Coal-fired boiler flue gas emission regulation and control method and system based on digital twinning

Abstract

The invention relates to the technical field of flue gas purification, and discloses a coal-fired boiler flue gas emission regulation and control method and system based on digital twinning, and the method comprises the steps: obtaining power grid dispatching data; setting a digital twinborn model of the boiler, and generating planned operation data of the boiler in the prediction time period according to the digital twinborn model and the power grid dispatching data; setting a deep peak regulation period according to the planned operation data; setting feed-forward signal data of the deep peak regulation time period; generating a first lag time according to the feedforward signal data; according to the planned operation data, the ammonia injection grid is divided into a plurality of partitions, and the specificity lag time of each partition is calculated; calculating a second lag time according to the first lag time and a specific lag time; and setting a flue gas denitration strategy according to the second lag time. According to the method, the delay time of ammonia spraying can be accurately judged under the condition of deep peak regulation, so that the ammonia spraying amount is matched with the actual working condition, and ammonia escape and substandard flue gas treatment are avoided.

Description

A method and system for controlling flue gas emissions from coal-fired boilers based on digital twins Technical Field

[0001] This application relates to the field of flue gas purification technology, and in particular to a method and system for regulating flue gas emissions from coal-fired boilers based on digital twins. Background Technology

[0002] Boiler flue gas emission treatment removes particulate matter, gaseous pollutants, and heavy metals from flue gas through physical, chemical, or biological methods to ensure compliance with emission standards. This includes initial dust removal of pollutants, followed by desulfurization and denitrification treatment of the flue gas. Boiler flue gas denitrification is a key measure for reducing nitrogen oxide emissions and improving air quality, and selective catalytic reduction using ammonia injection is the mainstream denitrification method.

[0003] Existing ammonia injection control systems suffer from the following technical defects: First, insufficient precision in ammonia injection timing control. Traditional ammonia injection control systems cannot accurately predict the ammonia injection lag time under different operating conditions, leading to a mismatch between ammonia injection and flue gas conditions. This may result in excessive ammonia injection causing ammonia escape, or insufficient ammonia injection leading to low emission efficiency. When the load changes rapidly, the fixed lag time setting cannot adapt to dynamically changing operating conditions. Second, poor uniformity of ammonia injection distribution. Existing systems typically employ a uniform ammonia injection strategy, ignoring the differences in flow characteristics between different zones within the flue gas duct, resulting in local over-injection or under-injection, which affects denitrification efficiency and increases the risk of ammonia escape. Especially when dealing with rapidly changing load conditions, existing technologies struggle to adjust the ammonia injection strategy in a timely manner. When abnormal flue gas parameters are detected, a long response time is usually required, failing to meet the stringent requirements of ultra-low emissions. Furthermore, insufficient coordination of ammonia injection between different zones leads to prominent local over-injection or under-injection problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a method and system for regulating flue gas emissions from coal-fired boilers based on digital twins, aiming to obtain a technical solution that can accurately determine the lag time of ammonia injection under deep peak shaving conditions and match the ammonia injection quantity with the operating conditions.

[0005] In some embodiments of this application, a method for regulating flue gas emissions from a coal-fired boiler based on digital twins is provided, characterized by comprising:

[0006] Obtain power grid dispatch data;

[0007] A digital twin model of the boiler is established, and the planned operation data of the boiler during the prediction period is generated based on the digital twin model and the power grid dispatch data.

[0008] Set deep peak shaving periods based on planned operation data;

[0009] Set the feedforward signal data for the deep peak-shaving period;

[0010] A first hysteresis time is generated based on the feedforward signal data;

[0011] Based on the planned operation data, the ammonia injection grid is divided into multiple zones and the specific lag time of each zone is calculated;

[0012] Calculate the second lag time based on the first lag time and the specific lag time;

[0013] The flue gas denitrification strategy is set according to the second lag time.

[0014] In some embodiments of this application, the step of setting the deep peak shaving period based on planned operation data includes:

[0015] Extract the load forecast curve from the planned operation data;

[0016] Calculate the load change rate of the load forecast curve;

[0017] The period during which the load change rate meets the first criterion is set as the deep peak shaving period.

[0018] In some embodiments of this application, the step of setting the feedforward signal data for the deep peak-shaving period includes:

[0019] Set each parameter in the planned operation data as a sub-parameter;

[0020] Calculate the correlation X between each sub-parameter and the ammonia injection rate data;

[0021] Construct a correlation sequence X, X=(X1,X2…Xi…Xn), where X1 is the correlation between the first sub-parameter and the ammonia injection rate data; Xi is the correlation between the i-th sub-parameter and the ammonia injection rate data; and n is the total number of sub-parameters.

[0022] The sub-parameters that meet the second judgment condition are set as feedforward signals;

[0023] The signal strength Q that generates the feedforward signal data;

[0024] Q= Si;

[0025] Where n is the total number of sub-parameters in the planned operation data; Xi is the weight value of the i-th sub-parameter; and Si is the i-th sub-parameter.

[0026] In some embodiments of this application, generating the first hysteresis time based on the feedforward signal data includes:

[0027] Obtain the reference lag time T 基 ;

[0028] Based on the preset signal strength-first correction coefficient relationship table, the first correction coefficient K1 is generated according to the signal strength.

[0029] Calculate the first lag time Z1;

[0030] Z1=T 基 *K1.

[0031] In some embodiments of this application, dividing the ammonia injection grid into multiple zones includes:

[0032] Extract flue gas concentration data, flue gas velocity data, and turbulence intensity data from the cross section where the ammonia injection grid is located in the planned operation data during the deep peak shaving period;

[0033] The partitions in the ammonia injection grid that satisfy the turbulence intensity data greater than the first threshold are set as Class I partitions;

[0034] The zones in the ammonia injection grid that meet the criteria of flue gas velocity data being greater than the second threshold or flue gas concentration data being greater than the third threshold are designated as Class II zones.

[0035] The zones in the ammonia injection grid that meet the conditions of flue gas velocity data being less than the second threshold and flue gas concentration data being less than the third threshold are designated as Class III zones.

[0036] In some embodiments of this application, calculating the specific lag time for each partition includes:

[0037] Construct a partition sequence F, F = (F1, F2, ..., Fi, ..., Fn), where Fi is the i-th partition; n is the total number of partitions; and Fn is the n-th partition.

[0038] Extract the flue gas velocity data of each zone and the distance data from the center of each zone to the flue gas concentration detection point;

[0039] Calculate the specific lag time Ti for each partition;

[0040] Ti = ki * Li / Vi;

[0041] Where Li is the distance data from the center of the i-th partition to the flue gas concentration detection point, Vi is the flue gas velocity data of the i-th partition, and ki is the partition correction coefficient of the i-th partition.

[0042] In some embodiments of this application, calculating the second lag time includes:

[0043] Generate the second lag time Z2i for each partition;

[0044] Z2i = Z1 + Ti.

[0045] In some embodiments of this application, the flue gas denitrification strategy includes:

[0046] Generate the priority sequence of each zone of the ammonia injection grid;

[0047] Based on the priority sequence and the second lag time, control commands are generated;

[0048] The control commands include:

[0049] The total ammonia injection volume of the ammonia injection grid is allocated according to the priority sequence.

[0050] In some embodiments of this application, the generation of the priority sequence for each partition of the ammonia injection grid includes:

[0051] Obtain the scheduled execution data for each partition;

[0052] Calculate the evaluation value P for each partition;

[0053] P=α*(Ci / Ca)+β*(Vi / Va)+γ*(1-Ei);

[0054] Where Ci is the flue gas concentration of the i-th partition, Ca is the average flue gas concentration of the ammonia injection section; Vi is the flue gas velocity of the i-th partition, Va is the average flue gas velocity of the ammonia injection section; Ei is the historical ammonia slip rate corresponding to the i-th partition; α is the first coefficient, β is the second coefficient, and γ is the third coefficient.

[0055] In some embodiments of this application, a digital twin-based flue gas emission control system for coal-fired boilers includes:

[0056] Data unit, used to acquire power grid dispatch data;

[0057] The central control unit is used to calculate the lag time based on the digital twin model;

[0058] The central control unit includes:

[0059] The first control module is used to set up a digital twin model of the boiler and generate planned operation data of the boiler within the predicted period based on the digital twin model and power grid dispatch data.

[0060] The second control module is used to set the deep peak shaving period based on the planned operation data, and to set the feedforward signal data for the deep peak shaving period.

[0061] The third control module is used to generate a first hysteresis time based on the feedforward signal data;

[0062] The fourth control module is used to divide the ammonia injection grid into multiple zones based on the planned operation data and calculate the specific lag time of each zone.

[0063] The fifth control module is used to calculate the second lag time based on the first lag time and the specific lag time;

[0064] An execution unit is used to set a flue gas denitrification strategy based on the second lag time. Compared with existing technologies, the flue gas emission control method for coal-fired boilers based on digital twins in this application embodiment has the following advantages:

[0065] A smart emission control system integrating feedforward prediction, dynamic lag compensation, and precise zone control was constructed. This method achieves accurate prediction of ammonia injection demand in the flue gas denitrification system by fusing boiler deep peak-shaving prediction operation data with digital twin simulation data. After feedforward prediction, the system can not only identify the differences in lag characteristics during different power change periods, but also accurately determine the dynamic lag time and control priority of each ammonia injection grid zone through the established CFD zoning model, greatly improving the matching accuracy and response speed between ammonia injection control and flue gas operating conditions.

[0066] Meanwhile, this method achieves dynamic optimization of the control strategy and rational allocation of ammonia. Based on the real-time flow characteristics, catalyst activity state, and ammonia slip risk of each ammonia injection zone, the system intelligently generates a zone control sequence, ensuring that ammonia injection tasks are preferentially allocated to zones with high lag time sensitivity and high mixing efficiency. This decision-making mechanism based on multi-dimensional evaluation significantly improves the utilization efficiency of ammonia injection resources, effectively ensuring denitrification efficiency while minimizing ammonia slip risk and operating costs. Attached Figure Description

[0067] Figure 1 is a flowchart illustrating a method for controlling flue gas emissions from a coal-fired boiler based on digital twins, as described in an embodiment of this application. Detailed Implementation

[0068] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0069] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0070] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0071] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0072] As shown in Figure 1, an embodiment of this application provides a method for controlling flue gas emissions from a coal-fired boiler based on digital twins, comprising:

[0073] Obtain power grid dispatch data;

[0074] A digital twin model of the boiler is set up, and the planned operation data of the boiler during the prediction period is generated based on the digital twin model and power grid dispatch data;

[0075] Set deep peak shaving periods based on planned operation data;

[0076] Set the feedforward signal data for the deep peak-shaving period;

[0077] The first hysteresis time is generated based on the feedforward signal data;

[0078] Based on the planned operation data, the ammonia injection grid is divided into multiple zones and the specific lag time of each zone is calculated;

[0079] The second lag time is calculated based on the first lag time and the specific lag time;

[0080] The flue gas denitrification strategy is set based on the second lag time.

[0081] Specifically, the power grid dispatch data is the boiler's power generation arrangement, which is obtained through the power grid dispatch center or plant-level monitoring information system. It is the demand plan for the power grid load for the next 5 days. The boiler's operating status is controlled based on this power grid dispatch data. When a large power change is required in a short period of time, the boiler needs to be deeply dispatched.

[0082] Specifically, a digital twin model is a boiler structure built in CFD software that is completely identical to the real boiler, using grid dispatch data and physical design parameters (such as furnace structure and heating surface arrangement) and historical operating big data (such as combustion parameters and flue gas resistance characteristics under different loads). This model is used to simulate the changes in flue gas during actual operation, thereby obtaining planned operating data. The planned operating data includes the expected flue gas flow rate, expected flue gas temperature, expected initial NOx concentration, expected steam parameters, flue gas concentration data, flue gas velocity data, and flue gas concentration and velocity data at various locations on the ammonia injection grid cross-section.

[0083] Specifically, feedforward signal data refers to the operational data generated during boiler operation that shows a high degree of similarity in changing trends to the ammonia injection rate.

[0084] In some embodiments of this application, deep peak shaving periods are set based on planned operation data, including:

[0085] Extract the load forecast curve from the planned operation data;

[0086] Calculate the load change rate of the load forecast curve;

[0087] The period during which the load change rate meets the first criterion is set as the deep peak shaving period.

[0088] Specifically, deep peak shaving periods are situations where large power fluctuations occur within a short period of time, under which the ammonia injection rate needs to be significantly changed.

[0089] Specifically, the first criterion is that the rate of change is greater than 3 MW / min.

[0090] In some embodiments of this application, the feedforward signal data for the deep peak-shaving period is set, including:

[0091] Set each parameter in the planned operation data as a sub-parameter;

[0092] Calculate the correlation X between each sub-parameter and the ammonia injection rate data;

[0093] Construct a correlation sequence X, X=(X1,X2…Xi…Xn), where X1 is the correlation between the first sub-parameter and the ammonia injection rate data; Xi is the correlation between the i-th sub-parameter and the ammonia injection rate data; and n is the total number of sub-parameters.

[0094] The sub-parameters that meet the second judgment condition are set as feedforward signals;

[0095] The signal strength Q that generates the feedforward signal data;

[0096] Q= Si;

[0097] Where n is the total number of sub-parameters in the planned operation data; Xi is the weight value of the i-th sub-parameter; and Si is the i-th sub-parameter.

[0098] Specifically, the absolute value of the correlation X between this sub-parameter and the ammonia injection data is greater than a preset threshold, preferably 0.7.

[0099] Specifically, the signal strength Q is the strength of the feedforward signal data, representing the magnitude of the data represented by the feedforward signal data.

[0100] Specifically, the sub-parameters are multiple types of operational data in the planned operation data, such as: load change rate, change in flue gas concentration, and change in flue gas flow rate.

[0101] Specifically, the correlation is calculated using the Pearson correlation coefficient algorithm.

[0102] In some embodiments of this application, generating a first hysteresis time based on feedforward signal data includes:

[0103] Obtain the reference lag time T 基 ;

[0104] Based on the preset signal strength-first correction coefficient relationship table, the first correction coefficient K1 is generated according to the signal strength.

[0105] Calculate the first lag time Z1;

[0106] Z1=T 基 *K1.

[0107] Specifically, the baseline lag time is the average lag time under normal operating conditions simulated by a digital twin model established using CFD software. The normal operating conditions are determined based on the most common operating conditions of the boiler and are provided by the power plant staff.

[0108] Specifically, the signal strength-first correction coefficient relationship table was provided by the power plant's technicians and experts in the field.

[0109] In some embodiments of this application, the ammonia injection grid is divided into multiple zones, including:

[0110] Extract flue gas concentration data, flue gas velocity data, and turbulence intensity data from the cross section where the ammonia injection grid is located in the planned operation data during the deep peak shaving period;

[0111] The partitions in the ammonia injection grid that satisfy the turbulence intensity data greater than the first threshold are set as Class I partitions;

[0112] The zones in the ammonia injection grid that meet the criteria of flue gas velocity data being greater than the second threshold or flue gas concentration data being greater than the third threshold are designated as Class II zones.

[0113] The zones in the ammonia injection grid that meet the conditions of flue gas velocity data being less than the second threshold and flue gas concentration data being less than the third threshold are designated as Class III zones.

[0114] Specifically, the concentration at the ammonia injection grid varies, and the ammonia injection grid is spatially divided according to the flue gas concentration data. The control lag time at different flue gas concentrations is related to and different from the flue gas concentration. Each zone contains multiple ammonia injection grid nozzles.

[0115] Specifically, one category of zones: These zones have unstable flow fields, and the mixing state of ammonia and flue gas is difficult to predict, requiring special attention. The first threshold can be determined based on the standard deviation of historical flow field simulation data.

[0116] Category II zones: These zones have short flue gas residence times, requiring earlier or more ammonia injection.

[0117] Category III Zoning: Zoning that meets this condition is designated as Category III zoning. These zoning zones have longer flue gas residence times, allowing for slightly later or less frequent ammonia injection. Finally, the ammonia injection grid is divided according to the zoning conditions, forming multiple independently controlled zones. For example, the entire grid is divided into n zones: F1, F2, ..., Fi, ..., Fn.

[0118] In some embodiments of this application, the specific lag time for each partition is calculated, including:

[0119] Construct a partition sequence F, F = (F1, F2, ..., Fi, ..., Fn), where Fi is the i-th partition; n is the total number of partitions; and Fn is the n-th partition.

[0120] Extract the flue gas velocity data of each zone and the distance data from the center of each zone to the flue gas concentration detection point;

[0121] Calculate the specific lag time Ti for each partition;

[0122] Ti = ki * Li / Vi;

[0123] Where Li is the distance data from the center of the i-th partition to the flue gas concentration detection point, Vi is the flue gas velocity data of the i-th partition, and ki is the partition correction coefficient of the i-th partition.

[0124] Specifically, the flue gas concentration detection point is the NOx measurement point at the outlet of the SCR reactor.

[0125] Specifically, a partition is considered as a regular geometric shape (such as a rectangle, circle, or polygon), and its geometric center is the center of the partition.

[0126] In some embodiments of this application, calculating the second lag time includes:

[0127] Generate the second lag time Z2i for each partition;

[0128] Z2i = Z1 + Ti.

[0129] Specifically, the first lag time is the overall lag velocity to the ammonia injection grid, which is calculated based on the average flow velocity throughout the flue. It serves as the unified time reference for ammonia injection control throughout the entire system.

[0130] Specifically, the second lag time is a time offset that represents how much earlier or later the ammonia injection command needs to be issued compared to the base time for each partition i.

[0131] In some embodiments of this application, the flue gas denitrification strategy includes:

[0132] Generate the priority sequence of each zone of the ammonia injection grid;

[0133] Control commands are generated based on the priority sequence and the second lag time.

[0134] Control commands include:

[0135] The total ammonia injection volume of the ammonia injection grid is allocated according to a priority sequence.

[0136] Specifically, the priority sequence is a ranking of the importance and urgency of each partition.

[0137] In some embodiments of this application, the priority sequence of each partition of the ammonia injection grid is generated, including:

[0138] Obtain the scheduled execution data for each partition;

[0139] Calculate the evaluation value P for each partition;

[0140] P=α*(Ci / Ca)+β*(Vi / Va)+γ*(1-Ei);

[0141] Where Ci is the flue gas concentration of the i-th partition, Ca is the average flue gas concentration of the ammonia injection section; Vi is the flue gas velocity of the i-th partition, Va is the average flue gas velocity of the ammonia injection section; Ei is the historical ammonia slip rate corresponding to the i-th partition; α is the first coefficient, β is the second coefficient, and γ is the third coefficient.

[0142] Specifically, α + β + γ = 1, where α is preferably 0.2, β is preferably 0.3, and γ is preferably 0.5.

[0143] Specifically, a higher evaluation value P indicates greater importance and priority in resource allocation and rapid response.

[0144] The total ammonia injection volume will be allocated based on a priority sequence. Zones with higher priority will receive a higher-than-average ammonia injection volume weight to ensure the denitrification effect of key zones.

[0145] When the calculated total ammonia injection demand exceeds the preset supply threshold of the ammonia injection system (when the ammonia pumps have reached maximum output), based on a priority sequence, the ammonia injection demand of high-priority zones is prioritized, temporarily sacrificing the supply to low-priority zones, with the goal of optimizing overall denitrification efficiency. Allocation must be based on the corresponding lag time. For example, if the ammonia injection data for the next 3 minutes has a lag time of 1 minute, then ammonia injection should be performed in the second minute of the next period, taking the lag time into account.

[0146] In some embodiments of this application, the data unit is used to acquire power grid dispatch data;

[0147] The central control unit is used to calculate the lag time based on the digital twin model;

[0148] The central control unit includes:

[0149] The first control module is used to set up the digital twin model of the boiler and generate the planned operation data of the boiler within the predicted period based on the digital twin model and the power grid dispatch data.

[0150] The second control module is used to set the deep peak shaving period based on the planned operation data and to set the feedforward signal data for the deep peak shaving period.

[0151] The third control module is used to generate the first lag time based on the feedforward signal data;

[0152] The fourth control module is used to divide the ammonia injection grid into multiple zones based on the planned operation data and calculate the specific lag time of each zone.

[0153] The fifth control module is used to calculate the second lag time based on the first lag time and the specific lag time;

[0154] The execution unit is used to set the flue gas denitrification strategy based on the second lag time.

[0155] The above are merely preferred embodiments of this application. It should be noted that, for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A method for regulating flue gas emissions from a coal-fired boiler based on digital twins, characterized in that, include: Obtain power grid dispatch data; A digital twin model of the boiler is established, and planned operation data of the boiler within the predicted period is generated based on the digital twin model and power grid dispatch data; a deep peak shaving period is set based on the planned operation data; feedforward signal data for the deep peak shaving period is set; and a first lag time is generated based on the feedforward signal data. Based on the planned operation data, the ammonia injection grid is divided into multiple zones and the specific lag time of each zone is calculated; Calculate the second lag time based on the first lag time and the specific lag time; The flue gas denitrification strategy is set according to the second lag time.

2. The method for regulating flue gas emissions from a coal-fired boiler based on digital twins as described in claim 1, characterized in that, The step of setting deep peak shaving periods based on planned operation data includes: extracting the load forecast curve from the planned operation data; calculating the load change rate of the load forecast curve; and setting the period in which the load change rate meets the first judgment condition as the deep peak shaving period.

3. The method for regulating flue gas emissions from a coal-fired boiler based on digital twins as described in claim 2, characterized in that, The process of setting the feedforward signal data for the deep peak-shaving period includes: setting each parameter in the planned operation data as a sub-parameter; calculating the correlation X between each sub-parameter and the ammonia injection rate data; constructing a correlation sequence X, X=(X1,X2…Xi…Xn), where X1 is the correlation between the first sub-parameter and the ammonia injection rate data; Xi is the correlation between the i-th sub-parameter and the ammonia injection rate data; n is the total number of sub-parameters; setting the sub-parameters that meet the second judgment condition as feedforward signals; and generating the signal strength Q of the feedforward signal data; Q= Si; where n is the total number of sub-parameters in the planned operation data; Xi is the weight value of the i-th sub-parameter; Si is the i-th sub-parameter.

4. The method for controlling flue gas emissions from a coal-fired boiler based on digital twins as described in claim 3, characterized in that, The step of generating the first lag time based on the feedforward signal data includes: obtaining a reference lag time T. 基 Based on the preset signal strength-first correction coefficient relationship table, generate the first correction coefficient K1 according to the signal strength; calculate the first lag time Z1; Z1=T 基 *K1.

5. The method for controlling flue gas emissions from a coal-fired boiler based on digital twins as described in claim 4, characterized in that, The process of dividing the ammonia injection grid into multiple zones includes: extracting flue gas concentration data, flue gas velocity data, and turbulence intensity data from the cross section where the ammonia injection grid is located from the planned operation data during the deep peak shaving period; setting the zones in the ammonia injection grid that meet the condition that the turbulence intensity data is greater than a first threshold as Class I zones; setting the zones in the ammonia injection grid that meet the condition that the flue gas velocity data is greater than a second threshold or the flue gas concentration data is greater than a third threshold as Class II zones; and setting the zones in the ammonia injection grid that meet the condition that the flue gas velocity data is less than a second threshold and the flue gas concentration data is less than a third threshold as Class III zones.

6. The method for controlling flue gas emissions from a coal-fired boiler based on digital twins as described in claim 5, characterized in that, The calculation of the specific lag time for each partition includes: constructing a partition sequence F, F=(F1,F2…Fi…Fn), where Fi is the i-th partition; n is the total number of partitions; Fn is the n-th partition; extracting the flue gas velocity data and the distance data from the center of each partition to the flue gas concentration detection point; calculating the specific lag time Ti for each partition; Ti=ki*Li / Vi; where Li is the distance data from the center of the i-th partition to the flue gas concentration detection point, Vi is the flue gas velocity data of the i-th partition, and ki is the partition correction coefficient of the i-th partition.

7. The method for regulating flue gas emissions from a coal-fired boiler based on digital twins as described in claim 6, characterized in that, The calculation of the second lag time includes: generating the second lag time Z2i for each partition; Z2i = Z1 + Ti.

8. The method for controlling flue gas emissions from a coal-fired boiler based on digital twins as described in claim 7, characterized in that, The flue gas denitrification strategy includes: generating a priority sequence for each zone of the ammonia injection grid; generating control commands based on the priority sequence and a second lag time; the control commands include: allocating the total ammonia injection amount of the ammonia injection grid according to the priority sequence.

9. The method for regulating flue gas emissions from a coal-fired boiler based on digital twins as described in claim 8, characterized in that, The process of generating the priority sequence for each zone of the ammonia injection grid includes: acquiring the planned operation data for each zone; calculating the evaluation value P for each zone; P=α*(Ci / Ca)+β*(Vi / Va)+γ*(1-Ei); where Ci is the flue gas concentration of the i-th zone, Ca is the average flue gas concentration of the ammonia injection section; Vi is the flue gas velocity of the i-th zone, Va is the average flue gas velocity of the ammonia injection section; Ei is the historical ammonia slip rate corresponding to the i-th zone; α is the first coefficient, β is the second coefficient, and γ is the third coefficient.

10. A digital twin-based flue gas emission control system for coal-fired boilers, used to execute a digital twin-based flue gas emission control method for coal-fired boilers as described in any one of claims 1-9, characterized in that, include: Data unit, used to acquire power grid dispatch data; The central control unit is used to calculate the lag time based on the digital twin model; The central control unit includes: a first control module for setting a digital twin model of the boiler and generating planned operation data of the boiler within a predicted period based on the digital twin model and power grid dispatch data; a second control module for setting a deep peak shaving period based on the planned operation data and setting feedforward signal data for the deep peak shaving period; a third control module for generating a first lag time based on the feedforward signal data; a fourth control module for dividing the ammonia injection grid into multiple zones based on the planned operation data and calculating the specific lag time of each zone; a fifth control module for calculating a second lag time based on the first lag time and the specific lag time; and an execution unit for setting a flue gas denitrification strategy based on the second lag time.

Images