Denitration system inlet nitric oxide disturbance and over-emission control method and system

Abstract

The invention discloses a denitration system inlet nitric oxide disturbance and super-emission control method and system, and the method comprises the steps: adjusting the operation oxygen amount and the over fire air ratio in real time based on the measured values of the carbon monoxide concentration and the nitric oxide concentration of the denitration system inlet; based on denitration system operation data, combining actual sampling delay time to carry out sample data alignment and construct a training sample set; constructing a prediction model based on the training sample set, obtaining a nitrogen oxide concentration prediction value of the inlet of the denitration system through the prediction model, and updating the prediction model in real time; based on the difference between the average value of the outlet nitrogen oxide concentration of each partition and a set value and the difference between the concentration value of the nitrogen oxide at the outlet of the denitration system and the set value, and by combining the predicted value of the nitrogen oxide concentration at the inlet of the denitration system, treatment measures after the nitrogen oxide concentration super-emission are set in a segmented mode. The control performance of the denitration system can be improved, the ammonia spraying amount is reduced, the ammonia escape concentration is reduced, and the system automation input rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler denitration, and in particular to a method and system for controlling disturbance and over-emission of nitrogen oxides at an inlet of a denitration system. Background Art

[0002] In actual operation, denitrification systems suffer from measurement lag, uneven flow field distribution, and excessive ammonia injection. To address these issues, the denitrification system began integrating a precision ammonia injection system. This system implements zoned control by adding zoned valves to ensure uniformity in the denitrification flow field. A prediction model is trained offline using historical data, combining this prediction model with PID control to precisely control the amount of ammonia injected.

[0003] However, as the boiler operates flexibly, air distribution, coal feed rate, mill operation mode, coal quality composition, etc. change all the time, resulting in large changes in the nitrogen oxide concentration at the boiler outlet. In severe cases, the original emission nitrogen oxide concentration can change rapidly by 300 mg / Nm3 within 2 to 3 minutes. 3 The existing precise ammonia injection methods mentioned above still have the following problems, and the adjustment effect is not good under the flexible operation of the boiler:

[0004] (1) There is no means to stabilize the concentration of nitrogen oxides emitted by the boiler. It can only be passively adjusted after the fluctuation of nitrogen oxides at the denitrification inlet is found, and the system adjustment effect is poor;

[0005] (2) When the unit operating parameters change, the offline trained prediction model cannot be updated in real time, resulting in prediction deviations and affecting the control effect;

[0006] (3) Rapid changes in NOx at the denitrification system inlet can cause fluctuations in NOx at the outlet. Existing precision ammonia injection systems only use the prediction model results as feedforward regulation, without emergency measures in the event of NOx over-emission at the outlet. Insufficient valve adjustment can lead to prolonged NOx over-emission at the outlet. If the over-emission time is too long or the peak value is too high, the operator needs to switch off the automatic mode and manually adjust the valve. Summary of the Invention

[0007] In order to solve the above problems, the present invention proposes a method and system for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system, which can improve the control performance of the denitrification system, reduce the amount of ammonia sprayed, reduce the ammonia escape concentration, and increase the system automation input rate.

[0008] The technical solution adopted in the present invention is as follows:

[0009] A method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system, comprising:

[0010] Based on the measured values ​​of carbon monoxide concentration and nitrogen oxide concentration at the denitrification system inlet, the operating oxygen content and burnout air ratio are adjusted in real time;

[0011] Based on the denitrification system operating data and the actual sampling delay time, sample data is aligned and a training sample set is constructed. A prediction model is built based on the training sample set. The prediction model is used to obtain the predicted value of nitrogen oxide concentration at the denitrification system inlet, and the prediction model is updated in real time.

[0012] Based on the difference between the average nitrogen oxide concentration at the outlet of each zone and the set value, as well as the difference between the nitrogen oxide concentration at the outlet of the denitrification system and the set value, combined with the predicted nitrogen oxide concentration at the inlet of the denitrification system, the treatment measures after the nitrogen oxide concentration exceeds the discharge limit are set in sections.

[0013] Furthermore, the real-time adjustment of the operating oxygen amount and the overfire air ratio based on the measured values ​​of the carbon monoxide concentration and the nitrogen oxide concentration at the denitrification system inlet includes:

[0014] Calculating the average carbon monoxide concentration at the denitration system inlet within a preset time period. If the average carbon monoxide concentration is greater than a first threshold, adjusting the operating oxygen level according to the current operating conditions based on a curve showing the relationship between the oxygen reduction value and the carbon monoxide concentration.

[0015] The average nitrogen oxide concentration at the inlet of the denitrification system within a preset time is calculated. If the average nitrogen oxide concentration is greater than the second threshold, the burnout damper opening is increased and the secondary damper opening is decreased.

[0016] Furthermore, the method for generating the relationship curve between the oxygen reduction value and the carbon monoxide concentration includes:

[0017] The historical data of the carbon monoxide concentration and oxygen value at the boiler economizer outlet under all operating conditions are retrieved, and a curve of the relationship between the oxygen reduction value and the carbon monoxide concentration is analyzed and drawn, where the horizontal axis is the carbon monoxide concentration value and the vertical axis is the oxygen reduction value.

[0018] Furthermore, the sample data is aligned based on the denitration system operation data in combination with the actual sampling delay time to construct a training sample set, including:

[0019] Retrieving denitration system operating data, the denitration system operating data including boiler load, economizer outlet oxygen content, total coal quantity, burnout air volume, and denitration system inlet nitrogen oxide concentration;

[0020] Analyze and predict the delay time based on actual sampling influencing factors, including the length of the measurement pipeline and the time it takes for nitrogen oxides at the denitrification system inlet to change after operating parameters change;

[0021] The retrieved denitrification system operation data is organized into multiple training sample sets according to the delay time.

[0022] Furthermore, the prediction model is constructed based on the training sample set, a predicted value of the nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model, and the prediction model is updated in real time, including:

[0023] Based on the training sample set, the random forest algorithm is used for training to form a prediction model;

[0024] Real-time data is acquired and the predicted value of nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model; when the deviation between the predicted value and the actual value exceeds the allowable error, the prediction model is updated online.

[0025] Furthermore, after obtaining the predicted value of the nitrogen oxide concentration at the inlet of the denitration system through the prediction model, the ammonia flow rate that needs to be adjusted is calculated according to the predicted value of the nitrogen oxide concentration, and is adjusted in a feedforward manner.

[0026] Furthermore, the treatment measures after the nitrogen oxide concentration exceeds the discharge level are set in sections, including:

[0027] Compensate the ammonia flow rate according to the zoned nitrogen oxide concentration: Based on the fact that the nitrogen oxide concentration measurement pipelines at the denitrification system outlet and each zone outlet are shorter than the nitrogen oxide concentration measurement pipelines at the chimney outlet and have a relatively smaller measurement lag, the ammonia flow rate is compensated accordingly by the difference between the predicted value and the set value of the nitrogen oxide concentration emission at the denitrification system outlet and each zone outlet.

[0028] Furthermore, the treatment measures after the nitrogen oxide concentration exceeds the discharge level are set in sections, including:

[0029] Compensate the ammonia flow rate according to the nitrogen oxide concentration at the chimney outlet: Based on the relatively large measurement lag of the nitrogen oxide concentration at the chimney outlet, the ammonia flow rate is compensated in time by the difference between the predicted value and the set value of the nitrogen oxide concentration emission at the chimney outlet.

[0030] Furthermore, the stepwise setting of treatment measures after excessive discharge of nitrogen oxide concentrations also includes:

[0031] According to the changing trend of the predicted value of nitrogen oxide concentration at the inlet of the denitrification system, the segmented influence coefficient of the ammonia flow is adjusted dynamically in real time: when the changing rate of the predicted value of nitrogen oxide concentration is large, the segmented influence coefficient of the ammonia flow is increased; when the changing rate of the predicted value of nitrogen oxide concentration is small, the segmented influence coefficient of the ammonia flow is reduced.

[0032] A denitrification system inlet nitrogen oxide disturbance and over-emission control system, comprising:

[0033] a combustion adjustment module configured to adjust the operating oxygen content and the overfire air ratio in real time based on the measured values ​​of the carbon monoxide concentration and the nitrogen oxide concentration at the denitrification system inlet;

[0034] The model training and prediction module is configured to align sample data based on the denitrification system operating data and the actual sampling delay time to construct a training sample set; build a prediction model based on the training sample set, use the prediction model to obtain the predicted value of nitrogen oxide concentration at the denitrification system inlet, and update the prediction model in real time;

[0035] The over-emission control module is configured to set the treatment measures after the nitrogen oxide concentration exceeds the emission limit in sections based on the difference between the average nitrogen oxide concentration at the outlet of each zone and the set value, as well as the difference between the nitrogen oxide concentration at the outlet of the denitrification system and the set value, combined with the predicted nitrogen oxide concentration at the inlet of the denitrification system.

[0036] The beneficial effects of the present invention are:

[0037] 1. This invention incorporates combustion control into precise ammonia injection. This ensures stable NOx concentrations from boiler emissions through operational oxygen adjustment and real-time adjustments to the burnout damper and secondary damper. Real-time adjustments to the unit's oxygen content and burnout damper prevent significant fluctuations in NOx concentrations, reducing significant disturbances in NOx concentrations at the denitrification inlet by approximately 50%.

[0038] 2. This invention utilizes a random forest algorithm combined with actual sampling delay time to align sample data and predict the NOx concentration at the denitrification system inlet. The prediction model is then updated in real time, improving the accuracy of NOx concentration predictions and significantly enhancing the control accuracy and response time of the denitrification system. This online update of the prediction model ensures the system's adaptability to flexible operating conditions, with a steady-state error of less than ±10%.

[0039] 3. By integrating prediction results and the difference between actual emissions and setpoints into the existing control system to establish an emergency response method, this method enhances the system's logical stability and adaptability to complex operating conditions, reduces the adverse effects of measurement lag, and increases the system's automation rate. By incorporating emergency response measures after over-emissions, the automation rate of the denitrification system can reach 100%. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of a method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to Example 1 of the present invention.

[0041] Figure 2 This is a logic diagram of oxygen bias setting in embodiment 1 of the present invention.

[0042] Figure 3 This is a logic diagram of the burnout damper offset setting in Example 1 of the present invention.

[0043] Figure 41 is a graph showing the relationship between the oxygen reduction value and the CO concentration in Example 1 of the present invention.

[0044] Figure 5 This is a flow chart of the prediction model training of Example 1 of the present invention.

[0045] Figure 6 This is the intelligent feedforward + cascade PID control logic diagram of Example 1 of the present invention.

[0046] Figure 7 Schematic diagram of emergency treatment coefficients after over-discharge in Example 1 of the present invention.

[0047] Figure 8 This is a schematic diagram of the effect before the emergency treatment measures are applied in Example 1 of the present invention.

[0048] Figure 9 It is a schematic diagram of the effect after the emergency treatment measures are applied in Example 1 of the present invention. DETAILED DESCRIPTION

[0049] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific embodiments of the present invention are now described. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. That is, the embodiments described are only part of the embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment provides a method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system, comprising:

[0052] Based on the measured values ​​of carbon monoxide concentration and nitrogen oxide concentration at the denitrification system inlet, the operating oxygen amount and the burnout air ratio are adjusted in real time to stabilize the nitrogen oxide concentration at the denitrification system inlet;

[0053] Based on the denitrification system operating data and the actual sampling delay time, sample data is aligned and a training sample set is constructed. A prediction model is built based on the training sample set. The prediction model is used to obtain the predicted value of nitrogen oxide concentration at the denitrification system inlet, and the prediction model is updated in real time.

[0054] Based on the difference between the average nitrogen oxide concentration at the outlet of each zone and the set value, as well as the difference between the nitrogen oxide concentration at the outlet of the denitrification system and the set value, combined with the predicted nitrogen oxide concentration at the inlet of the denitrification system, the treatment measures after the nitrogen oxide concentration exceeds the discharge limit are set in sections.

[0055] Preferably, the operating oxygen amount and the overfire air ratio are adjusted in real time based on the measured values ​​of the carbon monoxide concentration and the nitrogen oxide concentration at the denitrification system inlet, including:

[0056] like Figure 2 As shown, the average carbon monoxide concentration at the inlet of the denitrification system within a preset time is calculated. If the average carbon monoxide concentration is greater than a first threshold, the operating oxygen amount is adjusted according to the current operating conditions based on the relationship curve between the oxygen reduction value and the carbon monoxide concentration;

[0057] like Figure 3 As shown, the average nitrogen oxide concentration at the inlet of the denitration system within a preset time is calculated. If the average nitrogen oxide concentration is greater than the second threshold, the burnout damper opening is increased and the secondary damper opening is decreased.

[0058] More preferably, the method for generating the relationship curve between the oxygen reduction value and the carbon monoxide concentration includes: retrieving the historical data of the carbon monoxide concentration value and the oxygen value at the outlet of the boiler economizer under all working conditions, analyzing and drawing the relationship curve between the oxygen reduction value and the carbon monoxide concentration, and recording it as ΔO2=f(CO). Figure 4 As shown, the horizontal axis is the carbon monoxide concentration value and the vertical axis is the oxygen reduction value.

[0059] Preferably, based on the denitrification system operation data, sample data alignment is performed in combination with the actual sampling delay time to construct a training sample set, including:

[0060] Retrieve the denitrification system operation data, including boiler load, oxygen content at the economizer outlet, total coal quantity, burnout air volume, and nitrogen oxide concentration at the denitrification system inlet;

[0061] The delay time is predicted based on the analysis of factors affecting actual sampling, including the length of the measurement pipeline and the time it takes for nitrogen oxides at the denitrification system inlet to change after changes in operating parameters;

[0062] The retrieved denitrification system operation data is organized into multiple training sample sets according to the delay time.

[0063] Preferably, a prediction model is constructed based on the training sample set, a predicted value of the nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model, and the prediction model is updated in real time, including:

[0064] like Figure 5 As shown, based on the training sample set, the random forest algorithm is used for training to form a prediction model PreModel; Figure 5 In the table, NH3Flow represents ammonia flow rate in kg / h; GasFlow represents flue gas flow rate in Nm 3 / h;

[0065] Real-time data is acquired and the predicted value of nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model PreModel; when the deviation between the predicted value and the actual value exceeds the allowable error, the prediction model PreModel is updated online.

[0066] More preferably, after obtaining the predicted value of the nitrogen oxide concentration at the inlet of the denitrification system through the prediction model, the ammonia flow rate that needs to be adjusted is calculated according to the predicted value of the nitrogen oxide concentration and added to the ammonia flow rate in a feedforward manner. Figure 6 At mark ① in .

[0067] Preferably, the treatment measures after the nitrogen oxide concentration exceeds the discharge level are set in stages, including:

[0068] Compensate ammonia flow rate based on zoned NOx concentration: Based on the fact that the NOx concentration measurement pipelines at the denitration system outlet and each zone outlet are shorter than those at the chimney outlet, resulting in a relatively smaller measurement lag, a small amount of ammonia flow rate is compensated by the difference between the predicted NOx concentration emission value and the set value at the denitration system outlet and each zone outlet;

[0069] Compensate the ammonia flow rate based on the nitrogen oxide concentration at the chimney outlet: The nitrogen oxide concentration at the chimney outlet is a direct indicator of environmental monitoring, and its measurement is seriously delayed. When the nitrogen oxide concentration at the chimney outlet is measured to be excessive, it may actually be seriously excessive. Therefore, when the nitrogen oxide concentration at the chimney outlet is high, it is necessary to respond promptly and increase the ammonia injection volume to quickly reduce the measured value to ensure that the average nitrogen oxide concentration does not exceed the standard.

[0070] According to the changing trend of the predicted value of nitrogen oxide concentration at the inlet of the denitrification system, the above two coefficients are adjusted dynamically in real time, such as Figure 7 As shown in the figure: when the change rate of the predicted value of nitrogen oxide concentration is large, the segmented influence coefficient of ammonia flow rate is increased; when the change rate of the predicted value of nitrogen oxide concentration is small, the segmented influence coefficient of ammonia flow rate is reduced.

[0071] The above treatment measures can improve the response speed of the system, reduce overshoot and undershoot, and ensure that the average value of nitrogen oxides does not exceed the environmental emission standards, such as Figure 8 and Figure 9 shown. Figure 9 The curves from top to bottom are the nitrogen oxide concentration at the reactor inlet, ammonia flow, load, and nitrogen oxide concentration at the chimney outlet (with Figure 8 Except for the inlet NOx curve, which has a range of 0-1000, the other curves have a range of 0-400.

[0072] Specifically, the application of the precise ammonia injection project of a pulverized coal boiler in a power station is used as an example for explanation. The method of this embodiment can be implemented by the following steps:

[0073] 1. Based on the historical data of CO and O2, analyze the relationship between CO content and reduction of O2 content;

[0074] 2. Set the judgment logic based on the nitrogen oxide content emitted by the boiler, and add the O2 bias, burnout damper, and secondary damper bias to the original logic; the O2 bias value is calculated based on the coordinate curve of CO and O2 content, and the opening and closing amplitudes of the burnout damper and secondary damper are set based on the opening experience during commissioning;

[0075] 3. Train the boiler nitrogen oxide concentration emission prediction model and stop training when the error of model training is less than 10%;

[0076] 4. Use the deviation between the predicted value and the set value to dynamically calculate the reducing agent consumption and add it to the control system in a feedforward manner. The implementation plan is shown in Figure 6 Mark ①;

[0077] 5. Apply the emergency treatment measures after over-discharge to the intelligent feedforward. The implementation plan is shown in Figure 7 , add location see Figure 6 Mark ②, the implementation effect diagram is as follows Figure 8 shown.

[0078] In summary, the present invention stabilizes the nitrogen oxide concentration at the boiler denitrification inlet by adjusting the operating oxygen content and burnout air ratio in real time. Data alignment is performed based on the measurement delay time, and a random forest algorithm is used to predict nitrogen oxide emissions over a period of time in real time. The prediction model is continuously updated in real time based on the operating conditions to ensure the accuracy of the prediction. Emergency treatment measures are designed for sudden denitrification over-emissions to improve the system response speed and ensure that the system emission average does not exceed the standard. Engineering verification has shown that the implementation of this control method has improved the control performance of the denitrification system, reduced the amount of ammonia sprayed, lowered the ammonia escape concentration, and increased the automation investment rate of the system.

[0079] Example 2

[0080] This embodiment provides a denitration system inlet nitrogen oxide disturbance and over-emission control system, including:

[0081] a combustion adjustment module configured to adjust the operating oxygen content and the overfire air ratio in real time based on the measured values ​​of the carbon monoxide concentration and the nitrogen oxide concentration at the denitrification system inlet;

[0082] The model training and prediction module is configured to align sample data based on the denitrification system operating data and the actual sampling delay time to construct a training sample set; build a prediction model based on the training sample set, use the prediction model to obtain the predicted value of nitrogen oxide concentration at the denitrification system inlet, and update the prediction model in real time;

[0083] The over-emission control module is configured to set the treatment measures after the nitrogen oxide concentration exceeds the emission limit in sections based on the difference between the average nitrogen oxide concentration at the outlet of each zone and the set value, as well as the difference between the nitrogen oxide concentration at the outlet of the denitrification system and the set value, combined with the predicted nitrogen oxide concentration at the inlet of the denitrification system.

[0084] Example 3

[0085] This embodiment is based on embodiment 1:

[0086] This embodiment provides a computer device including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the method for controlling nitrogen oxide disturbance and excess emission at the inlet of a denitration system according to Embodiment 1. The computer program may be in source code form, object code form, an executable file, or some intermediate form.

[0087] Example 4

[0088] This embodiment is based on embodiment 1:

[0089] This embodiment provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to embodiment 1. The computer program may be in source code form, object code form, executable file, or some intermediate form. The storage medium includes: any entity or device capable of carrying computer program code, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content contained in the storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the storage medium does not include electric carrier signals and telecommunication signals.

[0090] It should be noted that, for the sake of simplicity, the aforementioned method embodiments are described as a series of action combinations. However, those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

Claims

1. A method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system, characterized in that: include: Based on the measured values ​​of carbon monoxide concentration and nitrogen oxide concentration at the denitrification system inlet, the operating oxygen content and burnout air ratio are adjusted in real time; Based on the denitrification system operation data, combined with the actual sampling delay time, sample data is aligned and a training sample set is constructed; A prediction model is built based on the training sample set, and the predicted value of nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model, and the prediction model is updated in real time; Based on the difference between the average nitrogen oxide concentration at the outlet of each zone and the set value, as well as the difference between the nitrogen oxide concentration at the outlet of the denitrification system and the set value, combined with the predicted nitrogen oxide concentration at the inlet of the denitrification system, the treatment measures after the nitrogen oxide concentration exceeds the discharge limit are set in sections.

2. A method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 1, characterized in that: The method of adjusting the operating oxygen amount and the overfire air ratio in real time based on the measured values ​​of the carbon monoxide concentration and the nitrogen oxide concentration at the denitrification system inlet includes: Calculating the average carbon monoxide concentration at the denitration system inlet within a preset time period. If the average carbon monoxide concentration is greater than a first threshold, adjusting the operating oxygen level according to the current operating conditions based on a curve showing the relationship between the oxygen reduction value and the carbon monoxide concentration. The average nitrogen oxide concentration at the inlet of the denitrification system within a preset time is calculated. If the average nitrogen oxide concentration is greater than the second threshold, the burnout damper opening is increased and the secondary damper opening is decreased.

3. A method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 2, characterized in that: The method for generating the relationship curve between the oxygen reduction value and the carbon monoxide concentration includes: The historical data of the carbon monoxide concentration and oxygen value at the boiler economizer outlet under all operating conditions are retrieved, and a curve of the relationship between the oxygen reduction value and the carbon monoxide concentration is analyzed and drawn, where the horizontal axis is the carbon monoxide concentration value and the vertical axis is the oxygen reduction value.

4. The method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 1, characterized in that: The method of aligning sample data based on the denitrification system operation data and combining the actual sampling delay time to construct a training sample set includes: Retrieving denitration system operating data, the denitration system operating data including boiler load, economizer outlet oxygen content, total coal quantity, burnout air volume, and denitration system inlet nitrogen oxide concentration; Analyze and predict the delay time based on actual sampling influencing factors, including the length of the measurement pipeline and the time it takes for nitrogen oxides at the denitrification system inlet to change after operating parameters change; The retrieved denitrification system operation data is organized into multiple training sample sets according to the delay time.

5. The method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 1, characterized in that: The method of constructing a prediction model based on the training sample set, obtaining a predicted value of the nitrogen oxide concentration at the inlet of the denitrification system through the prediction model, and updating the prediction model in real time includes: Based on the training sample set, the random forest algorithm is used for training to form a prediction model; Real-time data is acquired and the predicted value of nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model; when the deviation between the predicted value and the actual value exceeds the allowable error, the prediction model is updated online.

6. The method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 1, characterized in that: After obtaining the predicted value of the nitrogen oxide concentration at the inlet of the denitrification system through the prediction model, the ammonia flow rate that needs to be adjusted is calculated according to the predicted value of the nitrogen oxide concentration, and is adjusted in a feedforward manner.

7. The method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 1, characterized in that: The treatment measures after the nitrogen oxide concentration exceeds the discharge limit are set in sections, including: Compensate the ammonia flow rate according to the zoned nitrogen oxide concentration: Based on the fact that the nitrogen oxide concentration measurement pipelines at the denitrification system outlet and each zone outlet are shorter than the nitrogen oxide concentration measurement pipelines at the chimney outlet and have a relatively smaller measurement lag, the ammonia flow rate is compensated accordingly by the difference between the predicted value and the set value of the nitrogen oxide concentration emission at the denitrification system outlet and each zone outlet.

8. A method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 7, characterized in that: The treatment measures after the nitrogen oxide concentration exceeds the discharge limit are set in sections, including: Compensate the ammonia flow rate according to the nitrogen oxide concentration at the chimney outlet: Based on the relatively large measurement lag of the nitrogen oxide concentration at the chimney outlet, the ammonia flow rate is compensated in time by the difference between the predicted value and the set value of the nitrogen oxide concentration emission at the chimney outlet.

9. A method for controlling nitrogen oxide disturbance and over-emission at the inlet of a denitrification system according to claim 8, characterized in that: The stepwise setting of treatment measures after the nitrogen oxide concentration exceeds the discharge limit also includes: According to the changing trend of the predicted value of nitrogen oxide concentration at the inlet of the denitrification system, the segmented influence coefficient of the ammonia flow is adjusted dynamically in real time: when the changing rate of the predicted value of nitrogen oxide concentration is large, the segmented influence coefficient of the ammonia flow is increased; when the changing rate of the predicted value of nitrogen oxide concentration is small, the segmented influence coefficient of the ammonia flow is reduced.

10. A denitrification system inlet nitrogen oxide disturbance and over-emission control system, characterized in that: include: a combustion adjustment module configured to adjust the operating oxygen content and the overfire air ratio in real time based on the measured values ​​of the carbon monoxide concentration and the nitrogen oxide concentration at the denitrification system inlet; The model training and prediction module is configured to align sample data and construct a training sample set based on the denitrification system operation data and the actual sampling delay time; A prediction model is built based on the training sample set, and the predicted value of nitrogen oxide concentration at the inlet of the denitrification system is obtained through the prediction model, and the prediction model is updated in real time; The over-emission control module is configured to set the treatment measures after the nitrogen oxide concentration exceeds the emission limit in sections based on the difference between the average nitrogen oxide concentration at the outlet of each zone and the set value, as well as the difference between the nitrogen oxide concentration at the outlet of the denitrification system and the set value, combined with the predicted nitrogen oxide concentration at the inlet of the denitrification system.

Images