Incinerator denitration control method and device, storage medium and electronic equipment

Through real-time parameter acquisition and target prediction model optimization algorithm, the urea injection amount is dynamically adjusted, which solves the problems of high urea consumption and ammonia escape in the denitrification control of the waste incinerator and achieves efficient and stable denitrification control effect.

CN120802607APending Publication Date: 2025-10-17CHINA ENFI ENG CORP +1
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202510715301.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-30
Publication Date
2025-10-17

AI Technical Summary

Technical Problem

The denitrification control of existing waste incinerators relies on manual operation, resulting in high urea consumption and serious ammonia escape, increasing operating costs and bringing environmental pollution risks.

Method used

The incinerator parameters are acquired in real time, the concentration is predicted using a pre-trained target prediction model, and the optimal control parameters are solved through a target optimization algorithm. The urea injection amount is dynamically optimized, and a multi-objective collaborative control strategy is combined to regulate the incinerator combustion and denitrification process.

Benefits of technology

On the premise of ensuring that NOx emission concentration meets the standard, urea consumption is significantly reduced, control accuracy and system economy are improved, the risk of ammonia escape is reduced, and operational stability is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120802607A_ABST
    Figure CN120802607A_ABST
Patent Text Reader

Abstract

The invention discloses an incinerator denitration control method and device, a storage medium and electronic equipment. The method relates to the technical field of incinerator-SNCR (selective non-catalytic reduction) denitration control, and comprises the following steps: acquiring current operating parameters and current control parameters of an incinerator in real time; based on the current operation parameter, the current control parameter and a control target, adopting a pre-trained target prediction model for prediction to obtain a concentration prediction value corresponding to the control target; based on the concentration predicted value and a preset threshold value corresponding to the control target, a target optimization algorithm is adopted to solve a preset optimization function, the optimal parameter value of a target control parameter is obtained when the function value of the preset optimization function is minimum, and the preset optimization function comprises the target control parameter; and controlling the denitration process of the incinerator by adopting the optimal parameter value. According to the method, the control precision, the economical efficiency and the environmental protection performance of the incinerator denitration control system can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of incinerator-SNCR denitration control, and particularly relates to an incinerator denitration control method and device, a storage medium and electronic equipment. BACKGROUND

[0002] At present, domestic waste incineration flue gas purification control mainly relies on the combination of PID automatic control and manual intervention. The operator needs to monitor the concentration of main pollutants through the DCS system, and manually adjusts combined with on-site inspection and chemical test data. However, due to the complex working condition of waste incineration and the harsh environment, part of the sensor and actuator is easy to be damaged, which causes the PID control system to be unable to operate stably for a long time. Under dynamic conditions such as large flow flue gas fluctuation, the traditional PID control response lags behind, and the control precision and stability are poor. In actual operation, most waste incineration power plants still highly rely on the experience of operators for adjustment, which has large operation load and strong dependence on personnel skills and state. In order to ensure that the emission meets the standard, the conservative strategy of over-injection of urea solution is often used, which causes high urea consumption and serious ammonia escape, not only increases the operation cost, but also brings the risk of secondary environmental pollution. SUMMARY

[0003] Therefore, the present application provides an incinerator denitration control method and device, a storage medium and electronic equipment, which mainly aims to solve the problem that the current incinerator denitration control process relies on technical personnel, uses the conservative strategy of over-injection of urea solution, causes high urea consumption and serious ammonia escape, not only increases the operation cost, but also brings the risk of secondary environmental pollution.

[0004] To solve the above problems, the present application provides an incinerator denitration control method, which comprises the following steps:

[0005] Real-time acquisition of current operation parameters and current control parameters of the incinerator;

[0006] Based on the current operation parameters, the current control parameters and the control target, a pre-trained target prediction model is used for prediction to obtain a concentration prediction value corresponding to the control target;

[0007] Based on the concentration prediction value and a preset limit value corresponding to the control target, a target optimization algorithm is used to solve a preset optimization function to obtain an optimal parameter value of the target control parameter, which makes the function value of the preset optimization function minimum, and the preset optimization function includes the target control parameter;

[0008] The optimal parameter value is used to control the denitration process of the incinerator.

[0009] Optionally, before predicting based on the current operating parameter, the current control parameter and a control target by using a pre-trained target prediction model, the method further comprises: constructing different prediction models;

[0010] The constructing different prediction models specifically comprises:

[0011] Obtaining historical operating data of the incinerator in different operating time periods;

[0012] Preprocessing the historical operating data to obtain preprocessed data;

[0013] Extracting features from the preprocessed data to obtain feature data;

[0014] Based on the feature data and a predetermined lag step, a time series model is constructed by using a target algorithm to obtain prediction models corresponding to different operating time periods respectively;

[0015] The feature extraction method comprises one or more of a tree model-based feature importance method, a recursive feature elimination method and a correlation analysis method.

[0016] Optionally, the preprocessing the historical operating data to obtain preprocessed data specifically comprises:

[0017] Performing outlier processing on the historical operating data by using a quartile range method to obtain first operating data;

[0018] Performing missing value processing on the first operating data by using a target difference algorithm to obtain second operating data, the target difference algorithm comprising one or more of a linear difference algorithm, a nearest neighbor difference algorithm and a trend prediction interpolation algorithm;

[0019] Performing filter denoising processing on the second operating data by using a sliding window average method to obtain the preprocessed data.

[0020] Optionally, before the time series model is constructed by using the target algorithm based on the feature data and a predetermined lag step to obtain prediction models corresponding to different operating time periods respectively, the method further comprises: determining the predetermined lag step;

[0021] The determining the predetermined lag step specifically comprises:

[0022] Based on historical urea solution flow and historical flue gas outlet nitrogen oxide concentration, a cross-correlation function is used for calculation and processing to obtain a cross-correlation degree between the historical urea solution flow and the historical flue gas outlet nitrogen oxide concentration;

[0023] The lag step corresponding to the maximum cross-correlation degree is determined as the predetermined lag step.

[0024] Optionally, when the control target comprises nitrogen oxide concentration and ammonia concentration of the flue gas purification control process, the target control parameter comprises a urea solution flow parameter;

[0025] The target optimization algorithm is used to solve a preset optimization function based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain an optimal parameter value of the target control parameter at which a function value of the preset optimization function is minimum, specifically comprising:

[0026] A first target optimization model is obtained based on the control target;

[0027] The first target optimization model is solved by using a particle swarm optimization algorithm based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain an optimal parameter value corresponding to the urea solution flow parameter.

[0028] Optionally, when the control target comprises nitrogen oxide concentration and ammonia concentration of the flue gas purification control process, and hearth temperature, main steam flow and outlet oxygen content of the incinerator combustion control process, the target control parameter comprises primary and secondary air volume, urea solution flow and grate speed;

[0029] The target optimization algorithm is used to solve a preset optimization function based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain an optimal parameter value of the target control parameter at which a function value of the preset optimization function is minimum, specifically comprising:

[0030] A second target optimization model is obtained based on the control target;

[0031] The second target optimization model is solved by using a genetic algorithm based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain optimal parameter values corresponding to the primary and secondary air volume, the urea solution flow and the grate speed respectively.

[0032] Optionally, the method further comprises:

[0033] An actual concentration value of the control target after the denitration process of the incinerator is controlled by using the optimal parameter value is obtained;

[0034] An error calculation process is performed based on the concentration prediction value and the actual concentration value, to obtain a prediction error;

[0035] The concentration prediction value output by the target prediction model at the next moment is corrected based on the prediction error.

[0036] To solve the above problems, the present application provides a denitration control device for an incinerator, comprising:

[0037] An acquisition module is used to obtain the current operating parameters and current control parameters of the incinerator in real time;

[0038] A prediction module, configured to perform prediction based on the current operating parameters, the current control parameters, and the control target using a pre-trained target prediction model to obtain a concentration prediction value corresponding to the control target;

[0039] a solving module, configured to solve a preset optimization function using a target optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain an optimal parameter value of a target control parameter that minimizes the function value of the preset optimization function, wherein the preset optimization function includes the target control parameter;

[0040] A control module is used to control the denitration process of the incinerator using the optimal parameter value.

[0041] In order to solve the above problems, the present application provides a storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned incinerator denitrification control method are implemented.

[0042] In order to solve the above problems, the present application provides an electronic device, which includes at least a memory and a processor. The memory stores a computer program, and the processor implements the steps of the above-mentioned incinerator denitrification control method when executing the computer program on the memory.

[0043] Beneficial effects of the present application: The present application constructs a time series prediction model that adapts to different seasonal operating conditions to make real-time predictions of the nitrogen oxide concentration at the chimney outlet, providing accurate feedforward data support for incinerator combustion regulation and urea injection control. Taking into full consideration the large delay, large inertia and strong nonlinear characteristics of the selective non-catalytic reduction (SNCR) denitrification process, dynamic optimization regulation is achieved based on model predictive control. Under a wide range of variable load operating conditions, it can significantly reduce urea consumption while ensuring that the nitrogen oxide NOx emission concentration meets the standard, thereby improving the economy and operational stability of the denitrification system. The present application uses a multi-objective collaborative control strategy to achieve joint optimization of the incineration process and the denitrification process, and can simultaneously control the urea flow rate and the furnace temperature, grate speed, and primary and secondary air volume in the incinerator. The incinerator combustion control parameters and ammonia slip constraints are introduced into the optimization objective function at the same time. While ensuring that the NOx emission concentration meets the standard, the urea injection amount is dynamically adjusted, which can not only suppress the generation of NOx from the source, but also effectively control the ammonia slip risk, thereby improving the control accuracy, economy and environmental performance of the overall system.

[0044] The above description is only a summary of the technical solutions of the present application. In order to enable one skilled in the art to better understand the technical means of the present application, the contents of the specification can be implemented, and in order to enable the above and other purposes, features and advantages of the present application to be more apparent and easy to understand, the specific embodiments of the present application are described below. BRIEF DESCRIPTION OF DRAWINGS

[0045] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of preferred embodiments and are not meant to limit the present application. Furthermore, the same reference numerals in the drawings indicate the same or similar components. In the drawings:

[0046] Figure 1 A flowchart of a method for controlling denitration of an incinerator is shown;

[0047] Figure 2 A flowchart of a method for controlling denitration of an incinerator is shown;

[0048] Figure 3 A block diagram of a device for controlling denitration of an incinerator is shown. DETAILED DESCRIPTION

[0049] The various aspects and features of the present application are described herein with reference to the accompanying drawings.

[0050] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be taken as limiting, but merely as exemplification of preferred embodiments. One skilled in the art will envision other modifications within the scope and spirit of the application.

[0051] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and, together with the general description of the application given above, and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0052] These and other characteristics of the present application will become apparent upon consideration of the following detailed description taken in conjunction with the accompanying drawings.

[0053] It should also be understood that, although the present application has been described with reference to certain specific examples, various modifications thereof will be readily apparent to those skilled in the art without departing from the spirit and scope of the present application.

[0054] The above and other aspects, features, and advantages of the present application will become apparent upon consideration of the following detailed description taken in conjunction with the accompanying drawings, when considered in light of the following detailed description.

[0055] The specific embodiments of the application will be described with reference to the drawings; however, it is to be understood that the application is merely an example of the application and can be implemented in various ways. Well-known and / or repetitive functions and structures are not described in detail to avoid unnecessary or redundant details that would obscure the application. Therefore, the specific structural and functional details of the application described herein are not intended to be limiting, but are merely presented as a basis and representative basis for the claims and for teaching one skilled in the art to use the application in substantially any appropriate detailed structure.

[0056] The specification can use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," which can refer to one or more embodiments according to the application.

[0057] The embodiments of the application provide a method for controlling denitration of an incinerator, as shown in the figure, comprising: Figure 1

[0058] Step S101: Real-time acquisition of current operating parameters and current control parameters of the incinerator;

[0059] In the specific implementation process, the current operating parameters include operating parameters such as furnace temperature, NOx concentration, flue gas temperature, and flue gas flow; and the current control parameters include control parameters such as primary and secondary air volume, urea solution flow, and grate speed.

[0060] Step S102: Based on the current operating parameters, the current control parameters, and the control target, a pre-trained target prediction model is used for prediction to obtain a concentration prediction value corresponding to the control target;

[0061] In the specific implementation process, based on the current operating time of the incinerator and the target operating period corresponding to the current operating time; based on the control target and the target operating period, a target prediction model is determined. The operating period can be divided according to seasons, for example: spring period, summer period, autumn period, and winter period; each operating period corresponds to a different prediction model. The target operating period in which the current operating time is located can be determined through the current operating time, and the target prediction model can be determined through the target operating period.

[0062] Step S103: Based on the concentration prediction value and the preset limit value corresponding to the control target, a target optimization algorithm is used to solve a preset optimization function, to obtain an optimal parameter value of a target control parameter that minimizes the function value of the preset optimization function, the target control parameter being included in the preset optimization function;

[0063] ​In the implementation process, when the control target includes the concentration of nitrogen oxides and the concentration of ammonia in the flue gas purification control process, the target control parameter includes a urea solution flow parameter; a first target optimization model is obtained based on the control target; the first target optimization model is solved by using a particle swarm optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain an optimal parameter value corresponding to the urea solution flow parameter. When the control target includes the concentration of nitrogen oxides and the concentration of ammonia in the flue gas purification control process and the hearth temperature, the main steam flow and the outlet oxygen content in the incinerator combustion control process, a second target optimization model is obtained based on the control target; the second target optimization model is solved by using a genetic algorithm based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain optimal parameter values corresponding to the primary and secondary air volume, the urea solution flow and the speed of the grate respectively. The optimization algorithm can also select a bee colony algorithm and a grey wolf algorithm to solve the first target optimization model and the second target optimization model to obtain the optimal parameter values of the target control parameters. The optimization algorithm can be selected according to actual needs.

[0064] Step S104: controlling the denitration process of the incinerator by using the optimal parameter value.

[0065] In the implementation process, the optimal parameter values of the target control parameters obtained by solving are applied to the DCS system to adjust the urea solution flow or other variables, and then prediction and optimization are performed again at the next time step. Through the rolling time domain strategy, the control strategy can be adjusted according to real-time feedback to ensure that the concentration of exhaust gas is within the reference trajectory range.

[0066] The application can provide accurate feedforward data support for incinerator combustion adjustment and urea injection control by constructing a time series prediction model suitable for different seasonal conditions to predict the concentration of nitrogen oxides at the chimney outlet in real time. The large delay, large inertia and strong nonlinearity characteristics in the selective non-catalytic reduction (SNCR) denitration process are fully considered, and dynamic optimization adjustment is realized based on model predictive control. Under a wide range of variable load operating conditions, the consumption of urea can be significantly reduced while ensuring that the concentration of NOx emissions meets the standard, thereby improving the economic efficiency and operating stability of the denitration system. The application realizes joint optimization of the incineration process and the denitration process through a multi-objective collaborative control strategy, which can control the urea flow and the hearth temperature, the speed of the grate and the primary and secondary air volume in the incinerator at the same time. The incinerator combustion control parameters and the ammonia escape constraint are introduced into the optimization objective function at the same time, so that the urea injection amount is dynamically adjusted while ensuring that the concentration of NOx emissions meets the standard, thereby inhibiting the generation of NOx from the source and effectively controlling the risk of ammonia escape, thereby improving the control accuracy, economic efficiency and environmental performance of the overall system.

[0067] Yet another embodiment of the present application provides another incinerator denitration control method, as shown in the accompanying drawings, comprising: Figure 2

[0068] Step S201: Construct a prediction model corresponding to each different operation time period;

[0069] ​In the specific implementation process of the present step, historical operation data of the incinerator in different operation time periods is obtained; the historical operation data includes: pusher speed, drying grate speed, combustion grate speed, burnout grate speed, primary air volume, secondary air volume, temperature of each measuring point in the hearth, SNCR urea solution flow, flue gas flow, oxygen content at the outlet of the waste heat boiler, flue gas temperature, main steam flow, NOx concentration and NH3 concentration at the outlet of the chimney, etc.; the historical operation data is preprocessed to obtain preprocessed data; specifically, the quartile range method is used to process outliers of the historical operation data to obtain first operation data; by calculating the quartiles (Q1 and Q3) and the interquartile range (IQR = Q3-Q1) of the data, the range of outliers is determined, and data lower than Q1-1.5*IQR or higher than Q3+1.5*IQR is determined as outliers and is removed or replaced to ensure the integrity and consistency of the data, thereby obtaining the first operation data. The target difference algorithm is used to process missing values of the first operation data to obtain second operation data, and the target difference algorithm includes one or several of linear difference algorithm, nearest neighbor difference algorithm and trend prediction interpolation algorithm; specifically, for missing values caused by sensor failure or communication interruption, interpolation method is used for filling, including linear interpolation, nearest neighbor interpolation or trend prediction interpolation based on historical data, to ensure the continuity and integrity of the data sequence. For data with long missing time, the data is directly removed to obtain the second operation data. The sliding window average method is used to filter and denoise the second operation data to obtain the preprocessed data. Specifically, in order to eliminate high-frequency noise and random fluctuations in the data, the sliding window average method is used to filter and smooth the data every 10 seconds, the moving average value in a fixed time window is calculated, the noise in the data is effectively reduced, and the overall trend and key features of the data are retained, thereby obtaining the preprocessed data. Feature extraction is performed on the preprocessed data to obtain feature data; specifically, the purpose of feature selection is to select key variables directly related to nitrogen oxide generation and denitration control from a large number of operation parameters, so as to reduce the complexity of the model, improve the prediction accuracy and system interpretability. The method of feature selection is not specifically required in the present application, and the most suitable algorithm model can be selected according to the specific scene and requirements. For example: the variable importance ranking is output by using random forest, XGBoost, etc. model training based on tree model feature importance, and several variables with high importance are selected. The recursive feature elimination method starts from all available variables (furnace temperature, flue gas temperature, flue gas pressure, flue gas flow, etc.), repeatedly fits the model, evaluates different importance, and removes the least important variables until a specified number of variables or an ideal model performance level is reached. The correlation analysis method takes the outlet NOx concentration as the target, calculates the Pearson correlation coefficient of each variable and the NOx concentration, and selects high-correlation variables as features (|r|>0.5).Determining the predetermined hysteresis step size specifically includes: based on the historical urea solution flow rate and the historical chimney outlet nitrogen oxide concentration, using a cross-correlation function to perform calculation processing to obtain the cross-correlation between the historical urea solution flow rate (X) and the historical chimney outlet nitrogen oxide concentration (Y); the mathematical expression of the cross-correlation function can be expressed as the following formula (1):

[0070]

[0071] Where CCF(τ) is the cross-correlation; is the historical average flow rate of urea solution; is the average value of nitrogen oxide concentration at the chimney outlet in history; σ X is the standard deviation of historical urea solution flow rate; σ Y is the standard deviation of the nitrogen oxide concentration at the historical chimney outlet. The lag step corresponding to the maximum cross-correlation is determined as the predetermined lag step; based on the characteristic data and the predetermined lag step, a target algorithm is used to construct a time series model to obtain prediction models corresponding to different operating time periods; wherein, the feature extraction method includes one or more of a feature importance method based on a tree model, a recursive feature elimination method, and a correlation analysis method. The target algorithm can be an extreme gradient boosting algorithm XGBoost, a lightweight gradient boosting machine algorithm LightGBM, an autoregressive integrated moving average model (ARIMA), a long short-term memory network model LSTM, a temporal convolutional network TCN, and other algorithms. The most suitable algorithm model can be selected according to specific needs and actual effects.

[0072] Step S202: obtaining the current operating parameters and current control parameters of the incinerator in real time;

[0073] During the specific implementation of this step, the current operating parameters include furnace temperature, NOx concentration, flue gas temperature, flue gas flow and other operating parameters; the current control parameters include primary and secondary air volume, urea solution flow, grate speed and other control parameters.

[0074] Step S203: performing prediction using a pre-trained target prediction model based on the current operating parameters, the current control parameters, and the control target to obtain a concentration prediction value corresponding to the control target;

[0075] In the implementation, the target prediction model is determined based on the control target and a target operation time period corresponding to the current operation time of the incinerator. The operation time period can be divided according to seasons, for example, a spring time period, a summer time period, an autumn time period and a winter time period, and each operation time period corresponds to a different prediction model. The target operation time period in which the current operation time is located can be determined according to the current operation time, and the target prediction model can be determined according to the target operation time period.

[0076] In step S204, when the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process, a first target optimization model is obtained based on the control target.

[0077] In the implementation, when the incinerator is not considered, the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process. The mathematical expression of the first target optimization model can be shown in formula (2) as follows:

[0078]

[0079] wherein λ1 is the ammonia concentration weight, λ2 is the nitrogen oxide concentration weight, Δ u is a control variable, and R is the weight of the control variable. When the incinerator is not considered, the control variable is the nitrogen oxide concentration change amount and the ammonia concentration change amount; NH 3k+i is the ammonia concentration prediction value; is the preset limit value of the ammonia concentration; NOx k+1 is the nitrogen oxide concentration prediction value; is the preset limit value of the nitrogen oxide concentration; k is an absolute time index of the current time, indicating a time point of a prediction start sequence, and i is a relative time index in a prediction time domain, indicating a future time step number starting from the current time.

[0080] In step S205, the first target optimization model is solved by using a particle swarm optimization algorithm based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain an optimal parameter value corresponding to the urea solution flow parameter.

[0081] In the implementation process, the algorithm parameters are initialized, including population size, maximum iteration number, inertia weight, learning factor, speed limit, and position limit; the population is initialized within the speed limit and position limit range, a plurality of particles are randomly generated, the position (parameter value corresponding to the target control parameter) of each particle is randomly initialized within the allowed range; the speed of each particle is initialized to 0 or a small random value; the individual optimal position is initialized as the current position; the global optimal position is initialized by evaluating the objective function J; the position of each particle is iteratively updated, and the position of each individual and the global optimal position are updated in each iteration round until the maximum iteration round is reached, and the parameter value of the target control parameter corresponding to the final global optimal position is determined as the global optimal parameter value.

[0082] Step S206: When the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process and the hearth temperature, the main steam flow, and the outlet oxygen content of the incinerator combustion control process, a second target optimization model is obtained based on the control target;

[0083] In the implementation process, if the incinerator combustion control and the flue gas purification control are considered at the same time, the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process and the hearth temperature, the main steam flow, and the outlet oxygen content of the incinerator combustion control process; the mathematical expression of the second target optimization model can be shown in the following formula (3):

[0084]

[0085] Wherein, λ1 is the ammonia concentration weight, λ2 is the nitrogen oxide concentration weight, Δ u is a control variable, and R is the weight of the control variable; if the incinerator combustion control and the flue gas purification control are considered at the same time, the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process and the hearth temperature, the main steam flow, and the outlet oxygen content of the incinerator combustion control process; NH 3k+i is the predicted value of the ammonia concentration; is the preset limit value of the ammonia concentration; NOx k+1 is the predicted value of the nitrogen oxide concentration; is the preset limit value of the nitrogen oxide concentration; T k+1 is the predicted parameter value of the hearth temperature, is the preset limit value of the hearth temperature; Q k+1 is the predicted parameter value of the main steam flow; is the preset limit value of the main steam flow; O k+1 is the predicted value of the outlet oxygen content; is a preset limit value of the outlet oxygen content; k is an absolute time index of the current time, indicating a time point of a prediction start timing sequence; and i is a relative time index in a prediction time domain, indicating a future time step number starting from the current time.

[0086] Step S207: The second target optimization model is solved by using a genetic algorithm based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain optimal parameter values corresponding to the primary and secondary air flow, the urea solution flow, and the grate speed, respectively.

[0087] In the implementation process, the algorithm parameters of the genetic algorithm are initialized, including chromosome coding, population size, maximum iteration number, selection method, crossover method, crossover probability, mutation method, and mutation probability; selection operation, crossover operation, mutation operation, evaluation of offspring fitness, environmental selection, determination of the current optimal individual to the next generation are performed in each iteration round, until the maximum iteration round is reached, and the chromosome with the highest fitness in the final population is obtained as the optimal solution, to obtain the optimal parameter values corresponding to the primary and secondary air flow, the urea solution flow, and the grate speed, respectively. The method for solving the second target optimization model also includes a particle swarm algorithm, a bee colony algorithm, and a grey wolf algorithm, which can be selected according to actual needs.

[0088] Step S208: The denitration process of the incinerator is controlled by using the optimal parameter values.

[0089] In the implementation process, the optimal parameter values of the target control parameters obtained by solving are applied to the DCS system to adjust the urea solution flow or other variables, and then prediction and optimization are performed again at the next time step. Through the rolling horizon strategy, the control strategy can be adjusted according to real-time feedback to ensure that the emission gas concentration is within the reference trajectory range.

[0090] Step S209: The actual concentration value of the control target after the denitration process of the incinerator is controlled by using the optimal parameter values is obtained.

[0091] In the implementation process, the nitrogen oxide concentration and the ammonia concentration at the chimney outlet and the furnace temperature, the main steam flow, and the outlet oxygen content of the incinerator combustion control process can be monitored in real time by using a flue gas analyzer, a temperature sensor, a flowmeter, an oxygen content analyzer, and a pressure sensor, to obtain the actual concentration value of the control target after the denitration process of the incinerator is controlled by using the optimal parameter values.

[0092] Step S210: Error calculation processing is performed based on the concentration prediction value and the actual concentration value, to obtain a prediction error.

[0093] In the implementation, the concentration prediction value and the actual concentration value are time-aligned to obtain an aligned concentration prediction value and an aligned actual concentration value; and the aligned concentration prediction value and the aligned actual concentration value are calculated to obtain a prediction error corresponding to different error indicators, including mean absolute error, root mean square error, mean absolute percentage error, and maximum absolute error.

[0094] Step S211: correcting the target prediction model based on the prediction error.

[0095] In the implementation, the prediction error is feature extracted to obtain a time-domain error feature and a frequency-domain error feature; and the target prediction model is corrected based on the time-domain error feature and the frequency-domain error feature. Specifically, when the error is greater than a first preset threshold, the model is retrained to reconstruct the target prediction model; when the error is greater than a second preset threshold and less than the first preset threshold, the target prediction model is incrementally learned to correct and update the target prediction model; and when the error is less than the second preset threshold, a model coefficient of the target prediction model is adjusted to correct and update the target prediction model. The first preset threshold and the second preset threshold can be set according to actual needs.

[0096] The application can provide accurate feedforward data support for combustion adjustment of the incinerator and control of the urea injection amount by constructing a time series prediction model suitable for different seasonal conditions to predict the concentration of nitrogen oxides at the outlet of the chimney in real time. The model predictive control is used to realize dynamic optimization adjustment by fully considering the large delay, large inertia, and strong nonlinear characteristics in the selective non-catalytic reduction (SNCR) denitration process. Under a wide range of variable load operation conditions, the urea consumption can be significantly reduced while ensuring that the concentration of NOx emissions meets the standard, thereby improving the economy and operation stability of the denitration system. The application can realize joint optimization of the incineration process and the denitration process by using a multi-objective collaborative control strategy, and can control the urea flow rate and the hearth temperature, the speed of the grate, and the primary and secondary air flow rates in the incinerator at the same time. The incinerator combustion control parameters and the ammonia escape constraint are introduced into the optimization objective function at the same time, so that the urea injection amount is dynamically adjusted while ensuring that the concentration of NOx emissions meets the standard, thereby inhibiting the generation of NOx from the source and effectively controlling the risk of ammonia escape, and thereby improving the control accuracy, economy, and environmental performance of the overall system.

[0097] Another embodiment of the application provides an incinerator denitration control device, as shown in Figure 3 The device comprises:

[0098] The acquisition module 1 is configured to acquire current operation parameters and current control parameters of the incinerator in real time.

[0099] a prediction module 2 configured to predict, based on the current operating parameter, the current control parameter, and a control target, a concentration prediction value corresponding to the control target by using a pre-trained target prediction model;

[0100] a solving module 3 configured to solve a preset optimization function by using a target optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain an optimal parameter value of a target control parameter that minimizes a function value of the preset optimization function, the preset optimization function including the target control parameter;

[0101] a control module 4 configured to control a denitration process of the incinerator by using the optimal parameter value.

[0102] In specific implementation, the incinerator denitration control device further includes a model construction module, which is specifically configured to construct different prediction models; the construction of different prediction models specifically includes: obtaining historical operating data of the incinerator in different operating time periods; pre-processing the historical operating data to obtain pre-processed data; performing feature extraction on the pre-processed data to obtain feature data; and constructing a time series model based on the feature data and a predetermined lag step by using a target algorithm to obtain prediction models corresponding to different operating time periods respectively; wherein the feature extraction method includes one or more of a tree model-based feature importance method, a recursive feature elimination method, and a correlation analysis method.

[0103] In specific implementation, the model construction module is further configured to: perform outlier processing on the historical operating data by using a quartile range method to obtain first operating data; perform missing value processing on the first operating data by using a target difference algorithm to obtain second operating data, the target difference algorithm including one or more of a linear difference algorithm, a nearest neighbor difference algorithm, and a trend prediction interpolation algorithm; and perform filter denoising processing on the second operating data by using a sliding window average method to obtain the pre-processed data.

[0104] In specific implementation, the model construction module is further configured to: determine the predetermined lag step; and the determination of the predetermined lag step specifically includes: based on historical urea solution flow and historical flue gas outlet nitrogen oxide concentration, performing calculation and processing by using a cross-correlation function to obtain a cross-correlation degree between the historical urea solution flow and the historical flue gas outlet nitrogen oxide concentration; and determining a lag step corresponding to the maximum cross-correlation degree as the predetermined lag step.

[0105] In the implementation process, the solving module 3 is specifically configured to: when the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process, the target control parameter includes a urea solution flow parameter; the target optimization algorithm is used to solve the preset optimization function based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain an optimal parameter value of the target control parameter at which a function value of the preset optimization function is minimized, specifically including: obtaining a first target optimization model based on the control target; the particle swarm optimization algorithm is used to solve the first target optimization model based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain the optimal parameter value corresponding to the urea solution flow parameter.

[0106] In the implementation process, the solving module 3 is further configured to: when the control target includes the nitrogen oxide concentration and the ammonia concentration of the flue gas purification control process and the hearth temperature, the main steam flow and the outlet oxygen content of the incinerator combustion control process, the target control parameter includes a primary and secondary air volume, a urea solution flow and a grate speed; the target optimization algorithm is used to solve the preset optimization function based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain an optimal parameter value of the target control parameter at which a function value of the preset optimization function is minimized, specifically including: obtaining a second target optimization model based on the control target; the genetic algorithm is used to solve the second target optimization model based on the concentration prediction value and the preset limit value corresponding to the control target, to obtain optimal parameter values corresponding to the primary and secondary air volume, the urea solution flow and the grate speed respectively.

[0107] In the implementation process, the device further includes a correction module, which is specifically configured to: obtain an actual concentration value of the control target after the denitration process of the incinerator is controlled by using the optimal parameter value; perform error calculation processing based on the concentration prediction value and the actual concentration value, to obtain a prediction error; and correct the target prediction model based on the prediction error.

[0108] The application predicts the concentration of nitrogen oxides at the chimney outlet in real time by constructing a time series prediction model suitable for different seasonal conditions, thereby providing accurate feedforward data support for combustion adjustment of the incinerator and control of the urea injection amount. The large delay, large inertia and strong nonlinearity characteristics in the SNCR denitration process are fully considered, and dynamic optimization adjustment is realized based on model predictive control. Under a wide range of variable load operating conditions, the urea consumption can be significantly reduced while ensuring that the NOx emission concentration meets the standard, thereby improving the economic efficiency and operating stability of the denitration system. The incinerator combustion control parameters and ammonia escape constraints are introduced into the optimization objective function in the application, and a multi-objective collaborative control strategy is used to realize joint optimization of the incineration process and the denitration process. While ensuring that the NOx emission concentration meets the standard, the urea injection amount is dynamically adjusted, which can not only inhibit the generation of NOx from the source but also effectively control the risk of ammonia escape, thereby improving the control accuracy, economic efficiency and environmental performance of the overall system.

[0109] Another embodiment of the application provides a storage medium storing a computer program, wherein the computer program is executed by a processor to implement the following method steps:

[0110] Step one, real-time acquisition of current operating parameters and current control parameters of the incinerator;

[0111] Step two, prediction based on the current operating parameters, the current control parameters and a control target using a pre-trained target prediction model to obtain a concentration prediction value corresponding to the control target;

[0112] Step three, solving of a preset optimization function using a target optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target to obtain an optimal parameter value of a target control parameter that minimizes the function value of the preset optimization function, wherein the target control parameter is included in the preset optimization function;

[0113] Step four, control of the denitration process of the incinerator using the optimal parameter value.

[0114] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0115] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of functional units and modules is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the above-mentioned functions.

[0116] The specific implementation process of the above method steps can be referred to the above-mentioned embodiment of the incinerator denitration control method, and this embodiment will not be repeated here.

[0117] The application constructs a time series prediction model suitable for different seasonal conditions to predict the concentration of nitrogen oxides at the chimney outlet in real time, and provides accurate feedforward data support for combustion adjustment of the incinerator and control of the urea injection amount. The dynamic optimization adjustment is realized based on model predictive control, fully considering the large delay, large inertia and strong nonlinearity characteristics in the SNCR denitration process. Under a wide range of variable load operation conditions, the urea consumption can be significantly reduced while ensuring that the NOx emission concentration meets the standard, thereby improving the economy and operation stability of the denitration system. The application realizes the joint optimization of the incineration process and the denitration process through a multi-objective collaborative control strategy, and can control the urea flow rate, the furnace temperature, the grate speed and the primary and secondary air flow rates in the incinerator at the same time. The incinerator combustion control parameters and the ammonia escape constraints are introduced into the optimization objective function, so that the urea injection amount is dynamically adjusted, the generation of NOx is inhibited from the source, and the ammonia escape risk is effectively controlled, thereby improving the control accuracy, economy and environmental performance of the overall system.

[0118] Another embodiment of the application provides an electronic device, which can be a server. The electronic device includes a processor, a memory, a network interface and a database connected by a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external client through a network connection. The electronic device program is executed by the processor to implement the functions or steps of the server side of the incinerator denitration control method.

[0119] In one embodiment, an electronic device is provided, which can be a client. The electronic device includes a processor, a memory, a network interface, a display screen and an input device connected by a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external server through a network connection. The electronic device program is executed by the processor to implement the functions or steps of the client side of the incinerator denitration control method.

[0120] Another embodiment of the application provides an electronic device, which includes at least a memory and a processor. The memory stores a computer program, and the processor implements the following method steps when executing the computer program on the memory:

[0121] Step one, real-time acquisition of current operating parameters and current control parameters of the incinerator;

[0122] Step two, based on the current operating parameters, the current control parameters and the control target, a pre-trained target prediction model is used for prediction to obtain a concentration prediction value corresponding to the control target;

[0123] Step three, based on the concentration prediction value and a preset limit value corresponding to the control target, a target optimization algorithm is used to solve a preset optimization function to obtain an optimal parameter value of the target control parameter which makes the function value of the preset optimization function minimum, and the preset optimization function includes the target control parameter;

[0124] Step four, using the optimal parameter value to control the denitration process of the incinerator.

[0125] The specific implementation process of the above method steps can be referred to the embodiments of any incinerator denitration control method described above, which will not be repeated here.

[0126] The present application constructs a time series prediction model suitable for different seasonal conditions to predict the concentration of nitrogen oxides at the chimney outlet in real time, providing accurate feedforward data support for incinerator combustion regulation and urea injection control. Fully considering the large delay, large inertia and strong nonlinearity characteristics in the SNCR denitration process, dynamic optimization adjustment is realized based on model predictive control. Under wide range variable load operation conditions, it can significantly reduce urea consumption while ensuring that the NOx emission concentration meets the standard, improving the economic efficiency and operation stability of the denitration system. The present application realizes the joint optimization of the incineration process and the denitration process through a multi-objective collaborative control strategy, which can control the urea flow rate and the furnace temperature, the speed of the grate, and the primary and secondary air flow rate in the incinerator at the same time. The incinerator combustion control parameters and ammonia escape constraints are introduced into the optimization objective function at the same time, which dynamically adjusts the urea injection amount while ensuring that the NOx emission concentration meets the standard, not only inhibiting the generation of NOx from the source, but also effectively controlling the risk of ammonia escape, thereby improving the control accuracy, economic efficiency and environmental performance of the overall system.

[0127] The above embodiments are only exemplary embodiments of the present application and are not used to limit the present application, the protection scope of the present application is defined by the claims. Those skilled in the art can make various modifications or equivalent replacements to the present application within the spirit and protection scope of the present application, and such modifications or equivalent replacements should also be considered to fall within the protection scope of the present application.

Claims

1. A method for controlling denitrification in an incinerator, characterized in that: include: Obtain the current operating parameters and current control parameters of the incinerator in real time; Based on the current operating parameters, the current control parameters and the control target, a pre-trained target prediction model is used to perform prediction to obtain a concentration prediction value corresponding to the control target; Solving a preset optimization function using a target optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target to obtain an optimal parameter value of a target control parameter when the function value of the preset optimization function is minimized, wherein the preset optimization function includes the target control parameter; The optimal parameter value is used to control the denitrification process of the incinerator.

2. The method according to claim 1, wherein Before performing prediction using a pre-trained target prediction model based on the current operating parameters, the current control parameters, and the control target, the method further includes: constructing different prediction models; The construction of different prediction models specifically includes: Obtain historical operating data of the incinerator in different operating time periods; Preprocessing the historical operation data to obtain preprocessed data; Performing feature extraction on the preprocessed data to obtain feature data; Based on the characteristic data and the predetermined lag step length, a target algorithm is used to construct a time series model to obtain prediction models corresponding to different operating time periods; Among them, the feature extraction method includes one or more of a feature importance method based on a tree model, a recursive feature elimination method, and a correlation analysis method.

3. The method according to claim 2, wherein The preprocessing of the historical operation data to obtain preprocessed data specifically includes: Performing outlier processing on the historical operating data using the interquartile range method to obtain first operating data; Performing missing value processing on the first operating data using a target difference algorithm to obtain second operating data, wherein the target difference algorithm includes one or more of a linear difference algorithm, a nearest neighbor difference algorithm, and a trend prediction interpolation algorithm; The second operating data is filtered and denoised using a sliding window averaging method to obtain the preprocessed data.

4. The method according to claim 2, wherein Before constructing a time series model using a target algorithm based on the characteristic data and the predetermined hysteresis step length to obtain prediction models corresponding to different operating time periods, the method further includes: determining the predetermined hysteresis step length; The determining of the predetermined hysteresis step size specifically includes: Based on the historical urea solution flow rate and the historical chimney outlet nitrogen oxide concentration, a cross-correlation function is used to perform calculation processing to obtain a cross-correlation between the historical urea solution flow rate and the historical chimney outlet nitrogen oxide concentration; The lag step length corresponding to the maximum cross-correlation is determined as the predetermined lag step length.

5. The method according to claim 1, wherein When the control target includes the nitrogen oxide concentration and ammonia concentration in the flue gas purification control process, the target control parameter includes a urea solution flow parameter; The method of solving a preset optimization function based on the concentration prediction value and the preset limit value corresponding to the control target using a target optimization algorithm to obtain an optimal parameter value of the target control parameter when the function value of the preset optimization function is minimized specifically includes: Acquire a first target optimization model based on the control target; The first target optimization model is solved using a particle swarm optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target to obtain an optimal parameter value corresponding to the urea solution flow parameter.

6. The method according to claim 1, wherein When the control targets include the nitrogen oxide concentration and ammonia concentration in the flue gas purification control process and the furnace temperature, main steam flow rate and outlet oxygen content in the incinerator combustion control process, the target control parameters include the primary and secondary air volumes, urea solution flow rate and grate speed; The method of solving a preset optimization function based on the concentration prediction value and the preset limit value corresponding to the control target using a target optimization algorithm to obtain an optimal parameter value of the target control parameter when the function value of the preset optimization function is minimized specifically includes: Based on the control objective, obtaining a second objective optimization model; Based on the concentration prediction value and the preset limit value corresponding to the control target, the second target optimization model is solved using a genetic algorithm to obtain the optimal parameter values ​​corresponding to the primary and secondary air volumes, the urea solution flow rate and the grate speed respectively.

7. The method according to claim 1, wherein The method further comprises: Obtaining an actual concentration value of a control target after controlling the denitration process of the incinerator using the optimal parameter value; Performing error calculation based on the concentration prediction value and the actual concentration value to obtain a prediction error; The target prediction model is corrected based on the prediction error.

8. An incinerator denitrification control device, characterized in that: include: An acquisition module is used to obtain the current operating parameters and current control parameters of the incinerator in real time; A prediction module, configured to perform prediction based on the current operating parameters, the current control parameters, and the control target using a pre-trained target prediction model to obtain a concentration prediction value corresponding to the control target; a solving module, configured to solve a preset optimization function using a target optimization algorithm based on the concentration prediction value and a preset limit value corresponding to the control target, to obtain an optimal parameter value of a target control parameter that minimizes the function value of the preset optimization function, wherein the preset optimization function includes the target control parameter; A control module is used to control the denitration process of the incinerator using the optimal parameter value.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the incinerator denitrification control method according to any one of claims 1 to 7 are implemented.

10. An electronic device, characterized in that: The method comprises at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the incinerator denitrification control method according to any one of claims 1 to 7 when executing the computer program on the memory.

Citation Information

Patent Citations

  • PM2.5 mask service life indicator card

    CN106404626A

  • Life-indicatable PM2.5 mask

    CN106501142A

  • Predictive control-based denitration regulation and control optimization system and method for cement decomposing furnace

    CN114609986A

  • Cooperative optimization control method for boiler combustion and denitration process

    CN115145152A

  • Denitration control method, device, storage medium, electronic equipment and system

    CN116236889A