Glass kiln flue gas treatment method and device, electronic equipment and storage medium

Abstract

The invention provides a glass kiln flue gas treatment method and device, electronic equipment and a storage medium, and relates to the technical field of flue gas treatment.The glass kiln flue gas treatment method is characterized in that glass formula information of production plan scheduling is obtained; predicting the smoke component change trend after formula switching based on the incidence relation between the glass formula components and the pollutant concentration in the smoke; calculating and setting target operation parameters of each stage of processing unit; acquiring flue gas component online monitoring data and real-time operation parameters of each stage of flue gas treatment unit; calculating the component deviation between the actual flue gas component and the predicted flue gas component; calculating parameter deviation; and according to the component deviation and the parameter deviation, correcting the operating parameter set value of the downstream flue gas treatment unit. The glass kiln flue gas treatment method and device, the electronic equipment and the storage medium have the advantages that flue gas component changes can be quickly responded, collaborative optimization of multiple stages of treatment units is achieved, and the overall treatment efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of flue gas treatment, and in particular to a glass kiln flue gas treatment method, device, electronic equipment and storage medium. Background Art

[0002] In order to meet increasingly stringent environmental emission standards, glass factories generally adopt multi-stage series flue gas treatment systems. These systems integrate multiple treatment units to synergistically remove various pollutants such as particulate matter, sulfur dioxide, nitrogen oxides, etc. in the flue gas.

[0003] At present, the widely used flue gas treatment systems usually adopt relatively independent control modes for each level of treatment units, such as denitrification units, dust removal units, desulfurization units, etc. This means that the adjustment of the operating parameters of each unit mainly depends on its own outlet flue gas monitoring data or preset experience values, and there is a lack of effective linkage and coordination between units. Under relatively stable production conditions, this independent control mode can still maintain a certain treatment effect. However, in the production mode where the glass formula is frequently switched, its disadvantages are fully exposed.

[0004] When the upstream kiln quickly switches the glass formula, the flue gas composition and concentration will fluctuate rapidly and significantly. This sudden disturbance is transmitted to the inlet of the downstream flue gas treatment system, causing the inlet flue gas characteristics of each treatment unit to change frequently. Due to the independence of each unit control, it is difficult for the downstream flue gas treatment unit to quickly respond to the drastic fluctuations in the upstream flue gas composition. The adjustment of operating parameters often lags behind the flue gas changes, resulting in a decrease in the overall treatment efficiency of the system, and even short-term excessive emissions.

[0005] In order to cope with flue gas fluctuations and maintain emission compliance, operators have to invest a lot of energy and frequently manually adjust the operating parameters of each level of treatment units. This manual intervention method is not only cumbersome and labor-intensive, but also difficult to ensure the timeliness and accuracy of adjustments, which can easily cause energy waste and secondary pollution. More importantly, under conditions where flue gas components fluctuate rapidly, independent control modes make it difficult to achieve coordinated optimization among multi-level treatment units, unable to fully tap the processing potential of the entire system, and difficult to balance the dual goals of rapid responsiveness and stable emission compliance.

[0006] The existing technology has the following major defects: each processing unit adopts an independent control mode and lacks effective linkage and coordination; it is difficult for the downstream flue gas treatment unit to quickly respond to the drastic fluctuations in the upstream flue gas composition; the manual intervention adjustment method is cumbersome and inaccurate, and easily causes energy waste; it is difficult to achieve coordinated optimization among multi-level processing units, and the processing potential of the entire system cannot be fully utilized; it is difficult to achieve the dual goals of rapid responsiveness and stable emission compliance.

[0007] In view of the above problems, the existing technology needs to be improved urgently. Summary of the Invention

[0008] The purpose of this application is to provide a method, device, electronic device and storage medium for treating glass furnace flue gas, which has the advantages of being able to quickly respond to changes in flue gas composition, realizing collaborative optimization of multi-level treatment units, and improving the overall treatment efficiency.

[0009] This application provides a method for treating glass furnace flue gas, and the technical solution is as follows: The steps of this method include: Obtain the glass formula information scheduled in the production plan; Based on the glass formula information, predict the change trend of flue gas composition after formula switching according to the correlation between glass formula components and pollutant concentrations in the flue gas; According to the predicted change trend of flue gas composition after formula switching and the process characteristics and operation parameter thresholds of each level of flue gas treatment unit, calculate and set the target operation parameters of each treatment unit; Obtain the online monitoring data of flue gas composition and real-time operation parameters of each level of flue gas treatment unit; Calculate the composition deviation between the actual flue gas composition and the predicted flue gas composition according to the online monitoring data of flue gas composition and the predicted change trend of flue gas composition after formula switching; Calculate the parameter deviation according to the target operation parameters and the real-time operation parameters; Modify the set value of the operation parameters of the downstream flue gas treatment unit according to the composition deviation and the parameter deviation.

[0010] Furthermore, this application also proposes that the steps of predicting the change trend of flue gas composition after formula switching according to the correlation between glass formula components and pollutant concentrations in the flue gas include: Establish a polynomial regression model between the glass formula component vector and the flue gas composition concentration vector, and the model function form is as follows: cj = f_j(R) = ∑_{i=1}^{n} a_{ji} * ri + ∑_{i=1}^{n} ∑_{k=1}^{n} b_{jik} * ri * rk + d_j, where cj is the predicted concentration of the jth flue gas component, R = [r1, r2,..., rn] is the glass formula component vector, ri represents the proportion of the ith formula component, a_{ji} is the linear regression coefficient, b_{jik} is the quadratic regression coefficient, indicating the influence of the interaction between the ith and kth formula components on the concentration of the jth flue gas component, and d_j is the bias term; Convert the glass formula information scheduled in the production plan into the glass formula component vector R; Substitute the glass formula composition vector R into the polynomial regression model to calculate the flue gas composition concentration vector, which is used as the prediction result of the change trend of the flue gas composition after the formula switch.

[0011] Furthermore, the present application also proposes that the step of calculating and setting the target operating parameters of each treatment unit according to the predicted change trend of the flue gas composition after the formula switch and the process characteristics and operating parameter thresholds of each flue gas treatment unit includes: Obtain the pollutant removal efficiency curves of each flue gas treatment unit under different flue gas compositions, and the relationship between the operating parameters of each flue gas treatment unit and the pollutant removal efficiency; Based on the predicted change trend of the flue gas composition after the formula switch, combined with the pollutant removal efficiency curves of each flue gas treatment unit, calculate the expected pollutant removal amounts of each flue gas treatment unit under different operating parameters; According to the operating parameter thresholds of each flue gas treatment unit, determine the upper and lower bound constraints of the operating parameters of each flue gas treatment unit; With the goal of minimizing the total energy consumption, while satisfying that the emission concentrations of each pollutant are lower than the emission standards, and the upper and lower bound constraints of the operating parameters of each unit, calculate the target operating parameters of each flue gas treatment unit.

[0012] Furthermore, the present application also proposes that the step of calculating the target operating parameters of each flue gas treatment unit with the goal of minimizing the total energy consumption, while satisfying that the emission concentrations of each pollutant are lower than the emission standards, and the upper and lower bound constraints of the operating parameters of each unit includes: Construct a multi-objective optimization model with the objective function of minimizing the total energy consumption. The objective function is expressed as: min ∑_{i=1}^{m} P_i(u_i), where P_i(u_i) is the energy consumption function of the i-th flue gas treatment unit, u_i is the operating parameter of the i-th flue gas treatment unit, and m is the number of flue gas treatment units; Set the pollutant emission concentration constraint, expressed as: g_j(u_1, u_2, ..., u_m) ≤ C_j, where g_j(u_1, u_2, ..., u_m) is the emission concentration function of the j-th pollutant, and C_j is the emission standard of the j-th pollutant; Determine the upper and lower bound constraints of the operating parameters of each flue gas treatment unit, expressed as: u_{i,min} ≤ u_i ≤ u_{i,max}, where u_{i,min} and u_{i,max} are the lower and upper limits of the operating parameter of the i-th flue gas treatment unit respectively; Iteratively solve the multi-objective optimization model. In each iteration, calculate the gradients of the objective function and the constraint conditions, and update the operating parameters of each flue gas treatment unit until the convergence condition is satisfied to obtain the target operating parameters of each flue gas treatment unit.

[0013] Further, the present application also proposes that the step of correcting the set value of the operating parameters of the downstream flue gas treatment unit according to the composition deviation and parameter deviation includes: Obtain the control target type and the corresponding control target priority, where the control target type includes at least one of pollutant removal efficiency, energy consumption, and equipment life; Determine the first weight of the pollutant removal efficiency deviation, the second weight of the energy consumption deviation, and the third weight of the equipment life deviation according to the control target priority; Calculate the pollutant removal efficiency deviation according to the deviation between the actual flue gas composition and the predicted flue gas composition, and calculate the energy consumption deviation and equipment life deviation according to the deviation between the actual operating parameters and the target operating parameters; Calculate the correction amount of the operating parameters of each stage of the downstream flue gas treatment unit based on the first weight, the second weight, the third weight, the pollutant removal efficiency deviation, the energy consumption deviation, and the equipment life deviation; Superimpose the correction amount of the operating parameters on the set value of the operating parameters of each stage of the flue gas treatment unit to obtain the corrected set value of the operating parameters.

[0014] Further, the present application also proposes that the step of correcting the set value of the operating parameters of the downstream flue gas treatment unit according to the composition deviation and parameter deviation includes: Define the rolling optimization window length, which represents the number of future time steps that need to be predicted and optimized starting from the current moment; At each control moment, perform the following steps: Obtain the predicted data of the flue gas composition change trend and the dynamic response characteristic parameters of the treatment unit at the next k moments defined by the rolling optimization window length; Based on the composition deviation, parameter deviation, predicted data of the flue gas composition change trend, and dynamic response characteristic parameters of the treatment unit, establish a rolling optimization objective function, where the objective function includes a quadratic term of the increment of the correction amount of the operating parameters, a quadratic term of the pollutant removal efficiency deviation, and a quadratic term of the energy consumption deviation, and add constraints on the change rate of the operating parameters of each stage of the flue gas treatment unit and upper and lower limits of the correction amount; Solve the rolling optimization objective function to obtain the optimal sequence of the increment of the correction amount of the operating parameters from the current moment to the next k - 1 moments; Extract the increment of the correction amount of the operating parameters corresponding to the current moment in the optimal sequence, superimpose it on the set value of the operating parameters of the downstream flue gas treatment unit, and update the set value of the operating parameters.

[0015] Further, the present application also proposes that the step of extracting the increment of the correction amount of the operating parameters corresponding to the current moment in the optimal sequence, superimposing it on the set value of the operating parameters of the downstream flue gas treatment unit, and updating the set value of the operating parameters includes: Determine whether the current moment is the last moment of the rolling optimization window; If the current moment is not the last moment of the rolling optimization window, extract the increment of the operating parameter correction amount corresponding to the current moment from the optimal sequence, and multiply the increment of the operating parameter correction amount by a preset relaxation factor to obtain an adjusted increment of the operating parameter correction amount; Superimpose the adjusted increment of the operating parameter correction amount on the set value of the operating parameter of the downstream flue gas treatment unit to obtain an updated set value of the operating parameter for flue gas treatment at the next control moment.

[0016] Furthermore, the present application also proposes a flue gas treatment device for a glass furnace, which is used for flue gas treatment under the condition of glass formula switching. The device includes: A first acquisition module for acquiring glass formula information scheduled by the production plan; A first prediction module for predicting the change trend of flue gas composition after formula switching based on the correlation between the glass formula components and the pollutant concentration in the flue gas according to the glass formula information; A first calculation module for calculating and setting the target operating parameters of each treatment unit according to the predicted change trend of flue gas composition after formula switching and the process characteristics and operating parameter thresholds of each stage of flue gas treatment unit; A second acquisition module for acquiring the online monitoring data of flue gas composition and the real-time operating parameters of each stage of flue gas treatment unit; A second calculation module for calculating the composition deviation between the actual flue gas composition and the predicted flue gas composition according to the online monitoring data of flue gas composition and the predicted change trend of flue gas composition after formula switching; A third calculation module for calculating the parameter deviation according to the target operating parameters and the real-time operating parameters; A first correction module for correcting the set value of the operating parameter of the downstream flue gas treatment unit according to the composition deviation and the parameter deviation.

[0017] Furthermore, the present application also proposes an electronic device, including a processor and a memory, where the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the above method are run.

[0018] Furthermore, the present application also proposes a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above method are run.

[0019] As can be seen from the above, a method, device, electronic device and storage medium for treating glass furnace flue gas provided by the present application can obtain glass formula information, predict the change trend of flue gas components, calculate target operating parameters, and dynamically correct the operating parameters according to actual monitoring data, achieving a rapid response to the change of flue gas components and the collaborative optimization of multi-level treatment units, and having the advantages of being able to quickly respond to the change of flue gas components, realizing the collaborative optimization of multi-level treatment units, and improving the overall treatment efficiency. Description of the Drawings

[0020] Figure 1 It is a schematic flow chart of a method for treating glass furnace flue gas provided by the present application.

[0021] Figure 2 It is a schematic structural diagram of a device for treating glass furnace flue gas provided by the present application.

[0022] In the figure: 210, the first acquisition module; 220, the first prediction module; 230, the first calculation module; 240, the second acquisition module; 250, the second calculation module; 260, the third calculation module; 270, the first correction module. Detailed Embodiments

[0023] Next, the technical solutions in the present application will be clearly and completely described in conjunction with the drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the present application usually described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided in the drawings below is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0024] It should be noted that: similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0025] In the glass manufacturing industry, recipe switching poses challenges to the flue gas treatment system. Existing multi-stage series flue gas treatment systems usually adopt the method of independent control of each unit, making it difficult to effectively cope with the rapid fluctuations of flue gas components. Specifically, when the upstream glass furnace switches recipes, the flue gas components and concentrations will change drastically. This sudden disturbance is transmitted to the inlet end of the downstream flue gas treatment system, resulting in frequent changes in the inlet flue gas characteristics of each treatment unit. Due to the independence of each unit's control, the downstream flue gas treatment units are difficult to quickly respond to the drastic fluctuations of the upstream flue gas components, and the adjustment of operating parameters often lags behind the flue gas changes, thus affecting the overall treatment efficiency and emission stability of the system.

[0026] In response, referring to Figure 1 , this application proposes a method for treating glass furnace flue gas for flue gas treatment under the condition of glass recipe switching. The steps of this method include: S110. Obtain the glass recipe information scheduled in the production plan; S120. Based on the glass recipe information, predict the change trend of flue gas components after recipe switching according to the correlation between the glass recipe components and the pollutant concentrations in the flue gas; S130. Calculate and set the target operating parameters of each treatment unit according to the predicted change trend of flue gas components after recipe switching and the process characteristics and operating parameter thresholds of each stage of flue gas treatment unit; S140. Obtain the online monitoring data of flue gas components and the real-time operating parameters of each stage of flue gas treatment unit; S150. Calculate the component deviation between the actual flue gas components and the predicted flue gas components according to the online monitoring data of flue gas components and the predicted change trend of flue gas components after recipe switching; S160. Calculate the parameter deviation according to the target operating parameters and the real-time operating parameters; S170. Correct the set value of the operating parameters of the downstream flue gas treatment unit according to the component deviation and the parameter deviation.

[0027] Among them, the glass recipe information refers to the glass raw material ratio data scheduled in the production plan, which can specifically be represented by the types of raw materials and their mass percentages. These information directly affect the flue gas components and are the key inputs for predicting the change trend of flue gas.

[0028] Among them, the change trend of flue gas components refers to the expected change situation of the pollutant concentrations in the flue gas after recipe switching, which can specifically be represented by time series data. By predicting this trend, adjustments can be made to the flue gas treatment system in advance to achieve feedforward control.

[0029] Among them, the process characteristics refer to the technical parameters and performance indicators of each stage of the flue gas treatment unit, which can be specifically described by the removal efficiency curve, energy consumption characteristics, etc. Combining these characteristics for parameter setting can ensure that each unit operates under the optimal conditions.

[0030] Among them, the operating parameter threshold refers to the parameter range for the safe and stable operation of each treatment unit, which can be specifically represented by the upper and lower limits. Considering these thresholds when calculating the target operating parameters can avoid overloading the equipment.

[0031] Among them, the composition deviation refers to the difference between the actual flue gas composition and the predicted value, which can be specifically represented by the concentration difference or percentage. By calculating this deviation, the accuracy of the prediction model can be evaluated and a basis for parameter correction can be provided.

[0032] Among them, the parameter deviation refers to the difference between the actual operating parameter and the target value, which can be specifically represented by the absolute difference or relative error. This deviation reflects the accuracy of the control system and is an important basis for parameter correction.

[0033] The core innovation of this application lies in proposing to establish the correlation between the glass formula composition and the pollutant concentration in the flue gas, thereby achieving accurate prediction of the flue gas changes. Based on the prediction results and combined with the process characteristics of each treatment unit, the target operating parameters are set in advance to achieve a rapid response to the flue gas fluctuations. At the same time, a real-time monitoring and deviation calculation mechanism is introduced to dynamically correct the operating parameters, ensuring the accuracy of the control. This method effectively solves the problems of lag in the response of the flue gas treatment system and poor coordination among units under the condition of frequent switching of glass formulas, and improves the overall treatment efficiency and emission stability of the system.

[0034] The working principle of this application can be described in detail as follows: First, obtain the glass formula information from the production management system, including the types and mass percentages of each raw material. Taking these data as inputs, based on the correlation between the glass formula composition and the pollutant concentration in the flue gas, predict the changing trend of the flue gas composition after formula switching, considering the interaction between raw materials, and the output is the predicted values of the pollutant concentrations in the flue gas for a period of time in the future, forming the flue gas composition change trend data.

[0035] Next, combined with the process characteristics and operating parameter thresholds of each stage of the flue gas treatment unit, the target operating parameters are calculated. Specifically, first, according to the predicted flue gas composition, using the removal efficiency curves of each unit, the minimum treatment efficiency required to meet the emission standards is calculated. Then, considering the energy consumption characteristics of each unit, under the premise of meeting the minimum treatment efficiency, the energy consumption is optimized. At the same time, ensure that the calculated parameters do not exceed the safe operating thresholds of the equipment. The output of this step is the target operating parameters of each treatment unit, such as the reductant injection amount of the denitration unit, the electric field intensity of the dust removal unit, etc.

[0036] During the actual operation process, the system continuously obtains the online monitoring data and real-time operating parameters of each stage of the flue gas treatment unit. By comparing the actual flue gas composition with the predicted value, the composition deviation is calculated. At the same time, compare the actual operating parameters with the target values and calculate the parameter deviation. These two types of deviation data reflect the accuracy of the prediction model and the precision of the control system.

[0037] Finally, based on the calculated composition deviation and parameter deviation, the system uses a preset control algorithm (such as PID control or fuzzy control) to correct the set values of the operating parameters of the downstream flue gas treatment unit. The purpose of the correction is to eliminate the influence of prediction errors and external disturbances, make the actual flue gas composition closer to the expected value, and at the same time keep the operating parameters of each unit within the optimal range.

[0038] In this way, the present application realizes the feedforward-feedback coordinated control of the flue gas treatment system. The feedforward control is based on recipe prediction and provides a fast response ability; the feedback control ensures long-term control precision through real-time correction. The combination of the two can not only cope with the rapid fluctuations of the flue gas but also ensure the stable operation of the system.

[0039] As a preferred implementation manner, the present application can be implemented on a glass production line, and the specific implementation steps are as follows: Obtain glass recipe information: Receive the production plan for the next 6 hours from the production management system, including the switching between two different recipes. Recipe A: silica sand 71.5%, soda ash 22.5%, dolomite 5%, feldspar 1%; Recipe B: silica sand 70%, soda ash 23%, dolomite 5.5%, feldspar 0.5%, sodium sulfate 1%.

[0040] Predict the changing trend of flue gas composition: Based on the correlation between the glass recipe composition and the pollutant concentration in the flue gas, predict the concentration changes of NOx, SO2, and particulate matter in the flue gas within the next 6 hours. The model output shows that within 2 hours after the recipe switch, the NOx concentration is expected to rise from 450 mg / m³ to 520 mg / m³, the SO2 concentration drops from 300 mg / m³ to 250 mg / m³, and the particulate matter concentration remains basically stable at about 30 mg / m³.

[0041] Calculate target operating parameters: Based on the predicted trend of flue gas composition changes and combined with the process characteristics of each treatment unit, the following target operating parameters are calculated: SCR denitration system: The injection rate of the reducing agent (ammonia water) gradually increases from 35 kg / h to 42 kg / h Electrostatic precipitator: Maintain the electric field strength at 4.5 kV / cm Wet desulfurization system: Adjust the slurry pH value from 5.8 to 5.6 Obtain real-time monitoring data: Obtain the online monitoring data of flue gas composition and actual operating parameters of each stage of the treatment unit every 5 minutes.

[0042] Calculate the deviation: Compare the actual flue gas composition with the predicted value and calculate the composition deviation. For example, 2 hours after the recipe switch, the actual NOx concentration is 540 mg / m³, with a deviation of 20 mg / m³ from the predicted value of 520 mg / m³. At the same time, calculate the difference between the actual operating parameter and the target value. For example, the actual ammonia water injection rate of the SCR system is 40 kg / h, with a deviation of 2 kg / h from the target value of 42 kg / h.

[0043] Correct the operating parameters: Based on the calculated deviation, use the PID control algorithm to correct the operating parameters of the downstream flue gas treatment unit. For example, further adjust the ammonia water injection rate of the SCR system to 43 kg / h to cope with the situation where the actual NOx concentration is higher than expected.

[0044] In this way, the flue gas treatment system can anticipate the flue gas fluctuations caused by the recipe switch in advance and make fine adjustments during actual operation to ensure stable and compliant emissions.

[0045] In some of the above embodiments of the present application, based on the glass recipe information, the change trend of flue gas composition after the recipe switch is predicted according to the correlation between the glass recipe components and the pollutant concentration in the flue gas for predicting the change of flue gas composition. However, in this process, there are still challenges in how to accurately establish the correlation between the glass recipe components and the flue gas pollutant concentration, and how to efficiently perform the prediction calculation. A simple linear relationship may not accurately capture the complex recipe-flue gas composition interaction, while an overly complex model may lead to low calculation efficiency and is difficult to meet the requirements of real-time control.

[0046] In response to this, the present application further proposes steps for predicting the change trend of flue gas components after formula switching based on glass formula information and the correlation between glass formula components and pollutant concentrations in flue gas, including: establishing a polynomial regression model between the glass formula component vector and the flue gas component concentration vector, and the model function form is as follows: cj = f_j(R) = ∑_{i=1}^{n} a_{ji} * ri + ∑_{i=1}^{n} ∑_{k=1}^{n} b_{jik} * ri * rk + d_j, where cj is the predicted concentration of the jth flue gas component, R = [r1, r2, ..., rn] is the glass formula component vector, ri represents the proportion of the ith formula component, a_{ji} is the linear regression coefficient, b_{jik} is the quadratic regression coefficient, representing the influence of the interaction between the ith and kth formula components on the concentration of the jth flue gas component, and d_j is the bias term; converting the glass formula information scheduled in the production plan into the glass formula component vector R; substituting the glass formula component vector R into the polynomial regression model to calculate the flue gas component concentration vector, which is used as the prediction result of the change trend of flue gas components after formula switching.

[0047] The polynomial regression model proposed in the present application can effectively capture the non-linear relationship between glass formula components and flue gas pollutant concentrations. This model not only includes the linear term ∑_{i=1}^{n} a_{ji} * ri, but also introduces the quadratic term ∑_{i=1}^{n} ∑_{k=1}^{n} b_{jik} * ri * rk to characterize the interaction between different formula components. This structure enables the model to more accurately describe the complex formula-flue gas component relationship.

[0048] Specifically, the linear term a_{ji} * ri reflects the direct influence of a single formula component on flue gas components. For example, certain raw materials may directly cause an increase in specific pollutants. The quadratic term b_{jik} * ri * rk takes into account the combined effect of two formula components, which is very common in actual production. The combination of certain raw materials may produce synergistic or inhibitory effects, affecting the final flue gas components.

[0049] The bias term d_j plays a compensatory role in the model, which can adjust the baseline of the predicted value and improve the overall fitting accuracy of the model. In this way, the model can adapt to the change of flue gas components under different working conditions and improve the accuracy and robustness of prediction.

[0050] In practical applications, it is first necessary to collect historical production data, including different glass formulas and their corresponding flue gas composition monitoring data. Using these data, a polynomial regression model can be fitted through the least squares method or other optimization algorithms to determine the coefficients \(a_{ji}\), \(b_{jik}\), and \(d_j\). After the model training is completed, it can be used to predict the flue gas composition corresponding to a new glass formula.

[0051] When the production plan changes, the method proposed in this application can respond quickly. First, the new glass formula information is converted into a standardized formula composition vector \(R\). This step ensures the comparability between different formulas and provides a unified input format for the model. Subsequently, substituting the formula composition vector \(R\) into the trained polynomial regression model, the predicted flue gas composition concentration vector can be obtained.

[0052] The advantage of this prediction method lies in its high computational efficiency and the ability to give prediction results in real time. Compared with complex physical and chemical models, the computational complexity of the polynomial regression model is lower, making it suitable for online real-time applications. At the same time, since the model takes into account the interaction between formula components, the prediction results are more accurate than those of simple linear models, providing more reliable data support for subsequent flue gas treatment.

[0053] Furthermore, the method of this application also has good adaptability and scalability. When the production conditions change or new raw materials are introduced, only the historical data set needs to be updated and the model retrained to adapt to the new production situation. This flexibility enables the method to be continuously optimized and the prediction accuracy to be continuously improved.

[0054] As a preferred implementation, a rolling update strategy can be adopted to maintain the accuracy of the model. For example, the model is retrained every certain period (such as weekly or monthly) using the latest production data to capture the long-term change trends that may occur during the production process. This approach can ensure that the model always reflects the current production status and improve the reliability of the prediction.

[0055] By implementing the above method, this application can quickly and accurately predict the change trend of flue gas composition under the condition of frequent switching of glass formulas. This provides a reliable data basis for subsequent flue gas treatment, helping to improve the response speed and treatment efficiency of the entire flue gas treatment system. At the same time, due to the use of a polynomial regression model with relatively high computational efficiency, this method can meet the requirements of real-time control and provide strong support for the intelligent and refined control of the flue gas treatment system.

[0056] In some of the above embodiments of the present application, a step of calculating and setting the target operating parameters of each stage of the flue gas treatment unit is proposed according to the changing trend of the flue gas composition after the predicted recipe switch, as well as the process characteristics and operating parameter thresholds of each stage of the flue gas treatment unit, so as to adapt to the change in the flue gas composition caused by the glass recipe switch. However, in this process, how to accurately calculate the target operating parameters that can achieve efficient pollutant removal and low-energy consumption operation according to the predicted changing trend of the flue gas composition and the characteristics of each treatment unit, while also considering the limitations of the operating parameter thresholds of each unit, is a complex optimization problem.

[0057] In response to this, the present application further proposes that the steps of calculating and setting the target operating parameters of each stage of the flue gas treatment unit according to the changing trend of the flue gas composition after the predicted recipe switch, as well as the process characteristics and operating parameter thresholds of each stage of the flue gas treatment unit, include: obtaining the pollutant removal efficiency curves of each stage of the flue gas treatment unit under different flue gas compositions, and the relationship between the operating parameters of each stage of the flue gas treatment unit and the pollutant removal efficiency; based on the predicted changing trend of the flue gas composition after the recipe switch, combining with the pollutant removal efficiency curves of each stage of the flue gas treatment unit, calculating the expected pollutant removal amounts of each stage of the flue gas treatment unit under different operating parameters; determining the upper and lower limit constraints of the operating parameters of each stage of the flue gas treatment unit according to the operating parameter thresholds of each stage of the flue gas treatment unit; and calculating the target operating parameters of each stage of the flue gas treatment unit with the goal of minimizing the total energy consumption, while satisfying that the discharge concentrations of each pollutant are lower than the emission standards and the upper and lower limit constraints of the operating parameters of each unit.

[0058] This technical solution first obtains the pollutant removal efficiency curves of each stage of the flue gas treatment unit under different flue gas compositions, and the relationship between the operating parameters of each stage of the flue gas treatment unit and the pollutant removal efficiency. These relationships can be established through experimental data or historical operating data. For example, for the desulfurization unit, a relationship curve can be established between parameters such as the SO2 inlet concentration, liquid-gas ratio, and absorbent concentration and the SO2 removal efficiency. For the denitration unit, a relationship curve can be established between parameters such as the NOx inlet concentration, reductant injection amount, and reaction temperature and the NOx removal efficiency.

[0059] Based on the predicted changing trend of the flue gas composition after the recipe switch, combining with the pollutant removal efficiency curves of each stage of the flue gas treatment unit, calculate the expected pollutant removal amounts of each stage of the flue gas treatment unit under different operating parameters. Specifically, the predicted changing trend of the flue gas composition can be substituted into the removal efficiency curves of each unit to obtain the expected pollutant removal amounts under different operating parameters. This step takes into account the influence of the flue gas composition change on the performance of each treatment unit and provides a basis for subsequent optimization.

[0060] According to the operating parameter thresholds of each level of flue gas treatment unit, determine the upper and lower limit constraints of the operating parameters of each level of flue gas treatment unit. These constraints may include restrictions on the safe operation of the equipment, such as the upper and lower limits of the pH value of the desulfurization slurry, the maximum operating temperature of the denitrification catalyst, etc. By setting these constraints, ensure that the optimization results are within the practical operable range.

[0061] With the goal of minimizing total energy consumption, while satisfying the emission concentration of each pollutant below the emission standard and the upper and lower limits of the operating parameters of each unit, the target operating parameters of each level of flue gas treatment units are calculated. This step can use multi-objective optimization algorithms, such as sequential quadratic programming or genetic algorithms, to build an optimization model. The objective function can be expressed as the weighted sum of the energy consumption of each unit, and the constraints include the emission concentration limit and the operating parameter range. By solving this optimization problem, the operating parameter combination with the lowest energy consumption is obtained while meeting the emission standards and equipment limitations.

[0062] This technical solution achieves the coordinated optimization of multi-stage flue gas treatment units by comprehensively considering the flue gas composition change trend, the performance characteristics of each treatment unit, and the operation constraints. When the flue gas composition changes due to the glass formula switching, the target operating parameters with strong adaptability can be quickly calculated, which improves the system's responsiveness to flue gas fluctuations. At the same time, by minimizing the optimization goal of total energy consumption, the operating cost is reduced on the basis of ensuring that the emission meets the standard.

[0063] Furthermore, this method can be combined with online monitoring and feedback control to form a closed-loop control system. Specifically, the online monitoring data of flue gas composition and the actual operating parameters of each level of flue gas treatment unit can be obtained in real time, and the operating parameter setting values ​​of the downstream processing unit can be dynamically adjusted according to the deviation between the actual flue gas composition and the predicted flue gas composition, and the deviation between the actual operating parameters and the target operating parameters. This dynamic optimization mechanism can further improve the adaptability and stability of the system.

[0064] As a preferred implementation, a rolling optimization strategy can be used to achieve dynamic adjustment. For example, define a rolling optimization window, such as the next 4 hours, and perform an optimization calculation in each control cycle (such as 15 minutes). The optimization model can consider the predicted flue gas change trend within the window and the dynamic response characteristics of each processing unit to calculate a series of optimal operating parameter trajectories. Then, only the optimization results at the current moment are executed, and the optimization calculation is performed again in the next control cycle. This method can achieve predictive control of future changes while ensuring system stability.

[0065] In specific implementation, appropriate optimization algorithms and model structures can be selected according to actual working conditions. For example, for a system that includes three-stage treatment of denitrification, dust removal and desulfurization, the following optimization model can be established: Objective function: min J = w1 * E_DeNOx + w2 * E_Dust + w3 * E_DeSOx Where, E_DeNOx, E_Dust, and E_DeSOx represent the energy consumption of the denitration, dust removal, and desulfurization units respectively, and w1, w2, and w3 are the weight coefficients of the energy consumption of each unit.

[0066] Constraints: NOx_out ≤ NOx_limit; Dust_out ≤ Dust_limit; SO2_out ≤ SO2_limit; T_DeNOx_min ≤ T_DeNOx ≤ T_DeNOx_max; v_Dust_min ≤ v_Dust ≤ v_Dust_max; pH_DeSOx_min ≤ pH_DeSOx ≤ pH_DeSOx_max; Where, NOx_out, Dust_out, and SO2_out represent the outlet concentrations of each pollutant respectively, and the corresponding limit is the emission limit. T_DeNOx, v_Dust, and pH_DeSOx represent the reaction temperature of the denitration unit, the filtration speed of the dust removal unit, and the slurry pH value of the desulfurization unit respectively, and min and max represent their allowed minimum and maximum values.

[0067] By solving this optimization problem, the operating parameter combination with the lowest total energy consumption can be obtained on the premise of meeting the emission standards and equipment limitations. These parameters can be directly used to set the control objectives of each treatment unit or used as reference values for operators to adjust.

[0068] This technical solution realizes the coordinated control of multi-stage flue gas treatment units by establishing a multi-objective optimization model. Compared with the traditional independent control method of each unit, this solution can better adapt to the rapid changes in flue gas composition, improve the overall treatment efficiency and energy utilization rate of the system. At the same time, by considering the operating parameter constraints of each unit, the feasibility of the optimization results is ensured, and equipment damage or operating failures caused by improper parameter settings are avoided. In addition, the flexibility of this method enables it to adjust the optimization objectives and constraint conditions according to actual needs, and is applicable to flue gas treatment systems of different types of glass furnaces.

[0069] In some of the above embodiments of the present application, steps are proposed to calculate the target operating parameters of the flue gas treatment units at each level with the goal of minimizing total energy consumption while satisfying the emission concentration of each pollutant being lower than the emission standard, as well as the upper and lower limit constraints of the operating parameters of each unit, so as to optimize the operating efficiency of the flue gas treatment system. However, in this process, how to construct a suitable optimization model to balance the minimization of energy consumption and the satisfaction of pollutant emission standards while considering the operating parameter constraints of each treatment unit is a complex multi-objective optimization problem. In addition, how to design an efficient solution algorithm to iteratively calculate this optimization model to quickly obtain the optimal operating parameters of the flue gas treatment units at each level is also a technical problem that needs to be solved urgently.

[0070] In this regard, the present application further proposes to construct a multi-objective optimization model with the objective function of minimizing the total energy consumption. The objective function is expressed as: min ∑_{i=1}^{m} P_i(u_i), where P_i(u_i) is the energy consumption function of the i-th flue gas treatment unit, u_i is the operating parameter of the i-th flue gas treatment unit, and m is the number of flue gas treatment units; set pollutant emission concentration constraints, expressed as: g_j(u_1, u_2, ..., u_m) ≤ C_j, where g_j(u_1, u_2, ..., u_m) is the emission concentration function of the j-th pollutant, and C_j is the emission standard of the j-th pollutant; determine the upper and lower limit constraints of the operating parameters of each flue gas treatment unit, expressed as: u_{i,min} ≤ u_i ≤ u_{i,max}, where u_{i,min} and u_{i,max} are the lower and upper limits of the operating parameters of the i-th flue gas treatment unit respectively; the multi-objective optimization model is solved iteratively, and in each iteration, the gradients of the objective function and the constraints are calculated, and the operating parameters of each flue gas treatment unit are updated until the convergence conditions are met, thereby obtaining the target operating parameters of each level of flue gas treatment units.

[0071] One of the core innovations of this application is to construct a multi-objective optimization model with the objective function of minimizing the total energy consumption. The objective function is expressed as: min ∑_{i=1}^{m} P_i(u_i), where P_i(u_i) is the energy consumption function of the i-th flue gas treatment unit, u_i is the operating parameter of the i-th flue gas treatment unit, and m is the number of flue gas treatment units. This expression allows the energy consumption of the entire flue gas treatment system to be globally optimized, rather than just locally optimizing a single treatment unit.

[0072] The energy consumption function \(P_i(u_i)\) can take various forms to describe the energy consumption characteristics of different types of flue gas treatment units. For example, for the dust removal unit, the energy consumption function may be related to the fan power and the filtration pressure drop; for the desulfurization unit, the energy consumption may be related to the absorbent dosage and the circulation pump power. By establishing an accurate energy consumption function, the energy consumption under different operating parameters can be evaluated more precisely.

[0073] Setting the pollutant emission concentration constraint is the key to ensuring that the optimization results meet the environmental protection requirements. The constraint is expressed as: \(g_j(u_1, u_2, \ldots, u_m) \leq C_j\), where \(g_j(u_1, u_2, \ldots, u_m)\) is the emission concentration function of the \(j\)-th pollutant, and \(C_j\) is the emission standard of the \(j\)-th pollutant. This constraint condition takes into account the comprehensive treatment effect of multiple treatment units on a single pollutant, reflecting the integrity of the system.

[0074] The pollutant emission concentration function \(g_j\) can be constructed based on the removal efficiency and series relationship of each treatment unit. For example, for particulate matter, it may involve the synergistic effect of a high-temperature electrostatic precipitator and a wet desulfurization unit; for sulfur dioxide, the absorption efficiency of the desulfurization tower and the synergistic removal effect of the dust collector may need to be considered. By accurately modeling these relationships, the final emission concentration under different combinations of operating parameters can be predicted better.

[0075] Determining the upper and lower limit constraints of the operating parameters of each flue gas treatment unit is an important means to ensure the feasibility of the optimization results. The constraint is expressed as: \(u_{i,min} \leq u_i \leq u_{i,max}\), where \(u_{i,min}\) and \(u_{i,max}\) are the lower and upper limits of the operating parameters of the \(i\)-th flue gas treatment unit, respectively. These limitations may stem from equipment design specifications, safety operation requirements, or process stability considerations.

[0076] For example, for a selective catalytic reduction (SCR) denitration system, the catalyst bed temperature usually needs to be controlled within the range of 300 - 400 °C to ensure the denitration efficiency; for a wet desulfurization system, the slurry pH value may need to be controlled between 5.0 - 6.0 to balance the desulfurization efficiency and equipment corrosion. By setting these constraints, it can be ensured that the optimization results are within the practical and operable range.

[0077] Iteratively solving the multi-objective optimization model is the core step to obtain the optimal operating parameters. In each iteration, the gradients of the objective function and the constraint conditions are calculated, and the operating parameters of each flue gas treatment unit are updated until the convergence condition is met. This process can adopt various numerical optimization algorithms, such as the gradient descent method, Newton's method, or the interior point method, etc.

[0078] Specifically, the sequential quadratic programming (SQP) algorithm can be used to solve this nonlinear optimization problem. In each iteration, the SQP algorithm approximates the nonlinear problem as a quadratic programming problem, and then solves this quadratic programming problem to obtain the search direction and step size. This method can effectively handle nonlinear constraints and shows good convergence in many practical problems.

[0079] In practical applications, multiple convergence conditions can be set, such as the relative change of the objective function value is less than a certain threshold (for example, 0.1%), or the maximum change of the operating parameter is less than a specified value (for example, 0.5%). When these conditions are met, the algorithm is considered to have converged to the optimal solution.

[0080] This method can find the operating parameter combination that can minimize the total energy consumption while considering multiple constraints. Compared with the traditional unit independent control method, this global optimization method can better balance the overall system performance and the operating efficiency of each unit.

[0081] The technical solution of this application achieves overall optimization of the flue gas treatment system by constructing a multi-objective optimization model, taking into account multiple aspects such as minimizing energy consumption, pollutant emission compliance, and equipment operation constraints. This method can quickly calculate the optimal operating parameters of each level of flue gas treatment units under conditions where the glass formula is frequently switched, thereby improving the system's ability to adapt to fluctuations in flue gas composition.

[0082] Specifically, when the glass formula changes, the predicted results of the flue gas composition are input into the optimization model. The model recalculates the optimal operating parameters of each treatment unit based on the new flue gas composition. This process takes into account the mutual influence between different treatment units, such as changes in upstream denitrification efficiency affecting the load of downstream desulfurization units. Through global optimization, a set of operating parameters can be found that can achieve the lowest energy consumption and stable emission standards under new flue gas conditions.

[0083] In addition, the method of the present application has good scalability and adaptability. By adjusting the objective function and constraints, it can easily adapt to different types of flue gas treatment systems and different optimization objectives. For example, the consideration of equipment life or operating cost can be added to the objective function, or the deep treatment requirements of specific pollutants can be added to the constraints.

[0084] As a preferred implementation, the following specific embodiments may be considered: Assume that a glass furnace flue gas treatment system consists of three main units: a selective catalytic reduction (SCR) denitrification unit, an electrostatic precipitator (ESP), and a wet desulfurization tower. The main pollutants that need to be controlled by the system include nitrogen oxides (NOx), particulate matter (PM), and sulfur dioxide (SO2).

[0085] First, construct the energy consumption function: P_1(u_1) = a_1 * u_1^2 + b_1 * u_1 + c_1 / / SCR unit, where u_1 is the reductant injection amount; P_2(u_2) = a_2 * u_2^2 + b_2 * u_2 + c_2 / / ESP unit, where u_2 is the secondary voltage; P_3(u_3) = a_3 * u_3^2 + b_3 * u_3 + c_3 / / Desulfurization tower, where u_3 is the slurry circulation rate; Among them, a_i, b_i, c_i (i = 1, 2, 3) are fitting coefficients, which can be obtained by regression of historical operation data.

[0086] Then, set the pollutant emission concentration constraints: g_NOx(u_1, u_2, u_3) = C_in_NOx * (1 - η_1(u_1)) ≤ 50 mg / Nm³; g_PM(u_1, u_2, u_3) = C_in_PM * (1 - η_2(u_2)) * (1 - η_3_PM(u_3)) ≤ 20 mg / Nm³; g_SO2(u_1, u_2, u_3) = C_in_SO2 * (1 - η_3_SO2(u_3)) ≤ 100 mg / Nm³; Among them, C_in_X is the inlet concentration of pollutant X, and η_i is the removal efficiency function of each unit.

[0087] Next, determine the operating parameter constraints: 0.5 ≤ u_1 ≤ 2.0 / / SCR reductant injection amount, unit: NH3 / NOx molar ratio; 30 ≤ u_2 ≤ 80 / / ESP secondary voltage, unit: kV; 500 ≤ u_3 ≤ 2000 / / Desulfurization slurry circulation rate, unit: m³ / h; Finally, use the sequential quadratic programming algorithm to solve the optimization problem. In each iteration: Linearize the objective function and constraint conditions to construct a quadratic programming sub-problem.

[0088] Solve the sub-problem to obtain the search direction and step size.

[0089] Perform a one-dimensional search to determine the optimal step size.

[0090] Update the operating parameter values.

[0091] Check the convergence condition. If it is satisfied, stop. Otherwise, return to step 1.

[0092] Through this method, it is possible to find the combination of operating parameters that can minimize the total energy consumption while considering multiple constraints. For example, when the change in the glass formula leads to an increase in the NOx inlet concentration, the optimization algorithm may increase the SCR reductant injection amount, while slightly reducing the ESP voltage and the desulfurization slurry circulation amount to balance the increase in energy consumption brought about by the improvement in NOx removal efficiency. This global optimization method can better balance the overall system performance and the operating efficiency of each unit, and can achieve lower total energy consumption and more stable emission levels compared to the traditional unit-independent control method.

[0093] In some of the above embodiments of the present application, it is proposed to correct the set value of the operating parameters of the downstream treatment unit according to the composition deviation and parameter deviation to adapt to the change in the flue gas composition. However, in this process, the priority consideration of different control objectives is lacking, and it is difficult to weigh and optimize among multiple objectives. This may result in the corrected operating parameters not being able to meet the requirements in multiple aspects such as pollutant removal efficiency, energy consumption, and equipment life, affecting the performance and reliability of the overall system.

[0094] In response to this, the present application further proposes that the steps of correcting the set value of the operating parameters of the downstream treatment unit according to the composition deviation and parameter deviation include: obtaining the type of control objective and the corresponding control objective priority, where the type of control objective at least includes one of pollutant removal efficiency, energy consumption, and equipment life; determining the first weight of the pollutant removal efficiency deviation, the second weight of the energy consumption deviation, and the third weight of the equipment life deviation according to the control objective priority; calculating the pollutant removal efficiency deviation according to the deviation between the actual flue gas composition and the predicted flue gas composition, and calculating the energy consumption deviation and the equipment life deviation according to the deviation between the actual operating parameters and the target operating parameters; calculating the operating parameter correction amount of each stage of the downstream flue gas treatment unit based on the first weight, the second weight, the third weight, the pollutant removal efficiency deviation, the energy consumption deviation, and the equipment life deviation; and adding the operating parameter correction amount to the set value of the operating parameters of each stage of the flue gas treatment unit to obtain the corrected set value of the operating parameters.

[0095] In this technical solution, first, the type of control objective and its priority are obtained. The type of control objective can include multiple aspects such as pollutant removal efficiency, energy consumption, and equipment life. The setting of the priority can be flexibly adjusted according to the actual production requirements and environmental protection requirements. For example, during a period when environmental protection requirements are particularly strict, the pollutant removal efficiency can be set as the highest priority; during a period when energy costs are high, the priority of energy consumption can be increased; and during the equipment maintenance cycle, the priority of equipment life can be appropriately increased.

[0096] According to the set control target priorities, determine the weights of each deviation. Here, three weights are adopted: the first weight corresponds to the deviation of pollutant removal efficiency, the second weight corresponds to the deviation of energy consumption, and the third weight corresponds to the deviation of equipment life. The distribution of weights can adopt various methods, such as linear mapping or non-linear function mapping. For example, the target with the highest priority can be given a weight of 0.5, the second highest a weight of 0.3, and the lowest a weight of 0.2.

[0097] Next, calculate the deviation of pollutant removal efficiency based on the deviation between the actual flue gas composition and the predicted flue gas composition. This step can be achieved by comparing the actually monitored pollutant concentration with the concentration given by the prediction model. At the same time, calculate the deviation of energy consumption and the deviation of equipment life based on the deviation between the actual operating parameters and the target operating parameters. The deviation of energy consumption can be obtained by comparing the actual energy consumption with the expected energy consumption, while the deviation of equipment life can be estimated by monitoring the wear degree or operating time of key components.

[0098] Based on the weights and deviations calculated above, calculate the correction amounts of the operating parameters of the flue gas treatment units at each downstream stage. This step can adopt the method of weighted summation, that is, multiply each deviation by the corresponding weight and then add them to obtain the comprehensive correction amount. The specific calculation formula can be expressed as: Correction amount = w1 * deviation of pollutant removal efficiency + w2 * deviation of energy consumption + w3 * deviation of equipment life where w1, w2, and w3 are the first, second, and third weights respectively.

[0099] Finally, superimpose the calculated correction amounts of the operating parameters on the set values of the operating parameters of each flue gas treatment unit to obtain the corrected set values of the operating parameters. This step realizes the real-time adjustment of the operating parameters, enabling the system to better adapt to the changes in the flue gas composition.

[0100] By introducing a multi-objective trade-off mechanism, the technical solution of this application can, while ensuring the pollutant removal efficiency, take into account the requirements in multiple aspects such as energy consumption and equipment life. This method not only improves the adaptability and flexibility of the system but also enables more refined control in different production scenarios. For example, when the environmental protection pressure is high, the weight of pollutant removal efficiency can be increased to ensure compliance with emissions standards; during the period of high energy costs, the weight of energy consumption can be appropriately increased to reduce operating costs; when the equipment is about to enter the maintenance cycle, the weight of equipment life can be increased to extend the service life of the equipment.

[0101] Specifically, the technical solution of this application can be realized in the following ways: First, establish a control target priority matrix, which contains the priority settings of each control target in different production scenarios. For example: Scenario 1 (Strict environmental protection period): Pollutant removal efficiency (0.6), energy consumption (0.3), equipment life (0.1); Scenario 2 (Energy shortage period): Pollutant removal efficiency (0.4), energy consumption (0.5), equipment life (0.1); Scenario 3 (Equipment maintenance period): Pollutant removal efficiency (0.4), energy consumption (0.3), equipment life (0.3); During the operation of the system, select the corresponding priority settings according to the current production scenario.

[0102] Secondly, design a deviation calculation module that receives real-time monitoring data and the output of the prediction model, and calculates various deviations: Pollutant removal efficiency deviation = (Actual pollutant concentration - Predicted pollutant concentration) / Predicted pollutant concentration; Energy consumption deviation = (Actual energy consumption - Expected energy consumption) / Expected energy consumption; Equipment life deviation = (Actual operating time - Expected operating time) / Expected operating time; Then, design a parameter correction amount calculation module that calculates the comprehensive correction amount according to the priority weights and various deviations: Correction amount = 0.6 * Pollutant removal efficiency deviation + 0.3 * Energy consumption deviation + 0.1 * Equipment life deviation; Finally, apply the calculated correction amount to the operating parameters of each stage of the flue gas treatment unit: New operating parameter = Original operating parameter + Correction amount * Adjustment coefficient; Among them, the adjustment coefficient is used to control the amplitude of the correction to prevent drastic fluctuations in the parameters.

[0103] In this way, the technical solution of this application can achieve coordinated optimization control of multiple treatment units in the case of rapid changes in flue gas composition. This not only improves the system's rapid response ability to flue gas fluctuations, but also takes into account the requirements of multiple aspects such as energy consumption and equipment life while ensuring pollutant removal efficiency, thereby enhancing the overall performance and reliability of the entire flue gas treatment system.

[0104] In some of the above embodiments of this application, it is proposed to correct the set value of the operating parameters of the downstream treatment unit according to the composition deviation and parameter deviation to adjust the operating parameters of the flue gas treatment system. However, in this process, only considering the deviation at the current moment may lead to insufficiently smooth system response, making it difficult to cope with the future change trend of flue gas composition, thereby affecting the overall performance and stability of the system. In addition, simple parameter correction may not fully consider the dynamic response characteristics of the treatment unit, resulting in poor adjustment effects.

[0105] In response to this, the present application further proposes to define a rolling optimization window length, which represents the number of future time steps that need to be predicted and optimized starting from the current moment; at each control moment, the following steps are executed: Obtain the predicted data of the flue gas composition change trend and the dynamic response characteristic parameters of the processing unit for the next k moments defined by the rolling optimization window length; Based on the composition deviation, parameter deviation, predicted data of the flue gas composition change trend, and the dynamic response characteristic parameters of the processing unit, establish a rolling optimization objective function, where the objective function includes quadratic terms of the increment of the operating parameter correction amount, quadratic terms of the pollutant removal efficiency deviation, and quadratic terms of the energy consumption deviation, and add constraints on the change rate of the operating parameters of each stage of the flue gas treatment unit and upper and lower limits constraints on the correction amount; Solve the rolling optimization objective function to obtain the optimal sequence of the increment of the operating parameter correction amount from the current moment to the next k - 1 moments; Extract the increment of the operating parameter correction amount corresponding to the current moment in the optimal sequence, superimpose it on the set value of the operating parameter of the downstream flue gas treatment unit, and update the set value of the operating parameter.

[0106] This technical solution transforms the nonlinear time-varying system into a rolling optimization problem within a finite time domain by introducing a rolling optimization window and dynamic response characteristic parameters. The definition of the rolling optimization window length k ensures the matching of the optimization interval and the system inertia, enabling the control strategy to better adapt to the change trend of the flue gas composition.

[0107] Specifically, the rolling optimization window length k can be set according to the dynamic characteristics of the flue gas treatment system. For example, for a system with a slow response, a longer window length can be selected, such as 30 minutes or 1 hour; for a system with a fast response, a shorter window length can be selected, such as 10 minutes or 15 minutes. The selection of the window length needs to balance the prediction accuracy and the computational complexity.

[0108] At each control moment, first obtain the predicted data of the flue gas composition change trend for the next k moments. These data can be generated by time series analysis methods, such as deep learning models like autoregressive integrated moving average model (ARIMA) or long short-term memory network (LSTM). At the same time, obtain the dynamic response characteristic parameters of the processing unit, which can include the response time constant, gain coefficient, etc. of each processing unit, and are used to describe the dynamic response characteristics of the processing unit to input changes.

[0109] Based on the acquired data, a rolling optimization objective function is established. This objective function contains multiple quadratic terms, which respectively represent the increment of the operating parameter correction amount, the deviation of the pollutant removal efficiency, and the deviation of the energy consumption. The quadratic term of the increment of the operating parameter correction amount is used to limit the amplitude of parameter adjustment and avoid excessive adjustment causing system instability. The quadratic term of the pollutant removal efficiency deviation is used to ensure that the system achieves the expected pollutant removal effect. The quadratic term of the energy consumption deviation is used to control the energy consumption level of the system.

[0110] In addition, the objective function also includes the operating parameter change rate constraint and the upper and lower limit constraints of the correction amount. The change rate constraint can prevent the parameter adjustment from being too drastic, while the upper and lower limit constraints ensure that the adjusted parameters are still within the acceptable range of the equipment. The introduction of these constraint conditions makes the optimization result more in line with the actual operation requirements.

[0111] The rolling optimization objective function can be solved using quadratic programming or nonlinear programming algorithms, such as the interior point method or the sequential quadratic programming method. The solution result is an optimal sequence containing the increment of the operating parameter correction amount from the current moment to the future k - 1 moments. Extract the correction amount increment corresponding to the current moment from this sequence and superimpose it on the current operating parameter setting value to achieve parameter update.

[0112] The advantage of this rolling optimization method is that it not only considers the deviation at the current moment but also the predicted trend in a future period of time, making the control decision forward-looking. At the same time, by introducing dynamic response characteristic parameters, the control strategy can better adapt to the response characteristics of different processing units, improving the accuracy and effectiveness of adjustment.

[0113] As a preferred implementation method, the rolling optimization objective function can be expressed as: J = ∑_{i = 1}^k (α||Δu_i||^2 + β||η_i - η_d||^2 + γ||P_i - P_d||^2); where Δu_i represents the increment of the operating parameter correction amount at the i-th moment, η_i and η_d represent the actual and expected pollutant removal efficiencies at the i-th moment respectively, and P_i and P_d represent the actual and expected energy consumptions at the i-th moment respectively. α, β, and γ are weight coefficients used to balance the importance of different objectives.

[0114] The operating parameter change rate constraint can be expressed as: |Δu_i - Δu_{i - 1}| ≤ ε; where ε is the maximum allowable change rate.

[0115] The upper and lower limit constraints of the correction amount can be expressed as: Δu_min ≤ Δu_i ≤ Δu_max; Among them, Δu_min and Δu_max are the lower limit and upper limit of the correction amount respectively.

[0116] In practical applications, the objective function and constraint conditions can be appropriately adjusted and optimized according to the specific characteristics of the flue gas treatment system and control requirements. For example, for a system with three main treatment units including desulfurization, denitrification, and dust removal, different weight coefficients and constraint conditions can be set for each unit to reflect its importance and operation limitations in the entire treatment process.

[0117] Through this rolling optimization method, the flue gas treatment system can respond more flexibly and intelligently to the flue gas composition fluctuations caused by the glass formula switch. The system can optimize the control of energy consumption while ensuring the pollutant removal efficiency. Compared with the traditional control method based solely on the current deviation, this method has stronger predictability and adaptability, and can significantly improve the overall performance and stability of the system.

[0118] In some of the above embodiments of the present application, a step of extracting the increment of the operating parameter correction amount corresponding to the current moment in the optimal sequence, superimposing it on the set value of the operating parameter of the downstream flue gas treatment unit, and updating the set value of the operating parameter is proposed to correct the set value of the operating parameter of the downstream treatment unit. However, in this process, directly applying the optimized correction amount may cause the system response to be too intense and affect the system stability. In addition, at different moments of the rolling optimization window, the application strategy of the correction amount also needs to be treated differently to balance the rapid response and stability of the system.

[0119] In response to this, the present application further proposes to determine whether the current moment is the last moment of the rolling optimization window; if the current moment is not the last moment of the rolling optimization window, then extract the increment of the operating parameter correction amount corresponding to the current moment from the optimal sequence, and multiply the increment of the operating parameter correction amount by a preset relaxation factor to obtain an adjusted increment of the operating parameter correction amount; superimpose the adjusted increment of the operating parameter correction amount on the set value of the operating parameter of the downstream flue gas treatment unit to obtain an updated set value of the operating parameter for flue gas treatment at the next control moment.

[0120] In this technical solution, first, it is determined whether the current moment is the last moment of the rolling optimization window. This determination step is of great significance because in rolling optimization control, there are often large errors in the prediction at the end of the optimization window. By distinguishing the position of the current moment in the optimization window, different correction strategies can be adopted, thereby improving the accuracy and stability of the control.

[0121] Specifically, if the current moment is not the last moment of the rolling optimization window, the increment of the operating parameter correction corresponding to the current moment is extracted from the optimal sequence. Here, the optimal sequence is obtained by solving the rolling optimization objective function and contains the increments of the operating parameter corrections from the current moment to the future k-1 moments. Extracting the correction increment at the current moment can ensure the timeliness and accuracy of the control action.

[0122] Furthermore, the extracted increment of the operating parameter correction is multiplied by a preset relaxation factor to obtain an adjusted increment of the operating parameter correction. Introducing the relaxation factor is a crucial step, whose purpose is to reduce the amplitude of the control action and avoid overly drastic system responses. The value of the relaxation factor can be adjusted according to system characteristics and control requirements, and usually a value between 0 and 1 is selected. For example, 0.8 can be chosen as the initial value and then fine-tuned according to the system response.

[0123] Finally, the adjusted increment of the operating parameter correction is superimposed on the set value of the operating parameters of the downstream flue gas treatment unit to obtain an updated set value of the operating parameters. This step realizes the real-time correction of the operating parameters while considering system stability. The updated set value will be used for flue gas treatment at the next control moment to ensure the continuity and smoothness of the control.

[0124] This technical solution effectively solves the problem of system instability that may be caused by directly applying the optimization results by judging the position of the current moment in the rolling optimization window, introducing the relaxation factor, and implementing parameter correction step by step. At the same time, by differentiating different moments in the optimization window, refined control of the correction application strategy is achieved.

[0125] In practical applications, the value of the relaxation factor can be flexibly adjusted according to the characteristics and control requirements of the flue gas treatment system. For example, for a treatment unit with a slow response, a larger relaxation factor (such as 0.9) can be selected to reduce the impact of the correction; while for a unit with a fast response, a smaller relaxation factor (such as 0.7) can be selected to improve the control sensitivity. In addition, the relaxation factor can also be dynamically adjusted according to the moment position in the rolling optimization window. For example, a smaller relaxation factor is used in the early stage of the window, and a larger relaxation factor is used in the later stage of the window to balance the responsiveness and stability of the control.

[0126] A specific embodiment of this technical solution is as follows: Assume that the length of the rolling optimization window is 10 time steps and the current is at the 5th time step. The system first determines that the current moment is not the last moment of the rolling optimization window. Then, the increment of the operating parameter correction corresponding to the current moment (the 5th time step) is extracted from the optimal sequence, assumed to be Δu = 2.5.

[0127] Next, select a preset relaxation factor, for example, α = 0.8. Multiply the increment of the operating parameter correction by the relaxation factor: Δu_adjusted = Δu * α = 2.5 * 0.8 = 2.0.

[0128] Finally, add the increment of the adjusted operating parameter correction to the current set value of the operating parameter. Assume that the current set value of the operating parameter for a certain downstream flue gas treatment unit is u_current = 50, then the updated set value of the operating parameter is: u_new = u_current + Δu_adjusted = 50 + 2.0 = 52.

[0129] This updated set value of the operating parameter 52 will be used for flue gas treatment at the next control moment (the 6th time step). In this way, the system can maintain its fast response ability while avoiding instability caused by excessive correction amounts.

[0130] By adopting this technical solution, the flue gas treatment system can respond more smoothly to changes in flue gas composition, reducing oscillations and fluctuations during the control process. At the same time, due to considering the characteristics of different moments in the rolling optimization window, the system can adopt more appropriate control strategies at different stages of the window, improving the overall control effect. This not only enhances the system's adaptability to upstream flue gas fluctuations but also helps to maintain a stable pollutant removal efficiency, reducing energy consumption and the risk of secondary pollution.

[0131] In a second aspect, the present application also proposes a flue gas treatment device for a glass furnace, which is used for flue gas treatment under the condition of glass formula switching. The device includes: A first acquisition module 210, configured to acquire glass formula information scheduled for production; A first prediction module 220, configured to predict the change trend of flue gas composition after formula switching based on the glass formula information and the correlation between the glass formula components and the pollutant concentration in the flue gas; A first calculation module 230, configured to calculate and set the target operating parameters of each treatment unit according to the predicted change trend of flue gas composition after formula switching and the process characteristics and operating parameter thresholds of each flue gas treatment unit; A second acquisition module 240, configured to acquire the online monitoring data of flue gas composition and the real-time operating parameters of each flue gas treatment unit; A second calculation module 250, configured to calculate the composition deviation between the actual flue gas composition and the predicted flue gas composition according to the online monitoring data of flue gas composition and the predicted change trend of flue gas composition after formula switching; A third calculation module 260, configured to calculate the parameter deviation according to the target operating parameters and the real-time operating parameters; The first correction module 270 is configured to correct the set value of the operating parameters of the downstream flue gas treatment unit according to the composition deviation and the parameter deviation.

[0132] By obtaining the glass recipe information, predicting the change trend of the flue gas composition, calculating the target operating parameters, and dynamically correcting the operating parameters according to the actual monitoring data, a rapid response to the change of the flue gas composition and the collaborative optimization of the multi-stage treatment unit are achieved. It has the advantages of being able to quickly respond to the change of the flue gas composition, realizing the collaborative optimization of the multi-stage treatment unit, and improving the overall treatment efficiency.

[0133] In a third aspect, the present application also provides an electronic device, including a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the above method are run.

[0134] Through the above technical solution, the processor and the memory are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanisms (not marked). The memory stores computer-readable instructions executable by the processor. When the electronic device runs, the processor executes the computer-readable instructions to execute the method in any optional implementation manner of the above embodiments to achieve the following functions: obtaining the glass recipe information of the production plan scheduling; based on the glass recipe information, predicting the change trend of the flue gas composition after the recipe switch according to the correlation between the glass recipe composition and the pollutant concentration in the flue gas; calculating and setting the target operating parameters of each treatment unit according to the predicted change trend of the flue gas composition after the recipe switch and the process characteristics and operating parameter thresholds of each stage of the flue gas treatment unit; obtaining the on-line monitoring data of the flue gas composition and the real-time operating parameters of each stage of the flue gas treatment unit; calculating the composition deviation between the actual flue gas composition and the predicted flue gas composition according to the on-line monitoring data of the flue gas composition and the predicted change trend of the flue gas composition after the recipe switch; calculating the parameter deviation according to the target operating parameters and the real-time operating parameters; and correcting the set value of the operating parameters of the downstream flue gas treatment unit according to the composition deviation and the parameter deviation.

[0135] In a fourth aspect, the present application also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method are run.

[0136] Through the above technical solution, when the computer program is executed by the processor, it executes the method in any optional implementation manner of the above embodiments to implement the following functions: obtaining the glass formula information for production plan scheduling; predicting the change trend of flue gas composition after formula switching based on the correlation between the glass formula components and the pollutant concentration in the flue gas according to the glass formula information; calculating and setting the target operating parameters of each treatment unit according to the predicted change trend of flue gas composition after formula switching and the process characteristics and operating parameter thresholds of each flue gas treatment unit; obtaining the on-line monitoring data of the flue gas composition and the real-time operating parameters of each flue gas treatment unit; calculating the composition deviation between the actual flue gas composition and the predicted flue gas composition according to the on-line monitoring data of the flue gas composition and the predicted change trend of flue gas composition after formula switching; calculating the parameter deviation according to the target operating parameters and the real-time operating parameters; and correcting the set value of the operating parameters of the downstream flue gas treatment unit according to the composition deviation and the parameter deviation.

[0137] Among them, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0138] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be in an electrical, mechanical or other form.

[0139] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0140] Furthermore, in each embodiment of this application, the various functional modules may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part. The above are only the embodiments of this application and are not used to limit the protection scope of this application. For those skilled in the art, various changes and modifications can be made to this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the protection scope of this application.

Claims

1. A glass furnace fume treatment method, used for fume treatment under glass recipe switching conditions, the method comprising the following steps: Obtain glass formula information for production planning and scheduling; According to the glass formula information, based on the correlation between glass formula components and pollutant concentrations in flue gas, the trend of flue gas composition changes after formula switching is predicted; Calculate and set the target operating parameters of each level of treatment units based on the predicted flue gas composition change trend after recipe switching and the process characteristics and operating parameter thresholds of each level of flue gas treatment units; Obtain online monitoring data of flue gas composition and real-time operating parameters of each level of flue gas treatment units; The deviation between the actual smoke composition and the predicted smoke composition is calculated based on the online monitoring data of smoke composition and the change trend of smoke composition after the predicted recipe is switched; Calculate parameter deviations based on target operating parameters and real-time operating parameters; According to the composition deviation and parameter deviation, the operating parameter setting values ​​of the downstream flue gas treatment unit are corrected.

2. According to the glass furnace fume treatment method of claim 1, the step of predicting the change trend of fume components after formula switching based on the correlation between glass formula components and pollutant concentrations in the fume according to the glass formula information comprises: A polynomial regression model between the glass formula component vector and the smoke component concentration vector is established. The model function form is as follows: cj = f_j(R) = ∑_{i=1}^{n} a_{ji} * ri + ∑_{i=1}^{n} ∑_{k=1}^{n} b_{jik}* ri * rk + d_j, where cj is the predicted concentration of the j-th smoke component, R = [r1, r2, ..., rn] is the glass formula component vector, ri represents the proportion of the i-th formula component, a_{ji} is the linear regression coefficient, b_{jik} is the quadratic regression coefficient, which represents the influence of the interaction between the i-th and k-th formula components on the j-th smoke component concentration, and d_j is the bias term; Convert the glass formula information of production planning into glass formula component vector R; The glass formula component vector R is substituted into the polynomial regression model to calculate the smoke component concentration vector as the prediction result of the smoke component change trend after the formula is switched.

3. According to the glass furnace fume treatment method of claim 1, the step of calculating and setting the target operating parameters of each level of the fume treatment unit according to the predicted fume component change trend after the recipe is switched and the process characteristics and operating parameter thresholds of each level of the fume treatment unit comprises: Obtain pollutant removal efficiency curves of flue gas treatment units at various levels under different flue gas components, as well as the relationship between operating parameters of flue gas treatment units at various levels and pollutant removal efficiency; Based on the predicted trend of flue gas composition changes after recipe switching and combined with the pollutant removal efficiency curves of each level of flue gas treatment units, the expected pollutant removal amount of each level of flue gas treatment units under different operating parameters is calculated; According to the operating parameter thresholds of the flue gas treatment units at each level, the upper and lower limit constraints of the operating parameters of the flue gas treatment units at each level are determined; With the goal of minimizing total energy consumption, while satisfying the emission concentration of each pollutant being lower than the emission standard and the upper and lower limit constraints of each unit's operating parameters, the target operating parameters of each level of flue gas treatment units are calculated.

4. According to the glass furnace fume treatment method of claim 3, the step of calculating the target operating parameters of each level of fume treatment unit with the goal of minimizing total energy consumption and satisfying the emission concentration of each pollutant below the emission standard and the upper and lower limit constraints of each unit operation parameter comprises: A multi-objective optimization model with the objective function of minimizing total energy consumption is constructed. The objective function is expressed as: min ∑_{i=1}^{m} P_i(u_i), where P_i(u_i) is the energy consumption function of the i-th flue gas treatment unit, u_i is the operating parameter of the i-th flue gas treatment unit, and m is the number of flue gas treatment units; Set the pollutant emission concentration constraint, expressed as: g_j(u_1, u_2, ..., u_m) ≤ C_j, where g_j(u_1, u_2, ..., u_m) is the emission concentration function of the j-th pollutant and C_j is the emission standard of the j-th pollutant; Determine the upper and lower limit constraints of the operating parameters of each flue gas treatment unit, expressed as: u_{i,min} ≤ u_i ≤ u_{i,max}, where u_{i,min} and u_{i,max} are the lower and upper limits of the operating parameters of the i-th flue gas treatment unit, respectively; The multi-objective optimization model is solved iteratively. In each iteration, the gradients of the objective function and the constraints are calculated, and the operating parameters of each flue gas treatment unit are updated until the convergence conditions are met, thereby obtaining the target operating parameters of each level of flue gas treatment units.

5. According to the glass furnace fume treatment method of claim 1, the step of correcting the operating parameter setting value of the downstream fume treatment unit according to the component deviation and the parameter deviation comprises: Obtaining a control target type and a corresponding control target priority, wherein the control target type includes at least one of pollutant removal efficiency, energy consumption, and equipment life; According to the control target priority, determine the first weight of the pollutant removal efficiency deviation, the second weight of the energy consumption deviation, and the third weight of the equipment life deviation; According to the deviation between the actual flue gas composition and the predicted flue gas composition, the deviation of pollutant removal efficiency is calculated; according to the deviation between the actual operating parameters and the target operating parameters, the deviation of energy consumption and equipment life are calculated; Calculate the operating parameter correction amount of each level of downstream flue gas treatment units based on the first weight, the second weight, the third weight, the pollutant removal efficiency deviation, the energy consumption deviation, and the equipment life deviation; The operating parameter correction amount is added to the operating parameter setting value of each level of flue gas treatment unit to obtain the corrected operating parameter setting value.

6. According to the glass furnace fume treatment method of claim 1, the step of correcting the operating parameter setting value of the downstream fume treatment unit according to the component deviation and the parameter deviation comprises: Define the rolling optimization window length, which represents the number of future time steps that need to be predicted and optimized from the current moment; At each control moment, the following steps are performed: Obtain the smoke composition change trend prediction data and processing unit dynamic response characteristic parameters for the next k moments defined by the rolling optimization window length; Based on the composition deviation, parameter deviation, flue gas composition change trend prediction data and processing unit dynamic response characteristic parameters, a rolling optimization objective function is established, wherein the objective function includes a quadratic term of an operating parameter correction increment, a quadratic term of a pollutant removal efficiency deviation, and a quadratic term of an energy consumption deviation, and adds constraints on the rate of change of operating parameters of flue gas treatment units at all levels and upper and lower limits of corrections; Solving the rolling optimization objective function to obtain the optimal sequence of operating parameter correction increments from the current moment to the future k-1 moments; The operating parameter correction increment corresponding to the current moment in the optimal sequence is extracted, added to the operating parameter setting value of the downstream flue gas treatment unit, and the operating parameter setting value is updated.

7. According to the glass furnace fume treatment method of claim 6, the step of extracting the operating parameter correction increment corresponding to the current moment in the optimal sequence and adding it to the operating parameter setting value of the downstream fume treatment unit to update the operating parameter setting value comprises: Determine whether the current moment is the last moment of the rolling optimization window; If the current moment is not the last moment of the rolling optimization window, extracting the operating parameter correction increment corresponding to the current moment from the optimal sequence, and multiplying the operating parameter correction increment by a preset relaxation factor to obtain an adjusted operating parameter correction increment; The adjusted operating parameter correction increment is added to the operating parameter setting value of the downstream flue gas treatment unit to obtain an updated operating parameter setting value for flue gas treatment at the next control moment.

8. A glass furnace fume treatment device, used for fume treatment under glass recipe switching conditions, characterized in that: The device includes: The first acquisition module is used to obtain glass formula information for production planning and scheduling; The first prediction module is used to predict the change trend of flue gas composition after the formula is switched based on the glass formula information and the correlation between the glass formula components and the pollutant concentration in the flue gas; The first calculation module is used to calculate and set the target operating parameters of each level of the processing unit according to the predicted trend of flue gas composition changes after the recipe is switched and the process characteristics and operating parameter thresholds of each level of the flue gas treatment unit; The second acquisition module is used to obtain the online monitoring data of flue gas components and real-time operating parameters of the flue gas treatment units at each level; The second calculation module is used to calculate the component deviation between the actual smoke component and the predicted smoke component based on the online monitoring data of the smoke component and the change trend of the smoke component after the predicted recipe is switched; A third calculation module, used for calculating parameter deviation according to the target operating parameter and the real-time operating parameter; The first correction module is used to correct the operating parameter setting value of the downstream flue gas treatment unit according to the component deviation and the parameter deviation.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1 to 7 are executed.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are executed.

Images