizing the design of adjustable guide vanes, multi-layer flow equalization plates with variable orifice ratios, and adjustable swirl guide vanes, and combining this with adjusting operating parameters such as flue gas flow distribution, the flow field distribution can be improved during operation. , All values ​​remain above the set target value, ensuring the uniformity of the flue gas flow field from both structural and operational perspectives.

[0077] Sulfuric Acid Production Stability Model and Evaluation Indicators: Flue gas parameters at key inlets of the sulfuric acid production system (e.g., converter inlet) are selected as the main factors for evaluating the stability of the sulfuric acid production process, including... Volume fraction flue gas temperature and flue gas flow rate Calculate the mean and standard deviation of these parameters within a given time window, and define the corresponding dimensionless volatility index. Taking into account the influence of various factors, a comprehensive index for acid production stability is constructed as follows: in, These are weighting coefficients (which can be determined based on the relative impact of each factor on stability). The smaller the value, the more stable the acid production process. Through the coordinated action of controlling the output and composition of flue gas at the smelting end, adjusting the intensity of waste heat recovery, and rationally distributing the load within the acid production system, the conversion rate is ensured. Maintaining a high and stable level throughout enhances the acid production process's resilience to upstream fluctuations.

[0078] Energy level matching model and evaluation index: The waste heat of flue gas is divided into three energy level regions according to temperature: high-grade, medium-grade, and low-grade. For example, a high-grade region is defined. medium grade range Low-grade range ( (A threshold set according to operating conditions). Within the selected time window, the total heat recovered at each energy level is calculated and denoted as... Total residual heat Based on this, the proportion of actual recovered heat at each of the high, medium, and low energy levels can be calculated. Based on the downstream energy-consuming equipment's demand for heat energy of different grades, the optimal energy level allocation ratio is set for the target. And define an energy level matching deviation index: The smaller the value, the higher the degree of matching between the actual utilization of waste heat energy and the ideal target. By adjusting the production and distribution ratio of steam at each pressure level, controlling the heat exchanger outlet temperature, and adjusting the valve opening of each energy-consuming branch, the destination of heat energy at each grade is optimized, thus... Minimize as much as possible to achieve cascaded optimization of waste heat utilization.

[0079] The four-property synergy comprehensive objective function: To comprehensively consider waste heat fluctuations, flow field uniformity, acid production stability, and energy level matching, this invention weights and fuses the above four types of indicators to construct a "four-property synergy" comprehensive evaluation objective function: in The weighting coefficients for each indicator item (which can be determined through experience or offline optimization based on production and operation priorities). The objective function is then optimized using online optimization algorithms or model predictive control methods. Perform multi-objective optimization to achieve the desired result while satisfying all operational constraints. Minimize these parameters to achieve the optimal state of synergistic optimization in all aspects, including waste heat fluctuation control, flow field uniformity, acid production process stability, and energy level matching utilization.

[0080] In summary, this embodiment, through the proposed optimized control method, enables the system to provide real-time feedback on changes in operating conditions and automatically adjust the control strategy, ensuring the stability and efficiency of the smelting flue gas waste heat recovery system during long-term operation. Especially when facing changes in different production stages, the system can automatically adapt to different operational needs, achieving the optimal state of synergistic optimization of the four properties by dynamically adjusting the weight coefficients in the objective function. This adaptive control not only ensures the efficient operation of the system under various operating conditions but also provides strong support for energy conservation, emission reduction, and green production. Through an integrated digital design platform and intelligent optimized control methods, synergistic optimization of the smelting process, waste heat recovery, acid production process, and energy utilization is achieved. This platform can not only monitor and adjust system parameters in real time but also automatically adjust the optimization strategy according to different operating conditions, ensuring a high degree of synergy among all aspects of the smelting flue gas waste heat recovery system during operation. This integrated and intelligent optimized design enables the system to achieve optimal economic and environmental benefits, providing a practical solution for the green and low-carbon transformation of the non-ferrous metal smelting industry (see appendix for specific principles). Figure 6 ). like Figure 7 As shown, this embodiment provides an embodiment of an electronic device 4, which includes a processor 41 and a memory 42 coupled to the processor 41.

[0081] The memory 42 stores program instructions for implementing the four-property collaborative digital design method for waste heat recovery in non-ferrous metal smelting according to any of the above embodiments.

[0082] The processor 41 is used to execute the program instructions stored in the memory 42 to lay out the four-property collaborative digital design method for waste heat recovery in non-ferrous metal smelting.

[0083] 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.

[0084] Furthermore, Figure 8 This 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.

[0085] 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.

[0086] 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.

[0087] 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 four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting, characterized in that, The aforementioned four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting includes: A digital model is constructed based on process parameter data and waste heat boiler data to calculate in real time the four optimization indicators of waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching; and a time series analysis and prediction model is constructed by retrieving process parameter data and waste heat boiler data. The key indicators for real-time computational optimization of the four properties are input into the time series analysis and prediction model for prediction; an optimization objective function is constructed based on the prediction results; constraints are set and weights are configured; and the optimal parameter combination is obtained through real-time process parameter data and waste heat boiler data.

2. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 1, characterized in that, The process of real-time calculation of key indicators for optimizing four properties—waste heat fluctuation, flow field uniformity, acid production stability, and energy level matching—includes the following steps: A digital model is constructed based on process parameters and waste heat boiler data to calculate the instantaneous sensible heat power of smelting flue gas, and to obtain its average value and standard deviation, thus obtaining the waste heat fluctuation index; by measuring the flue gas velocity and temperature of the waste heat boiler, the flow field uniformity index is calculated, thus obtaining the flow field uniformity. Based on flue gas Concentration, temperature, and flow rate data are used to calculate the stability index of the acid production system and obtain the acid production stability. The waste heat of the flue gas is divided into energy levels, the recovered heat of different energy levels is calculated, and the proportion of each energy level to the total waste heat is calculated. The energy level matching is obtained through the proportion of the total waste heat. Data on process parameters and waste heat boilers were retrieved to construct time series analysis and prediction models; historical time series data of the four optimization indicators were retrieved to train the time series analysis and prediction models.

3. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 2, characterized in that, The process of training a time series analysis and prediction model includes the following steps: Based on process parameter data and waste heat boiler data, a time series analysis and prediction model is constructed; the process parameter data is used as an external feature of the time series analysis and prediction model; in the historical four-indicator time series data, the past preset time series data of the four-indicator are retrieved according to a unified timestamp to form a dataset; The dataset is divided into training, validation, and test sets; the parameters of the time series analysis and prediction model are initialized; the training set data is input into the time series analysis and prediction model, and predictions are made based on the current parameters; Calculate the loss between the predicted result and the true value, and adjust the parameters of the time series analysis and prediction model according to the loss value; iterate sequentially until the preset number of iterations is reached; calculate various error indicators of the prediction results of the time series analysis and prediction model on the training set that has not participated in the training.

4. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 3, characterized in that, The process of adjusting the parameters of time series analysis and forecasting models based on the loss value includes the following steps: Construct a mechanism for calculating parameter adjustments. Take the loss value as input and calculate a set of adjustment data corresponding to the parameter dimensions of the time series analysis and prediction model. The adjustment data represents the value that each parameter should be modified in the current iteration to reduce the loss value. The obtained adjustment data is combined with the current parameter values ​​of the time series analysis and forecasting model to generate a new set of parameter values; After the parameter update is completed, the next iteration begins; the training set data is reprocessed using the updated time series analysis and prediction model parameters to obtain new prediction results; the loss value is recalculated based on the new prediction results, and the adjustment calculation and parameter update steps are repeated; the iteration process is repeated until the preset number of iterations is reached; after the iteration terminates, the final parameters of the time series analysis and prediction model are fixed.

5. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 4, characterized in that, The process of obtaining a set of adjustment data corresponding to the parameter dimensions of time series analysis and prediction models through calculation includes the following steps: Using the obtained loss value as input, a direction indicator is generated for each parameter of the time series analysis and prediction model through inverse calculation associated with the internal computational structure of the time series analysis and prediction model; the direction indicator indicates whether the parameter value should be modified in the positive or negative direction in order to reduce the loss value. Simultaneously, it receives two inputs: one is the training sample data in the formed dataset, and the other is the intermediate processing results of the samples under the current parameters of the time series analysis and prediction model; it matches and quantifies the features of the input data with the intermediate processing results, and calculates a non-negative intensity coefficient for each parameter with a determined direction; the intensity coefficient represents the urgency or influence weight of adjusting the parameter in the iteration. The directional identifier of each parameter is paired with the corresponding intensity coefficient, and the directional information and intensity information are merged into a single scalar value; the scalar value is the adjustment amount of the parameter.

6. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 5, characterized in that, The process of matching and quantifying the features of input data with intermediate processing results includes the following steps: Using the obtained parameter direction identifiers and the provided intermediate processing results; based on the adjustment direction specified by the direction identifiers, filter out the data segments related to the current role of each parameter from the intermediate processing results; perform aggregation operations on the data segments to generate an initial contribution factor for each parameter; Using the provided training sample data, calculate the statistical dispersion of each external feature sequence data and target sequence data in the current training batch to obtain a dispersion sequence corresponding to the feature dimension. The contribution factor of each parameter is multiplied by the dispersion value of the feature dimension of the input data associated with it to obtain a weighted contribution metric. A scaling operation is performed on the weighted contribution metrics of all parameters. All metrics are scaled proportionally so that the maximum value is mapped to a preset baseline intensity value, and the remaining values ​​are mapped proportionally. The final value of each parameter obtained after this mapping is the intensity coefficient of the parameter.

7. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 6, characterized in that, The process of calculating its statistical dispersion within the current training batch includes the following steps: Take the training sample data used, which includes the external feature sequence and the target sequence; for each dimension of the data in the sequence; calculate the maximum and minimum values ​​of that dimension of the data in all samples of the current training batch. Subtract the minimum value from the maximum value to obtain a difference; calculate the average value of this dimension of data across all samples in the current training batch. Divide the difference by the mean to obtain a relative fluctuation, which is used as the statistical dispersion of the data in that dimension; repeat this process for all dimensions to obtain the dispersion sequence corresponding to the feature dimension.

8. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 1, characterized in that, The process of constructing an optimization objective function based on the prediction results includes the following steps: Based on the specific needs of different production stages, a comprehensive optimization objective function is set; when setting the function, reasonable constraints or target value ranges are set for each indicator, and the weights are adjusted; during the optimization process, adjustable structures and operating condition variables are determined. By using real-time operating data for rolling optimization, the optimal combination of control parameters that minimizes the comprehensive objective function is found; the optimization process is dynamically corrected based on real-time feedback process data; each optimization is recalculated based on the current operating conditions; The optimized control strategy is sent to the underlying DCS / PLC control system to drive each actuator to adjust according to the optimized instructions. After entering closed-loop operation, the deviation between the prediction and the actual situation is compared based on the real-time feedback process data, and the strategy parameters are corrected through feedback.

9. The four-property synergistic digital design method for waste heat recovery in non-ferrous metal smelting according to claim 1, characterized in that, It also includes deploying various sensors at key locations in the smelting furnace, flue, waste heat boiler, acid production system, and steam pipeline network to collect real-time data on the temperature and flow rate of the smelting furnace flue gas. Concentration and Concentration process parameters and steam / hot water pressure, temperature and flow rate data of waste heat boiler.

10. A four-property collaborative digital design system for waste heat recovery in non-ferrous metal smelting, applied to the four-property collaborative digital design method for waste heat recovery in non-ferrous metal smelting as described in any one of claims 1 to 9, characterized in that, The non-ferrous metal smelting waste heat recovery four-property coordinated digital design system includes: The waste heat data acquisition module is used to deploy various sensors at key locations in the smelting furnace, flue, waste heat boiler, acid production system, and steam pipeline network to collect real-time data on the temperature and flow rate of the flue gas from the smelting furnace. Concentration and Concentration process parameters and steam / hot water pressure, temperature, and flow rate data for waste heat boilers; The four-property optimization index module is used to build digital models based on process parameter data and waste heat boiler data, and to calculate the four-property optimization indexes of waste heat fluctuation, flow field uniformity, acid production stability and energy level matching in real time; it also retrieves process parameter data and waste heat boiler data to build time series analysis and prediction models. The optimal adjustment process combination module is used to input the key indicators of the four-property optimization in real time into the time series analysis and prediction model for prediction; construct the optimization objective function based on the prediction results; set constraints and weights; and obtain the optimal parameter combination through real-time process parameter data and waste heat boiler data.

Images