Partitioned ammonia injection control method, electronic device, and storage medium

Abstract

The application discloses a partition ammonia injection control method, an electronic device and a storage medium, relates to the technical field of flue gas denitration, and comprises the following steps: based on a pre-constructed model, inputting current operation condition parameters of a flue gas denitration device, outputting NO x concentration prediction values of each partition in a flue; according to the NO x concentration prediction values, determining target ammonia injection amounts of each partition, and performing coarse adjustment on ammonia injection of each partition according to the target ammonia injection amounts; dynamically acquiring NO x concentration monitoring values of each partition, and according to the NO x concentration monitoring values, determining NO x non-uniformity coefficients of each partition; when the NO x non-uniformity coefficient is greater than a preset threshold value, according to the NO x concentration monitoring values, determining feedback ammonia injection amounts corresponding to each partition, and performing fine adjustment on ammonia injection of each partition according to the feedback ammonia injection amounts. The application effectively improves the ammonia injection accuracy under complex blending combustion conditions.

Description

Technical Field

[0001] This application relates to the field of flue gas denitrification technology, and in particular to a method for zoned ammonia injection control, electronic equipment, and storage medium. Background Technology

[0002] SCR (Selective Catalytic Reduction) flue gas denitrification systems in coal-fired power plants remove nitrogen oxides (NOx) from the flue gas by injecting ammonia as a reducing agent into the flue gas duct, under the action of a catalyst. x Reducing NO to nitrogen and water is a key technical means to achieve compliant emissions of flue gas pollutants. With the increasing proportion of high-moisture Indonesian coal and dried sludge, among other multi-fuel combinations, in thermal power units, NO levels in different regions of the SCR reactor inlet section are rising. x With significantly increased concentration distribution differences, zoned ammonia injection control technology has emerged to improve denitrification efficiency and control ammonia slip. This involves dividing the ammonia injection grid into multiple independent control zones, with each zone independently adjusting the ammonia injection rate to match the inlet NO₂ levels. x Spatial distribution.

[0003] However, the current method of zoned ammonia injection control still has a low accuracy rate in ammonia injection under mixed coal combustion conditions.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a zoned ammonia injection control method, electronic equipment, and storage medium to improve the accuracy of ammonia injection under complex co-firing conditions.

[0006] To achieve the above objectives, this application proposes a zoned ammonia injection control method, comprising the following steps: Based on a pre-built model, the current operating parameters of the flue gas denitrification device are input, and the NO values ​​of each zone in the flue gas duct are output. x Concentration prediction; According to NO x The concentration prediction value is used to determine the target ammonia injection amount for each zone, and the ammonia injection amount for each zone is roughly adjusted according to the target ammonia injection amount. Dynamically obtain the NO of each partition x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Inhomogeneity coefficient; When NO x When the non-uniformity coefficient is greater than a preset threshold, according to NO x The concentration monitoring value is used to determine the corresponding ammonia injection amount for each zone, and the ammonia injection for each zone is fine-tuned based on the feedback ammonia injection amount.

[0007] In one embodiment, according to NO x The steps for predicting concentration and determining the target ammonia injection amount for each zone include: Obtain the flue gas volume flow rate for each zone; According to NO x Based on the predicted concentration and flue gas volumetric flow rate, calculate the NO concentration for each zone. x Total amount; According to NO x The target ammonia injection amount for each zone is obtained by multiplying the total amount by the preset ammonia-nitrogen molar ratio.

[0008] In one embodiment, according to NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x The steps for determining the non-uniformity coefficient include: According to NO x Concentration monitoring values ​​to determine NO in the flue x Average concentration; According to NO x Concentration monitoring value and NO x The first deviation of the average concentration determines the NO concentration in each zone. x Concentration standard deviation; According to NO x Concentration standard deviation and NO x The ratio of the average concentrations yields the NO concentration for each zone. x Non-uniformity coefficient.

[0009] In one embodiment, according to NO x The steps for determining the corresponding ammonia injection rate for a given zone based on concentration monitoring values ​​include: Confirm NO x The second deviation between the concentration monitoring value and the preset outlet concentration target value; The proportional adjustment amount for each zone is determined based on the second deviation and the preset proportional coefficient. The integral adjustment amount of the partition is determined based on the cumulative amount of the second deviation and the preset integral coefficient. The feedback ammonia injection amount for each zone is obtained by summing the proportional and integral adjustment amounts.

[0010] In one embodiment, based on a pre-built model, the current operating parameters of the flue gas denitrification device are input, and the NO levels of each zone in the flue are output. x Before the concentration prediction step, the following steps are also included: Inlet and outlet measuring points are set up in each zone; Obtain the entry NO of each partition through the entry measurement point. x Concentration and flue gas flow parameters were measured, and the outlet NO of each zone was obtained through outlet measuring points.x concentration; Enter NO x Concentration, flue gas flow parameters, outlet NO x The concentration and the amount of ammonia injected in each zone were correlated to obtain the dataset used to build the model.

[0011] In one embodiment, the NO of each partition is dynamically obtained. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Following the step of determining the non-uniformity coefficient, the following steps are also included: When NO x When the non-uniformity coefficient is less than or equal to the preset threshold, obtain the current ammonia injection amount and flue gas volume flow rate of each zone; According to NO x Predicted concentration, NO x Based on the concentration monitoring value, the current ammonia injection rate, and the flue gas volume flow rate, determine the theoretical ammonia slip and the theoretical total ammonia slip for each zone; Based on the actual total ammonia slip and the theoretical total ammonia slip, the theoretical ammonia slip is corrected using a proportional correction method to obtain the actual ammonia slip for each zone. When the actual ammonia escape amount exceeds the preset zone escape threshold, the ammonia injection amount of the zone is reduced according to the preset reduction ratio.

[0012] In one embodiment, the method further includes: Dynamically acquire the operating monitoring parameters of the flue gas denitrification device, including load reduction rate, main fuel trip status, furnace negative pressure fluctuation amplitude, flame loss signal, average flame intensity, flue gas oxygen content fluctuation amplitude, and furnace temperature drop rate. When the operating monitoring parameters meet the preset abnormal conditions, the current ammonia injection amount of each zone is adjusted according to the preset fixed molar ratio.

[0013] In one embodiment, the method further includes: When the flue gas denitrification unit meets the preset stable operating conditions, the valve opening of each zone is adjusted from the current opening to the first opening and maintained for a preset duration to obtain the first outlet NO. x Concentration change rate, where the first opening degree is greater than the current opening degree; Adjust the valve opening to the second opening degree and maintain it for a preset time to obtain the second outlet NO. x Concentration change rate, wherein the second opening degree is less than the current opening degree; According to the first exit NO x Concentration change rate and second outlet NO xThe concentration change rate is used to determine the actual flow coefficient of the valves in each zone, and the correspondence between the valve opening and the ammonia injection amount is updated based on the actual flow coefficient of the valves. The correspondence is used for the step of coarse adjustment of ammonia injection in each zone based on the target ammonia injection amount, and / or for the step of fine adjustment of ammonia injection in each zone based on the feedback ammonia injection amount.

[0014] To achieve the above objectives, this application proposes an electronic device, including a memory and a processor, the processor being capable of executing the steps of the zoned ammonia injection control method described above.

[0015] Furthermore, to achieve the above objectives, this application proposes a readable storage medium, which is a computer-readable storage medium storing a program for implementing zoned ammonia injection control. The program for implementing zoned ammonia injection control is executed by a processor to implement the steps of the zoned ammonia injection control method as described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: by inputting the current operating parameters of the flue gas denitrification device based on a pre-built model, the NO in each zone of the flue gas duct is output. x Predicted NO concentration values ​​can be obtained before ammonia injection adjustments. x The spatial distribution of NO provides accurate feedforward data for subsequent zoned differentiated ammonia injection; based on NO x The concentration prediction value is used to determine the target ammonia injection rate for each zone. Based on the target ammonia injection rate, the ammonia injection rate for each zone is coarsely adjusted to ensure that the ammonia injection rate for each zone corresponds to the NO in the corresponding area. x The total quantity was initially matched in spatial distribution, enabling a rapid response to changes in operating conditions; NO was dynamically acquired for each zone. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Inhomogeneity coefficient, NO x The uniformity of distribution is quantified into a measurable control indicator, providing a clear basis for determining whether to initiate fine-tuning and avoiding ineffective adjustments when uniformity has already met the target; when NO x When the non-uniformity coefficient is greater than a preset threshold, according to NO x Based on the concentration monitoring values, the corresponding ammonia injection rate for each zone is determined, and the ammonia injection rate for each zone is fine-tuned according to the feedback ammonia injection rate, so that NO... x Zones with high concentrations of nitrogen oxides and nitrogen oxides were compensated by ammonia and NO injection. x The ammonia injection rate was reduced accordingly in zones with low concentrations, forming an independent closed-loop correction for each zone and improving the overall NO concentration at the outlet section. x Concentration uniformity. This application achieves uniformity of ammonia injection rate and inlet NO in each zone under mixed coal combustion conditions. xThe adaptive and precise matching of the distribution effectively improves the accuracy of ammonia injection under complex co-firing conditions. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the zoned ammonia injection control method according to an embodiment of this application. Figure 1 ; Figure 2 This is a schematic diagram of the zoning ammonia injection control method according to an embodiment of this application; Figure 3 This is a flowchart illustrating the zoned ammonia injection control method according to an embodiment of this application. Figure 2 ; Figure 4 This is a flowchart illustrating the zoned ammonia injection control method according to an embodiment of this application. Figure 3 ; Figure 5 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the zoned ammonia injection control method of this application embodiment.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] The term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of this application. The phrase "in an embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that excludes other embodiments.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] Currently, the accuracy of ammonia injection control methods in mixed coal combustion conditions remains low. On the one hand, existing control methods mostly rely on macroscopic parameters such as unit load and total air volume as the basis for control, lacking the ability to respond to changes in fuel composition. When the fuel structure changes, the inlet NO... x The actual distribution of the concentration field deviates significantly from the preset matching relationship of the control system, leading to a significant discrepancy between the ammonia injection and NO concentration. x Distribution mismatch. On the other hand, traditional methods typically rely on outlet NO after ammonia injection conditioning. x Correcting feedback signals from single or a small number of measuring points cannot fully reflect the NO levels in each zone of the outlet section. x Concentration distribution makes it difficult to determine the inlet NO concentration. x The system makes timely and accurate adaptive adjustments to changes in the field. The above factors lead to variations in ammonia injection rates and corresponding NO levels in different zones under blended coal combustion conditions. x Low overall matching degree, export NO x The high non-uniformity coefficient and prominent local ammonia escape problem limit the system's economic efficiency and environmental compliance.

[0025] This application provides a solution that, based on a pre-built model, inputs the current operating parameters of a flue gas denitrification device and outputs the NO levels for each zone in the flue gas duct. x Predicted NO concentration values ​​can be obtained before ammonia injection adjustments. x The spatial distribution of NO provides accurate feedforward data for subsequent zoned differentiated ammonia injection; based on NO x The concentration prediction value is used to determine the target ammonia injection rate for each zone. Based on the target ammonia injection rate, the ammonia injection rate for each zone is coarsely adjusted to ensure that the ammonia injection rate for each zone corresponds to the NO in the corresponding area. x The total quantity was initially matched in spatial distribution, enabling a rapid response to changes in operating conditions; NO was dynamically acquired for each zone. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Inhomogeneity coefficient, NO x The uniformity of distribution is quantified into a measurable control indicator, providing a clear basis for determining whether to initiate fine-tuning and avoiding ineffective adjustments when uniformity has already met the target; when NO x When the non-uniformity coefficient is greater than a preset threshold, according to NO x Based on the concentration monitoring values, the corresponding ammonia injection rate for each zone is determined, and the ammonia injection rate for each zone is fine-tuned according to the feedback ammonia injection rate, so that NO... x Zones with high concentrations of nitrogen oxides and nitrogen oxides were compensated by ammonia and NO injection. x The ammonia injection rate was reduced accordingly in zones with low concentrations, forming an independent closed-loop correction for each zone and improving the overall NO concentration at the outlet section. xConcentration uniformity. This application achieves uniformity of ammonia injection rate and inlet NO in each zone under mixed coal combustion conditions. x The adaptive and precise matching of the distribution effectively improves the accuracy of ammonia injection under complex co-firing conditions.

[0026] Based on this, the first aspect of the embodiments of this application provides a zoned ammonia injection control method, referring to... Figure 1 This includes the following steps: Step S10: Based on the pre-built model, input the current operating parameters of the flue gas denitrification device, and output the NO levels of each zone in the flue gas duct. x Concentration prediction; Optionally, the flue gas denitrification device refers to the selective catalytic reduction (SCR) denitrification system in a coal-fired power plant, which injects ammonia reducing agent into the flue gas to remove nitrogen oxides (NOx) in the flue gas under the action of a catalyst. x The process reduces the nitrogen and water content. The SCR system includes the boiler furnace, economizer outlet, SCR reactor inlet flue, ammonia injection grid, SCR catalyst bed, SCR reactor outlet flue, air preheater, inlet grid measuring point group, outlet grid measuring point group, ammonia slip monitoring points, zoned ammonia injection regulating valve group, liquid ammonia evaporator, ammonia injection optimization controller, and DCS (Distributed Control System). A zone refers to multiple independent control areas divided spatially from the SCR reactor inlet flue cross-section. Each zone corresponds to an independent ammonia injection branch pipe and regulating valve of the ammonia injection grid. The zones are arranged in an array along the width and height of the flue cross-section, covering the entire flue cross-section, as shown in the reference diagram. Figure 2 It can be set to have 12 partitions in total, consisting of four columns (A1~A4) and three rows (B1~B3).

[0027] Optionally, the current operating condition parameters refer to multi-dimensional parameters characterizing the current operating status of the flue gas denitrification device. In the embodiments of this application, the current operating condition parameters include, but are not limited to: unit load, fuel blending ratio, received nitrogen content of coal, total air volume, average furnace temperature, and burner tilt angle. The unit load reflects the boiler's evaporation and power generation, measured in MW. The fuel blending ratio refers to the percentage of the total mass of dried sludge fed into the furnace relative to the total mass of all fuel fed into the furnace. The measurement standard is the real-time weighing volume of the feed conveyor belt. The instantaneous mass flow rate is uploaded to the DCS by the raw coal belt scale and the sludge-dried feed screw scale in the coal conveying system. The DCS calculates the blending ratio based on the real-time flow rate, with a value range of 0-50%, updated every second. The nitrogen content of the received coal is automatically sampled online using an online CHNS elemental analyzer (using high-temperature combustion infrared-thermal conductivity method) on the raw coal and sludge mixed feed conveyor belt. Sampling is performed automatically every 2 hours. The instrument first measures the air-dried nitrogen content N_ad, and then, combined with the simultaneously measured received total moisture M_ar and air-dried moisture M_ad, calculates the received nitrogen content N_ar using the national standard conversion formula: N_ad × (100 - M_ar) / (100 - ... M_ad), after conversion, a set of real-time N_ar values ​​are updated every 2 hours and sent to the prediction model. If the analyzer temporarily fails, the average value of historical measured N_ar for the same coal type and blending ratio in the past 7 days is retrieved as a temporary substitute input to ensure uninterrupted operation of the model; the total air volume is obtained by the air volume measurement device at the outlet of the blower; the average furnace temperature is determined by the average value of the temperature measurement points in each layer of the furnace; the burner tilt angle reflects the tilt position of the burner and is obtained by the position feedback signal of the burner actuator.

[0028] Optionally, the model is a NO model constructed based on a neural network algorithm. x A concentration field prediction model is used to predict the NO concentration in each zone of the SCR reactor inlet section based on the current operating parameters of the flue gas denitrification unit. x Concentration distribution. This model can employ a BP (Back Propagation) neural network structure, comprising an input layer, hidden layers, and an output layer. The input layer receives current operating parameters, while the hidden layers perform nonlinear mapping and feature extraction on the input data. Two layers are used, each with 10 neurons, and the activation function is ReLU (Rectified Linear Unit) to effectively avoid the gradient vanishing problem under conditions of large flue gas fluctuations, enabling the model to learn stably in complex nonlinear feature spaces. The output layer outputs the NO concentration for each region. xConcentration prediction values. Based on 12 consecutive months of full-condition measured data from the unit, a total sample was constructed and divided into three parts: training set, validation set, and test set. The Levenberg-Marquardt (LM) algorithm was used as the backpropagation training function to train the model, and the accuracy of the trained model was verified using the test set to achieve NO. x Accuracy requirements for concentration prediction.

[0029] In one feasible embodiment, the aforementioned current operating condition parameters are collected in real time and input into the pre-built model. The model calculates and processes the input parameters, outputting the NO values ​​for each zone at the SCR reactor inlet section. x Predicted concentration values. This step allows for advance knowledge of the spatial distribution of NOx at the inlet, providing accurate feedforward data for subsequent zoned differentiated ammonia injection and overcoming the control lag problem caused by traditional methods relying solely on historical data or macroscopic average parameters.

[0030] It should be noted that the number of partitions can be determined according to the actual situation, such as the size of the flue cross-section. For example, it can be 12, 8 or 16. This application embodiment does not limit this.

[0031] It should be noted that the neural network model can be replaced with the support vector machine model to adapt to scenarios with small sample datasets, and this application does not limit this.

[0032] Step S20, according to NO x The concentration prediction value is used to determine the target ammonia injection amount for each zone, and the ammonia injection amount for each zone is roughly adjusted according to the target ammonia injection amount. Optionally, the target ammonia injection rate refers to the current predicted inlet NO. x Under the given concentration distribution conditions, the theoretical ammonia reduction dose that should be injected into each zone is given in kg / h. The process for determining the target ammonia injection dose takes into account the NO carried by the flue gas in each zone. x The difference in total volume causes the amount of ammonia injected to differ from the actual NO in each zone. x The loads are matched.

[0033] Optionally, after determining the target ammonia injection rate for each zone, coarse ammonia injection is performed on each zone based on the target ammonia injection rate and the preset correspondence between valve opening and ammonia injection rate. The preset correspondence between valve opening and ammonia injection rate refers to a valve characteristic mapping table stored within the control system. This mapping table records the actual ammonia injection rate corresponding to different openings of the regulating valves in each zone. This correspondence can be obtained through valve flow characteristic tests before system commissioning, or it can be corrected and updated online through subsequent valve flow self-calibration steps.

[0034] In one feasible embodiment, based on the partition NO output in step S10x The concentration prediction value is used to calculate the target ammonia injection rate for each zone. Then, based on the preset correspondence between valve opening and ammonia injection rate, the target ammonia injection rate for each zone is converted into a corresponding valve opening command. The regulating valves of each zone are then controlled to execute this opening command, completing the coarse adjustment of ammonia injection for each zone. This step ensures that the ammonia injection rate of each zone corresponds to the NO in its corresponding area. x The total amount has been initially matched in spatial distribution, enabling a rapid response to changes in operating conditions and effectively reducing the export NOx. x The overall deviation range of the distribution provides a good starting point for subsequent fine-tuning.

[0035] Step S30: Dynamically obtain the NO of each partition. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Inhomogeneity coefficient; Optionally, NO x Concentration monitoring values ​​refer to the NO concentration values ​​measured by placing them in various zones at the outlet section of the SCR reactor. x NO measured in real time by the measuring instrument x Concentration data, in mg / m³ 3 The division method of each zone at the exit section is consistent with that at the inlet section, forming a spatially corresponding grid measurement system.

[0036] Optionally, NO x The non-uniformity coefficient is used to quantify the NO in each zone of the outlet section. x A dimensionless index of the uniformity of NO concentration distribution; the smaller the value, the better. x The more uniform the concentration distribution across the cross section, the smaller the concentration difference between different zones; conversely, a more uneven distribution indicates a greater concentration difference between different zones.

[0037] In one feasible embodiment, after completing the ammonia injection coarse adjustment in step S20, the NO levels in each zone of the SCR reactor outlet section are collected in real time. x Concentration monitoring values, and based on NO in all the aforementioned zones. x Concentration monitoring values, calculate NO at the outlet section x The average concentration and the degree of deviation of each zone from this average are used to determine the NO concentration. x Non-uniformity coefficient. This step will reduce the export NO. x The uniformity of distribution is quantified into an evaluable control index, providing a clear basis for determining whether to initiate fine-tuning, avoiding ineffective adjustments when uniformity has already met the target, and saving control resources and ammonia reducing agent consumption.

[0038] Step S40, when NO xWhen the non-uniformity coefficient is greater than a preset threshold, according to NO x The concentration monitoring value is used to determine the corresponding ammonia injection amount for each zone, and the ammonia injection for each zone is fine-tuned based on the feedback ammonia injection amount.

[0039] Optionally, the preset threshold is a pre-set NO. x The upper limit of the non-uniformity coefficient: when the calculated non-uniformity coefficient exceeds this threshold, it indicates that the export NO x The uniformity of concentration distribution does not meet the requirements, necessitating fine-tuning. This preset threshold can be set according to the emission standards and actual operating requirements of the flue gas denitrification device, for example, 15%.

[0040] Optionally, the feedback ammonia injection volume refers to the amount based on the export NO. x The deviation between the concentration monitoring value and the target value is used to calculate the correction value of the ammonia injection amount for each zone. This feedback ammonia injection amount is used to compensate for the target ammonia injection amount determined in step S20, so that the actual ammonia injection amount of each zone can be adaptively adjusted according to the actual measurement results at the outlet.

[0041] Optionally, after determining the feedback ammonia injection quantity for each zone, the ammonia injection quantity for each zone is finely adjusted based on the correspondence between the feedback ammonia injection quantity and the valve opening and the ammonia injection quantity.

[0042] In one feasible embodiment, when the NO calculated in step S30 x When the unevenness coefficient is greater than a preset threshold, the system determines the partition's NO value based on the threshold value. x The deviation between the concentration monitoring value and the outlet concentration target value is used to calculate the feedback ammonia injection rate for each zone. Based on the preset correspondence between valve opening and ammonia injection rate, the feedback ammonia injection rate for each zone is converted into a corresponding valve opening correction command. The regulating valves of each zone are then controlled to execute the correction command based on the current opening of that zone, thus completing the fine-tuning of ammonia injection for each zone. Through this step, NO... x Zones with high concentrations (i.e., zones with insufficient ammonia injection) receive compensatory incremental ammonia injection correction, NO x Zones with low concentrations (i.e., zones with excessive ammonia injection) receive reduced ammonia injection correction, forming independent closed-loop corrections for each zone. This step effectively improves the overall NO concentration at the outlet section. x The uniformity of concentration reduces the risk of local over-spraying and under-spraying, while avoiding the problems of increased ammonia escape due to excessive ammonia spraying and substandard denitrification efficiency due to insufficient ammonia spraying.

[0043] Optionally, the ammonia injection fine-tuning is performed cyclically according to a preset adjustment cycle. In this embodiment, the adjustment cycle is set to 30 seconds, based on the following: Instrument hysteresis constraint: In this embodiment, the outlet NO xThe online analyzer has a delay of approximately 28 seconds from sampling to outputting a stable reading. Simultaneously, there is a pure transmission lag of 22 to 30 seconds in the flue gas flow from the ammonia injection grid to the outlet measuring point. If the adjustment cycle is less than or equal to this lag time, it will lead to frequent valve vibration and repeated over-adjustment of ammonia injection. Control stability constraints: If the adjustment cycle is less than 30 seconds, the flue gas generated by the previous valve adjustment has not yet reached the outlet measuring point, and the sampled data cannot accurately reflect the actual effect of this adjustment, easily causing frequent valve vibration, repeated over-adjustment of ammonia injection, and system oscillation. Response speed balance: If the adjustment cycle is greater than 30 seconds, the time for eliminating a single deviation is lengthened, and the system's ability to track changes in operating conditions decreases. Field verification: Comparative tests with different adjustment cycles were conducted on the 330MW unit described in this application embodiment. Under a 30-second cycle, the single-zone NO... x The average elimination time of concentration deviation is ≤3min, which is better than the 60s long cycle scheme, and there is no problem of frequent valve operation due to the short cycle.

[0044] It should be noted that the adjustment cycle can also be adaptively adjusted according to the specific flue gas velocity, analyzer response characteristics, etc., and the embodiments of this application do not limit this.

[0045] In one feasible implementation, refer to Figure 3 In step S10, based on the pre-built model, the current operating parameters of the flue gas denitrification device are input, and the NO levels of each zone in the flue are output. x Before the concentration prediction step, steps A10 to A30 are also included: Step A10: Set up entry and exit measurement points in each zone; Optionally, the inlet measuring point refers to the gas sampling point and flow parameter measurement point arranged at the cross-section of the inlet flue of the SCR reactor, used to obtain NO at the inlet of each zone. x Concentration and flue gas flow parameters. The outlet sampling point refers to the gas sampling point located at the cross-section of the SCR reactor outlet flue, used to obtain NO concentration at the outlet of each zone. x concentration.

[0046] In one feasible embodiment, grid measuring points are arranged at the inlet and outlet flue sections of the SCR reactor, respectively. The grid division method of the inlet and outlet flue sections is consistent, dividing the flue into multiple equal grid regions along the width and height directions. Each grid region corresponds to a partition. An inlet measuring point and an outlet measuring point are set at the geometric center of each partition, forming a one-to-one spatial mapping relationship between the inlet and outlet measuring points. This ensures the spatial comparability of inlet-side and outlet-side measurement data, providing a data foundation for subsequent calculations of ammonia slip in each partition.

[0047] Step A20: Obtain the entry NO of each partition through the entry measurement point. x Concentration and flue gas flow parameters were measured, and the outlet NO of each zone was obtained through outlet measuring points. x concentration; Optionally, the flue gas flow parameters include the flue gas velocity and temperature at the inlet of each zone, used to calculate the flue gas volumetric flow rate of each zone. Inlet NO x Concentration and export NO x Concentrations are all expressed in mg / m³ 3 Units.

[0048] In one feasible embodiment, under multiple preset operating conditions of the unit, the inlet NO of each zone is synchronously collected through the inlet measuring point of each zone. x Concentration, flue gas velocity, and flue gas temperature, while simultaneously collecting the outlet NO of the zone through the corresponding outlet measuring point. x Concentration. Based on 12 consecutive months of full-condition measured data from the unit, a total sample of 14,680 sets was constructed. The preset operating conditions covered different combinations of unit loads and different fuel blending ratios, such as three typical load types (330MW, 258MW, and 165MW) and four fuel types (pure coal, 20% blending, 35% blending, and 50% blending), totaling 12 benchmark operating conditions. The sampling synchronization deviation at each measuring point was controlled within a preset range (e.g., ≤2s) to ensure that the inlet and outlet data collected at the same time corresponded in time. This step, through simultaneous sampling across multiple operating conditions and zones, obtained raw data covering the entire operating range of the unit, providing sufficient training samples for building the predictive model.

[0049] Step A30, enter NO x Concentration, flue gas flow parameters, outlet NO x The concentration and the amount of ammonia injected in each zone were correlated to obtain the dataset used to build the model.

[0050] Optionally, a dataset refers to a collection of multiple sets of sample data, each set of sample data including entry points NO collected at the same time, in the same partition, and under the same operating conditions. x Concentration, flue gas velocity, flue gas temperature, outlet NO x The concentration and the current ammonia injection rate in the zone are interconnected and together constitute a complete modeling sample.

[0051] In one feasible embodiment, for each partition under each set of preset working conditions, the entry point NO collected under that working condition is... x Concentration, flue gas flow parameters (including flue gas velocity and flue gas temperature), outlet NO xThe concentration and the synchronously recorded current ammonia injection rate of the corresponding partition are correlated to form a set of sample data. Sample data from all partitions under all preset operating conditions are aggregated to form a dataset. This dataset is used to train a pre-built model: the current operating condition parameters in the dataset (such as unit load, fuel blending ratio, received nitrogen content of coal, total air volume, average furnace temperature, burner angle, etc.) are used as model inputs, and the corresponding partition's inlet NO... x Concentration is used as the model output label. The model is trained through supervised learning, enabling it to learn and memorize parameters from operating conditions to the inlet NO. x Mapping relationship of concentration spatial distribution.

[0052] In one feasible embodiment, the dataset is further divided into a training set (e.g., 70%), a validation set (e.g., 20%), and a test set (e.g., 10%), used for model training, parameter tuning, and accuracy verification, respectively. All input and output parameters are mapped to the [0,1] interval using the Min-Max normalization method to eliminate the influence of differences in the units of different parameters on model training. Subsequently, the LM algorithm is used as the backpropagation training function, which has fast convergence speed and high accuracy when processing medium-sized neural networks. At the same time, a momentum factor is added to optimize the weight update process to avoid the training process getting trapped in local minima. The convergence criterion is as follows: Global Mean Squared Error (MSE) ≤ 1 × 10⁻⁶ -4 The maximum number of iterations was set to 2000. The overfitting termination threshold was set to 20 consecutive iterations where the validation set error showed no decrease. Training was automatically terminated when the validation set error failed to improve for 20 consecutive iterations to prevent overfitting. The prediction accuracy of the trained model was then validated using the test set. The validation results were: Single partition NO x The relative average prediction error is ≤6.8%, and the overall average prediction error for all 12 zones is ≤5.2%. Compared with the traditional BP neural network model that uses only load and air volume as dual input parameters (with an average error of 32.1% on the same test set), the prediction accuracy of the model in this embodiment is improved by more than 30%.

[0053] Through steps A10 to A30, this embodiment of the application constructs a modeling dataset covering the entire operating range of the flue gas denitrification device, enabling the model to accurately learn the operating parameters and NO under the condition of blended coal combustion. x The complex nonlinear relationship between concentration spatial distributions provides a reliable data foundation for precise ammonia injection in different zones.

[0054] In one feasible implementation, in step S20, according to NO x The steps for predicting concentration values ​​and determining the target ammonia injection amount for each zone include steps S21 to S23: Step S21: Obtain the flue gas volume flow rate for each zone; Optionally, flue gas volumetric flow rate refers to the volume of flue gas flowing through each section cross-section per unit time, with units of m³. 3 / h. The method of obtaining the data is as follows: based on the flue gas velocity and temperature obtained through the inlet measuring point in step A20, combined with the cross-sectional area of ​​the zone, the measured operating flow rate is converted into the dry flue gas flow rate under standard conditions according to the gas state equation.

[0055] Step S22, according to NO x Based on the predicted concentration and flue gas volumetric flow rate, calculate the NO concentration for each zone. x Total amount; Optionally, the NO of each partition x Total amount refers to the amount of NO flowing through each section of the partition per unit time. x Mass, in units of kg / h.

[0056] Step S23, according to NO x The target ammonia injection amount for each zone is obtained by multiplying the total amount by the preset ammonia-nitrogen molar ratio.

[0057] Optionally, the ammonia-nitrogen molar ratio refers to the ratio of ammonia (NH3) to NO participating in the denitrification reaction. x The ratio of the number of moles is generally in the range of 0.8 to 1.0 and can be dynamically adjusted according to the working conditions.

[0058] In one feasible embodiment, the target ammonia injection amount q for each zone NH3,i Calculate using the following formula: q NH3,i = α × C in,i × η0× V gas,i Where, q NH3,i α represents the target ammonia injection amount for the i-th partition; α represents the ammonia-nitrogen molar ratio; C in,i For the entry point NO of the i-th partition x Predicted concentration value; η0 is the design denitrification efficiency; V gas,i Let be the flue gas volumetric flow rate of the i-th zone. From this formula, it can be seen that the target ammonia injection rate for each zone depends on the NO content of that zone. x Predicted NO concentration and flue gas volumetric flow rate. Due to different inlet NO levels in different zones... x Concentration and flue gas flow rate may vary significantly, therefore the target ammonia injection rate calculated for each zone will also differ, achieving a balance between ammonia injection rate and inlet NO. x The precise matching of spatial distribution differences avoids the problems caused by traditional methods that use the same ammonia injection rate or fixed ratio for all zones, resulting in inconsistent ammonia and NO injection. x Distribution mismatch problem.

[0059] Optionally, the target ammonia injection rate for a single zone can be set to 30-120% of the rated ammonia injection rate. The rated ammonia injection rate refers to the maximum ammonia injection rate corresponding to the zone's regulating valve when it is fully open under design conditions. This ensures that the zone maintains a minimum ammonia injection rate under all circumstances to prevent substandard denitrification efficiency, while allowing the zone to inject ammonia beyond the rated value when necessary to meet local high NO levels. x The required concentration of denitrification ensures that the ammonia injection rate in each zone always operates within a safe and effective range.

[0060] In one feasible implementation, in step S30, according to NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x The steps for determining the non-uniformity coefficient include steps S31 to S33: Step S31, according to NO x Concentration monitoring values ​​to determine NO in the flue x Average concentration; Optionally, NO x The average concentration refers to the NO concentration in all zones of the outlet section. x The arithmetic mean of the concentration monitoring values ​​reflects the overall NO concentration at the outlet section. x Concentration level, in mg / m³ 3 .

[0061] Step S32, according to NO x Concentration monitoring value and NO x The first deviation of the average concentration determines the NO concentration in each zone. x Concentration standard deviation; Optionally, the first deviation refers to the NO of each partition. x Concentration monitoring value relative to NO x The degree of deviation from the average concentration; NO x Concentration standard deviation is a dispersion index obtained by comprehensively statistically analyzing all deviations, reflecting the NO concentration in all zones of the outlet section. x Concentration monitoring value relative to NO x The overall dispersion of the average concentration.

[0062] Step S33, according to NO x Concentration standard deviation and NO x The ratio of the average concentrations yields the NO concentration for each zone. x Non-uniformity coefficient.

[0063] In one feasible embodiment, NO x Inhomogeneity coefficient C v Calculate using the following formula: C v = σ / NOx × 100% Among them, C v For export NO x Concentration non-uniformity coefficient, expressed as a percentage; σ represents the NO concentration at each measuring point at the outlet. x Standard deviation of concentration; NOx For export NO x The concentration arithmetic mean. Wherein, σ is calculated as follows: σ = sqrt( 1 / (n-1) × Σ(C i NOx ) 2 ) Among them, C i NO for the i-th measurement point x Concentration monitoring value; n is the total number of monitoring points (in this embodiment, n=12, consistent with the number of zones). This ratio C v Normalizing the absolute dispersion relative to the average concentration level eliminates the influence of factors such as unit load changes on the absolute value of the dispersion caused by fluctuations in the average concentration. This makes the non-uniformity coefficients calculated under different operating conditions comparable, facilitating the setting of a unified control threshold. v The smaller the value, the lower the NO value between the export zones. x The closer the concentrations, the more uniform the distribution; C v The larger the value, the higher the NO value between the export zones. x The greater the concentration difference, the more uneven the distribution.

[0064] In one feasible implementation, in step S40, according to NO x The steps for determining the corresponding ammonia injection amount for a given zone based on concentration monitoring values ​​include steps S41 to S44: Step S41, determine NO x The second deviation between the concentration monitoring value and the preset outlet concentration target value; Optionally, the preset outlet concentration target value refers to the NO concentration at the outlet of the SCR reactor. x The concentration control target is set in mg / m³. 3 It can be set according to air pollutant emission standards and the operational requirements of flue gas denitrification devices. The second deviation refers to the NO in each zone. x The difference between the concentration monitoring value and the preset outlet concentration target value.

[0065] Step S42: Determine the proportional adjustment amount of the partition based on the second deviation and the preset proportional coefficient; Optionally, the proportional adjustment amount is a correction value for the amount of ammonia injected, calculated proportionally based on the magnitude of the deviation in the current cycle, used to provide an immediate response to the degree to which the zone deviates from the target value in the current cycle.

[0066] Step S43: Determine the integral adjustment amount of the partition based on the cumulative amount of the second deviation and the preset integral coefficient. Optionally, the cumulative amount of the second deviation refers to the cumulative value of the deviation over multiple consecutive adjustment cycles in this zone; the integral adjustment amount is the ammonia injection correction value calculated based on the cumulative amount of deviation, used to eliminate long-term minor deviations.

[0067] Step S44: Based on the sum of the proportional adjustment amount and the integral adjustment amount, the feedback ammonia injection amount for the zone is obtained.

[0068] In one feasible embodiment, the feedback ammonia injection rate Δq for each zone i Calculated according to the following PI (Proportional-Integral) adjustment formula: Δq i = K p × (C) out,i out ) + K i × ∫( C out,i out ) dt Where, Δq i This is the correction value for the ammonia injection rate in the i-th partition, i.e., the feedback ammonia injection rate. A positive value increases ammonia injection, and a negative value decreases ammonia injection; K p K is the preset proportionality coefficient. i C is the preset integral coefficient; out,i For the i-th partition, the export NO x Concentration monitoring value; out The preset target value for outlet concentration; ∫( C out,i out ) dt is the exit NO of the i-th partition. x The cumulative deviation between the concentration monitoring value and the outlet concentration target value over time reflects the cumulative effect of the deviation in this zone over multiple consecutive adjustment cycles.

[0069] Among them, the proportional adjustment amount K p × (C) out,i outThe correction amount is given proportionally based on the magnitude of the deviation in the current cycle; the larger the deviation, the stronger the correction, which governs the dynamic response speed of the system; integral adjustment amount K i × ∫( C out,i out The dt parameter corrects for the cumulative amount of deviation, eliminating persistent small deviations (steady-state errors) and dominating the steady-state accuracy of the system. x Measured concentration C out,i Higher than the target value out The partition (positive deviation) obtains positive feedback on the ammonia injection amount (increases ammonia injection), NO x The zones with measured concentration values ​​lower than the target value (negative deviation) receive negative feedback ammonia injection amounts (reduced ammonia injection). The feedback ammonia injection amounts for each zone are independent and do not affect each other.

[0070] Optionally, K p K i The value can be adjusted according to different operating conditions. Under steady-state conditions, a combined tuning method is adopted, which combines theoretical initial calculation using the Ziegler-Nichols (ZN) critical proportionality method with trial and error correction in field engineering: by disconnecting the integral link and retaining only pure proportional control, the proportional coefficient is gradually increased until the system exhibits stable constant-amplitude oscillations. The critical gain Kcu is measured to be approximately 1.17 and the critical oscillation period Tu is approximately 90s. The theoretical initial values ​​Kp≈0.53 and Ki=1 / Ti≈0.0134 are calculated according to the ZN standard PI tuning formula. Based on this, considering the dual large hysteresis characteristics of flue gas transmission lag and denitrification reaction in the SCR system, as well as the NO under high-moisture sludge co-firing conditions, x Due to the frequent disturbances, the proportional and integral action intensities were increased based on the initial ZN theoretical values. After repeated field trials and verifications under multiple steady-state conditions, the steady-state parameter K was finally determined. p =0.7, K i =0.2. Under this set of parameters, partition NO x Concentration overshoot less than 8%, no sustained oscillation, outlet NO x The non-uniformity coefficient can be stably controlled within 15%. Under varying operating conditions and disturbances, the PI parameter is automatically lowered to reduce the regulation intensity and avoid drastic fluctuations in ammonia injection. Specifically: under moderate disturbance conditions (load change rate 5~10MW / min or blending ratio fluctuation 10%~25%), K p Lowered to 0.45, K i Reduced to 0.08; For severe disturbance conditions (load change rate greater than 10 MW / min or blending ratio fluctuation greater than 25%), K p Reduced to 0.3, K iThe value was lowered to 0.04. When the unit load remained stable for more than 3 minutes and the sludge blending ratio fluctuation remained less than 3%, the disturbance was considered to have been resolved, and the PI parameter was restored to the steady-state value.

[0071] In one feasible implementation, in step S30, the NO of each partition is dynamically obtained. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Following the step on the non-uniformity coefficient, steps S34 to S37 are also included: Step S34, when NO x When the non-uniformity coefficient is less than or equal to the preset threshold, obtain the current ammonia injection amount and flue gas volume flow rate of each zone; Optionally, when the NO calculated in step S30 x When the non-uniformity coefficient is less than or equal to the preset threshold, it indicates that the outlet NO x The concentration distribution uniformity has met the requirements, and no further fine-tuning of ammonia injection is needed. We will now proceed to the ammonia slip detection stage. The current ammonia injection rate refers to the actual ammonia reduction dose injected into each zone at the current moment. This can be read in real time by the ammonia flow meter of each zone, or calculated based on the correspondence between valve opening and ammonia injection rate.

[0072] Step S35, according to NO x Predicted concentration, NO x Based on the concentration monitoring value, the current ammonia injection rate, and the flue gas volume flow rate, determine the theoretical ammonia slip and the theoretical total ammonia slip for each zone; Optionally, theoretical ammonia slip refers to the amount of ammonia that escapes from the reactor in a single zone without being consumed in the denitrification reaction, estimated based on the principle of material conservation. The estimation logic is: the amount of ammonia entering the zone (i.e., the current ammonia injection rate in that zone) minus the amount of ammonia consumed in the denitrification reaction. The amount of ammonia consumed in the denitrification reaction is determined based on the inlet NO of that zone. x Concentration forecast and export NO x The difference in concentration monitoring values ​​and the flue gas volume flow rate of the zone are used to calculate the result.

[0073] In one feasible embodiment, the theoretical ammonia slip ε of each partition i Calculate using the following formula: ε i = q NH3,i (C in,i C out,i ) × V gas,i × 17 / 30 Where, ε i q represents the theoretical ammonia slip in the i-th partition; NH3,iC represents the current ammonia injection rate for the i-th partition; in,i NO is the entry point of the i-th partition. x Predicted concentration value; C out,i For the exit NO of the i-th partition x Concentration monitoring value; (C in,i C out,i This reflects the amount of NO actually reduced in the denitrification reaction within this zone. x Concentration difference; V gas,i Let be the flue gas volumetric flow rate of the i-th partition; 17 is the molar mass of NH3, and 30 is the molar mass of NO. The ratio of these two values ​​is used to measure the NO content. x The reduction amount is converted into the corresponding ammonia consumption, in kg / h.

[0074] Optionally, the theoretical total ammonia slip ε total_theory This refers to the sum of the theoretical ammonia slip values ​​from all the aforementioned zones, calculated using the following formula: ε total_theory = Σ ε i Step S36: Based on the actual total ammonia escape and the theoretical total ammonia escape, the theoretical ammonia escape is corrected using the proportional correction method to obtain the actual ammonia escape of each zone. In one feasible embodiment, due to factors such as flue gas flow deviation, sampling errors at measuring points, and instrument drift, there is usually a certain deviation between the sum of theoretical ammonia slip in each zone and the actual total ammonia slip measured at the SCR outlet. The actual total ammonia slip refers to the total ammonia slip concentration across the entire outlet cross-section measured by an ammonia slip measuring instrument located at the SCR outlet. The proportional correction method first calculates the global correction proportional coefficient k: k = ε_measured / ε_total_theory Then, the theoretical escape amounts of all partitions are corrected using a proportional weighting method to obtain the corrected actual ammonia escape amounts ε' of each partition. i : ε' i = k × ε i This correction method, while maintaining the relative distribution ratio of ammonia slip in each zone, uses the measured total amount as a constraint to comprehensively correct the estimated values ​​for each zone. The k value reflects the degree of systematic deviation between the theoretical estimate and the measured value. If k > 1, it indicates that the theoretical estimate is generally too low, and all zones are adjusted upward proportionally; if k < 1, it indicates that the theoretical estimate is generally too high, and all zones are adjusted downward proportionally. This proportional correction eliminates systematic deviations caused by factors such as flue gas flow deviation and sampling errors at measuring points. Furthermore, it requires less computation and lowers the controller's computing power, making it suitable for real-time periodic operations in DCS.

[0075] Step S37: When the actual ammonia escape amount is greater than the preset zone escape threshold, the ammonia injection amount of the zone is reduced according to the preset reduction ratio.

[0076] Optionally, the preset ammonia escape threshold is an upper limit for ammonia escape in a single zone (e.g., 2 ppm). When the actual ammonia escape in a zone exceeds this threshold, it is determined that the zone has local over-spraying. When a zone is determined to be locally over-sprayed, the system reduces the ammonia injection rate of that zone by a preset reduction ratio (e.g., 10%), and re-executes the above ammonia escape estimation process in the next control cycle until the actual ammonia escape in that zone drops below the preset ammonia escape threshold, while the ammonia injection rates of other zones that have not exceeded the threshold remain unchanged.

[0077] Optionally, when the measured total ammonia slip reaches a preset global warning value (e.g., 3 ppm), the total ammonia injection rate is limited, and the valve opening of the zone with the highest ammonia injection rate is reduced first. When the measured total ammonia slip reaches a preset global protection value (e.g., 5 ppm), secondary protection is triggered, gradually reducing the total ammonia injection rate and issuing an alarm signal. This step, in conjunction with the zone ammonia slip tracing and correction, forms a dual ammonia slip control mechanism: when local over-injection causes ammonia slip in individual zones to exceed the standard, the zone tracing and correction performs targeted reduction; when the overall ammonia slip exceeds the standard, the total quantity constraint protection performs global total quantity restriction, ensuring that ammonia slip is stable and controllable across the entire operating range.

[0078] Through steps S34 to S37, this embodiment of the application solves the technical problem of difficulty in detecting and locating "overall compliance, local over-compliance" in traditional methods by independently estimating, judging, and adjusting each zone, and effectively reduces the overall ammonia escape level and the risk of local over-spraying.

[0079] In one feasible implementation, the method further includes steps B10 to B20: Step B10: Dynamically acquire the operating monitoring parameters of the flue gas denitrification device. The operating monitoring parameters include load reduction rate, main fuel trip status, furnace negative pressure fluctuation amplitude, flame loss signal, average flame intensity, flue gas oxygen content fluctuation amplitude, and furnace temperature drop rate. Optionally, operational monitoring parameters refer to multi-dimensional real-time parameters used to monitor the operational safety of the flue gas denitrification unit and its upstream boiler. Among these, the load reduction rate refers to the magnitude of load reduction per unit time, measured in MW / min, and is calculated in real-time by the DCS system based on the actual load change rate over time; Main Fuel Trip (MFT) is also included. The Trip status refers to the status signal indicating whether the boiler main fuel trip protection action has been triggered. It is a Boolean value (triggered / not triggered) and is provided by the boiler combustion management system. The furnace negative pressure fluctuation amplitude refers to the amplitude of the instantaneous pressure fluctuation inside the furnace, in Pa, and is collected in real time by the furnace pressure transmitter. The flame loss signal refers to the flame extinguishing signal detected by the flame detectors corresponding to each coal mill. Each flame detector outputs an independent on / off signal. The average flame intensity refers to the average intensity of the signals from all furnace flame detectors, expressed as a percentage of the rated threshold (for example, if the rated threshold is 100%, a measured intensity of 60% of the rated value means that the flame intensity has dropped to 60% of the rated level). The flue gas oxygen content fluctuation amplitude refers to the amplitude of the instantaneous fluctuation of the oxygen content in the flue gas at the economizer outlet, in %, and is collected in real time by the economizer outlet flue gas oxygen content analyzer. The furnace temperature drop rate refers to the rate of decrease of the average furnace temperature per unit time, in °C / min, and is calculated from the rate of change of the average temperature measurement points in each layer of the furnace. All of the above-mentioned operational monitoring parameters can be acquired in real time through the DCS system, without the need for additional monitoring equipment. The judgment logic is clear and easy to implement in engineering.

[0080] Step B20: When the operating monitoring parameters meet the preset abnormal conditions, adjust the current ammonia injection amount of each zone according to the preset fixed molar ratio.

[0081] Optionally, the preset abnormal conditions refer to the quantitative judgment criteria used to determine that the flue gas denitrification device or its upstream boiler is in an abnormal operating state, including any one or more of the following situations: (1) The unit's load reduction rate exceeds the set rate threshold (e.g., 5MW / min); (2) MFT action trigger; (3) The instantaneous fluctuation amplitude of the negative pressure in the furnace exceeds the set pressure threshold (e.g., ±200Pa) and the duration exceeds the first set duration (e.g., 10s). (4) The number of flame loss signals corresponding to a single coal mill exceeds the set number (e.g., 3 channels), or the average flame intensity of the entire furnace is lower than the set intensity threshold (e.g., 60% of the rated threshold). (5) The instantaneous fluctuation of oxygen content in the flue gas at the economizer outlet exceeds the set oxygen content threshold (e.g., ±1.5%) and the duration exceeds the second set duration (e.g., 15s). (6) The average rate of temperature drop in the furnace exceeds the set temperature threshold (e.g., 80℃ / min).

[0082] When any of the aforementioned preset abnormal conditions are met, the ammonia injection fine-tuning described in step S40 and the ammonia escape zone control described in steps S34 to S37 are immediately suspended, and the ammonia-nitrogen molar ratio is fixed to a preset fixed molar ratio value (e.g., 0.85). In this mode, the ammonia injection amount in each zone changes synchronously with the flue gas volume of that zone, but the ammonia-nitrogen molar ratio remains constant, no longer based on NO. x The concentration prediction and monitoring values ​​are dynamically differentiated to avoid drastic fluctuations in ammonia injection and uncontrolled ammonia escape caused by fine-tuning under abnormal operating conditions such as unstable combustion and rapid load reduction. When the operating monitoring parameters remain stable within the preset normal range for a period of time (e.g., 5 minutes), the system automatically exits the fixed ammonia-nitrogen molar ratio mode and resumes the ammonia injection fine-tuning steps and subsequent steps.

[0083] Through steps B10 to B20, this embodiment of the application sets up an automatic switch to a safe and conservative control mode under abnormal operating conditions. All preset abnormal condition judgment thresholds are quantified hard thresholds, with clear judgment logic and no subjective manual judgment standards. The control logic can be directly programmed, achieving full operating condition coverage of "fine adjustment under normal operating conditions and safe protection under abnormal operating conditions", thereby improving the reliability and safety of the denitrification system throughout the entire operating range.

[0084] In one feasible embodiment, the method further includes steps C10 to C30: Step C10: When the flue gas denitrification device meets the preset stable operating conditions, adjust the valve opening of each zone from the current opening to the first opening and maintain it for a preset time to obtain the first outlet NO. x Concentration change rate, where the first opening degree is greater than the current opening degree; Optionally, the preset stable operating conditions refer to the set of conditions that ensure the unit's operating status meets the stability requirements for calibration operations. These conditions include: maintaining unit load fluctuations within a set range (e.g., ±3MW), maintaining fuel blending ratio fluctuations within a set range (e.g., ±2%), and ammonia slip concentration below a set threshold (e.g., 2ppm). Calibration operations are only initiated when these stable operating conditions are met to avoid interference from load changes or fuel fluctuations with the calibration results. The calibration operation is automatically triggered and executed according to a preset cycle (e.g., every 7 days), requiring no manual intervention or shutdown.

[0085] In one feasible embodiment, the first opening degree is greater than the current opening degree, for example, by 5% based on the current opening degree. The preset time period refers to the time (e.g., 3 minutes) required for the flue gas denitrification system to reach a new stable state after the valve opening degree is adjusted. Within this preset time period, the system maintains the valve opening degree unchanged, and collects the outlet NO corresponding to the zone after the system stabilizes. x The concentration data was used to calculate the rate of change of NOx concentration at the first outlet.

[0086] Step C20: Adjust the valve opening to the second opening degree and maintain it for a preset time to obtain the second outlet NO. x Concentration change rate, wherein the second opening degree is less than the current opening degree; In one feasible embodiment, a positive disturbance (increasing the valve opening) is first performed to collect data. Then, the valve opening is restored to its original opening and left to stand for a period of time (e.g., 5 minutes) to eliminate residual disturbance. Next, a negative disturbance (decreasing the valve opening) is performed. The second opening is less than the current opening, for example, reduced by 5% from the current opening. After maintaining the same preset duration, the outlet NO corresponding to that zone is collected. x Concentration data were used to calculate the NO concentration at the second outlet. x Concentration change rate. This was determined by collecting data on outlet NO under both positive and negative perturbations. x The concentration change rate can eliminate the directional influence of flue gas fluctuations on the measurement results (e.g., the flue gas's own NO content). x The slow upward or downward trend of concentration produces errors in the same direction in both positive and negative perturbations (this error can be offset by taking the combined result of the two perturbations), thus improving the accuracy of calibration.

[0087] Step C30, according to the first outlet NO x Concentration change rate and second outlet NO x The concentration change rate is used to determine the actual flow coefficient of the valves in each zone, and the correspondence between the valve opening and the ammonia injection amount is updated based on the actual flow coefficient of the valves. The correspondence is used for the step of coarse adjustment of ammonia injection in each zone based on the target ammonia injection amount, and / or for the step of fine adjustment of ammonia injection in each zone based on the feedback ammonia injection amount.

[0088] Optionally, the actual flow coefficient of a valve refers to a characteristic parameter reflecting the current actual flow capacity of the valve. As the operating time increases, factors such as valve core wear, media erosion, and valve stem jamming can cause the actual flow characteristics of the valve to deviate from the theoretical characteristics at the time of manufacture. This results in a deviation between the actual ammonia flow rate and the theoretical value at the same opening degree, with a deviation of up to ±15%.

[0089] In one feasible embodiment, based on the first outlet NO x The concentration change rate and the second outlet NOx concentration change rate are used to calculate the outlet NOx under each set of step perturbations. x Concentration change rate dNO x / dK (i.e., export NO) x (The change in concentration divided by the change in valve opening), and then combined with the outlet NO in the denitrification reaction kinetic model. x The correlation between the concentration change rate and the ammonia injection rate change rate can be used to deduce the true flow coefficient of the valve in this zone under the current condition: Under steady operating conditions, the outlet NO xThere is a definite kinetic relationship between the rate of change in concentration and the rate of change in ammonia injection rate, which can be determined by measuring the outlet NO. x The actual rate of change in concentration is used to calculate the actual change in ammonia injection rate based on the kinetic relationship. This change is then divided by the change in valve opening to obtain the actual flow coefficient of the valve near its current opening position. This actual flow coefficient is then used to replace the original theoretical flow coefficient or the previously calibrated flow coefficient in the controller, thus updating the correspondence between valve opening and ammonia injection rate for that zone.

[0090] Through steps C10 to C30, the embodiments of this application adopt a small opening step disturbance, which has a low disturbance amplitude and almost no interference with the denitrification index of the unit. It can be completed fully automatically online without the need for shutdown or load reduction for manual calibration. After calibration, the zonal flow error can be controlled within ±2%, ensuring the execution accuracy of ammonia injection coarse and fine adjustment commands during long-term operation.

[0091] In one feasible embodiment, the method further includes steps D10 to D40: Step D10: Obtain historical operating data of the flue gas denitrification unit; Optionally, historical operating data refers to the multi-dimensional operating records accumulated and stored by the flue gas denitrification unit during its past operation, including current operating condition parameters at each historical moment (such as unit load, fuel blending ratio, received nitrogen content of coal, total air volume, average furnace temperature, burner tilt angle, etc.), valve opening parameters corresponding to each zone, and corresponding NO. x Concentration monitoring values ​​(including inlet NO) x Concentration and export NO x Concentration) and ammonia slip monitoring values. The above historical operating data can be extracted from the historical database of the DCS system, covering various operating conditions during the long-term operation of the unit. The larger the data volume and the more comprehensive the operating conditions covered, the better the subsequent clustering and iteration effects.

[0092] Step D20: The historical operating data is divided into operating conditions using a clustering algorithm to obtain multiple groups of typical operating condition categories; Optionally, clustering algorithms are unsupervised learning algorithms used to divide historical operating data into several categories (clusters) based on the similarity of operating condition characteristics, so that data within the same category have similar operating condition characteristics, while data between different categories have obvious operating condition differences.

[0093] In one feasible embodiment, the K-means clustering algorithm is used, with the current operating condition parameters (such as load, co-firing ratio, etc.) at each historical moment as the clustering feature vector. The massive historical operating data is divided into a preset number of categories (e.g., 50 groups), with each category representing a typical operating condition. After clustering, each typical operating condition category corresponds to a cluster center in a high-dimensional space. The coordinates of the cluster center are the operating condition feature vector of that typical operating condition, representing the typical combination of values ​​for each operating parameter under that type of operating condition.

[0094] Step D30: Associate each group of typical working condition categories with the valve opening parameters corresponding to each group of typical working condition categories to build a working condition parameter library; In one feasible embodiment, for each typical operating condition category, the category with the best denitrification effect (i.e., outlet NO) is selected from the historical operating data belonging to that category. x The time period with the lowest non-uniformity coefficient and lowest ammonia slip is recorded, and the valve opening parameters actually used in each zone during that time period are extracted as the optimal valve opening benchmark for that typical operating condition. Then, the operating condition feature vectors of various typical operating conditions and their corresponding optimal valve opening benchmarks are associated and stored to construct an operating condition parameter library.

[0095] Once the operating condition parameter library is constructed, it can be embedded into a DCS system or an ammonia injection optimization controller. In the coarse ammonia injection adjustment stage of step S20, the target typical operating condition category to which the current operating condition belongs is first identified (by calculating the Euclidean distance between the current operating condition parameters and each cluster center, selecting the category with the closest distance as the target category). Then, the baseline valve opening parameters corresponding to the target typical operating condition category are called from the operating condition parameter library as the initial valve opening for each zone. Fine adjustments are then made based on measured data. Through this parameter library calling mechanism, the feedforward response time under varying operating conditions can be shortened to the second level, significantly improving the adjustment response speed under varying load and blending ratio conditions.

[0096] Step D40: Update the working condition parameter library to the model.

[0097] In one feasible embodiment, by updating the operating condition parameter library to the model, the model, upon receiving the current operating condition parameters, outputs the NO for each partition. x In addition to the concentration prediction value, the corresponding reference valve opening parameters can also be matched and called from the operating condition parameter library simultaneously, further improving the response speed and accuracy of the feedforward coarse adjustment.

[0098] Optionally, newly added operational data (including actual collected entry NO) can be automatically extracted within a preset period (e.g., monthly). x Concentration, export NO xconcentration, the actual ammonia injection amount of each zone and the corresponding operating condition parameters, etc.), the newly added operating data is input into the model as a new training sample, the prediction error of the model under the current parameters is calculated, and the weight parameters of the model are fine-tuned using the error back propagation algorithm (i.e., incremental training), so that the model gradually adapts to the latest system characteristic changes on the basis of retaining the original memory. Through this incremental training, the model can adapt to long-term slow-varying characteristics such as catalyst activity attenuation, flow field offset, and air preheater air leakage variation, so that the prediction accuracy of the model does not decrease with the extension of operating time.

[0099] Through steps D10 to D40, the embodiments of the present application, on the one hand, solidify the verified and effective operating condition parameters into a parameter library that can be called at the second level, and on the other hand, enable the model to continuously adapt to the long-term changes of the system through monthly incremental training, so that the control effect remains stable throughout the full life cycle, and the problem of fixed control parameters and year-by-year attenuation of operating effects in traditional methods is avoided.

[0100] Exemplarily, with reference to Figure 4 , a zoned ammonia injection control process is provided: first, the process starts, and operating condition identification is performed; then the prediction model outputs the target ammonia injection amount of each zone to realize feedforward prediction and preliminarily match the NO x load difference of different zones; coarse adjustment of ammonia injection is performed according to the target ammonia injection amount of each zone to narrow the NO x distribution deviation of each zone; the monitored NO x concentration value of each zone is obtained and used as the basis for feedback regulation; it is judged whether the NO x non-uniformity coefficient meets the standard; if the judgment result is no, fine adjustment of ammonia injection is performed, and after the fine adjustment of ammonia injection is completed, the process jumps to the step of obtaining the monitored NO x concentration value of each zone; if the judgment result is yes, the actual ammonia escape amount is calculated; it is judged whether the actual ammonia escape amount is greater than the preset zone escape threshold; if the judgment result is yes, the ammonia injection amount is reduced, and after the ammonia injection amount reduction is completed, the process jumps to the step of calculating the actual ammonia escape amount; if the judgment result is no, the operating condition parameters are solidified, and then the process ends.

[0101] To further verify the effect of the embodiments of the present application, a verification example is set, which is applied to an SCR denitration system supporting a 330MW subcritical tangentially-fired boiler, burning a mixed fuel of high-moisture Indonesian coal and dried sludge with a blending ratio of 50%. The system is equipped with a 12-zone ammonia injection grid, grid NO x measuring points at the inlet and outlet, an ammonia escape monitoring device and an ammonia injection optimization controller, and is controlled by the zoned ammonia injection control method described in the present application. The control effect of the present application is compared with the conventional ammonia injection control mode. The conventional ammonia injection control mode is: only taking the NO xThe average concentration is used as a feedback signal for closed-loop control of the total ammonia injection volume. The opening degree of the branch pipe valves in each zone of the ammonia injection grid is manually set to a fixed distribution ratio based on experience, and no adaptive adjustment is made during daily operation. Under the same operating conditions, the outlet NO is controlled. x With consistent average concentrations, the operating data for both control methods were recorded, and the results are shown in Table 1:

[0102] As can be seen, the zoned ammonia injection control method provided in this application achieves the control of ammonia injection rate in each zone and inlet NO under the condition of blended coal combustion. x The adaptive and precise matching of the distribution can simultaneously improve denitrification efficiency, reduce ammonia slip, and save ammonia consumption, resulting in good environmental and economic benefits.

[0103] In this embodiment, based on a pre-built model, the current operating parameters of the flue gas denitrification device are input, and the NO levels in each zone of the flue gas duct are output. x Predicted NO concentration values ​​can be obtained before ammonia injection adjustments. x The spatial distribution of NO provides accurate feedforward data for subsequent zoned differentiated ammonia injection; based on NO x The concentration prediction value is used to determine the target ammonia injection rate for each zone. Based on the target ammonia injection rate, the ammonia injection rate for each zone is coarsely adjusted to ensure that the ammonia injection rate for each zone corresponds to the NO in the corresponding area. x The total quantity was initially matched in spatial distribution, enabling a rapid response to changes in operating conditions; NO was dynamically acquired for each zone. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Inhomogeneity coefficient, NO x The uniformity of distribution is quantified into a measurable control indicator, providing a clear basis for determining whether to initiate fine-tuning and avoiding ineffective adjustments when uniformity has already met the target; when NO x When the non-uniformity coefficient is greater than a preset threshold, according to NO x Based on the concentration monitoring values, the corresponding ammonia injection rate for each zone is determined, and the ammonia injection rate for each zone is fine-tuned according to the feedback ammonia injection rate, so that NO... x Zones with high concentrations of nitrogen oxides and nitrogen oxides were compensated by ammonia and NO injection. x The ammonia injection rate was reduced accordingly in zones with low concentrations, forming an independent closed-loop correction for each zone and improving the overall NO concentration at the outlet section. x Concentration uniformity. This application achieves uniformity of ammonia injection rate and inlet NO in each zone under mixed coal combustion conditions. x The adaptive and precise matching of the distribution effectively improves the accuracy of ammonia injection under complex co-firing conditions.

[0104] A second aspect of this application provides an electronic device, including a memory and a processor, the processor being capable of executing the steps of the zoned ammonia injection control method described above. Reference is made below. Figure 5 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic device in these embodiments may include, but is not limited to, terminal devices such as smartphones, laptops, and tablets (PADs). Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0105] like Figure 5 As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: an input device 1007 including, for example, a touchscreen, physical buttons, an image sensor, a microphone, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although the diagrams show electronic devices with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented alternatively.

[0106] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0107] The electronic device provided in this application employs the zoned ammonia injection control method described in the above embodiments. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the zoned ammonia injection control method provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0108] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0109] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0110] A third aspect of this application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the zoned ammonia injection control method in the above embodiments.

[0111] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0112] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0113] The aforementioned computer-readable storage medium carries one or more programs that, when executed by an electronic device, cause the electronic device to: based on a pre-built model, input the current operating parameters of the flue gas denitrification device, and output the NO levels of each zone in the flue gas duct. x Predicted concentration values; based on NO x Predicted concentration values ​​are used to determine the target ammonia injection rate for each zone, and coarse adjustments to the ammonia injection rate for each zone are made based on the target ammonia injection rate; NO levels for each zone are dynamically acquired. x Concentration monitoring values, and based on NO x Concentration monitoring values ​​were used to determine the NO concentration in each zone. x Inhomogeneity coefficient; when NO x When the non-uniformity coefficient is greater than a preset threshold, according to NO x The concentration monitoring value is used to determine the corresponding ammonia injection amount for each zone, and the ammonia injection for each zone is fine-tuned based on the feedback ammonia injection amount.

[0114] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0115] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0116] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0117] The readable storage medium provided in this application is a computer-readable storage medium that stores a program implementing the partitioned ammonia injection control method described above. This program is executed by a processor to implement the steps of the partitioned ammonia injection control method described above. Compared with the prior art, the beneficial effects of the readable storage medium provided in this application are the same as those of the partitioned ammonia injection control method provided in the above embodiments, and will not be repeated here.

[0118] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for controlling ammonia injection in designated zones, characterized in that, Includes the following steps: Based on a pre-built model, the current operating parameters of the flue gas denitrification device are input, and the NO values ​​of each zone in the flue gas duct are output. x Concentration prediction; According to the NO x The concentration prediction value is used to determine the target ammonia injection amount for each zone, and the ammonia injection amount for each zone is coarsely adjusted according to the target ammonia injection amount. Dynamically obtain the NO of each partition. x Concentration monitoring value, and according to the NO x Concentration monitoring values ​​were used to determine the NO concentration in each of the aforementioned zones. x Inhomogeneity coefficient; When the NO x When the non-uniformity coefficient is greater than a preset threshold, according to the NO x The concentration monitoring value is used to determine the corresponding feedback ammonia injection amount for each zone, and the ammonia injection for each zone is fine-tuned based on the feedback ammonia injection amount.

2. The method as described in claim 1, characterized in that, The NO x The steps for determining the target ammonia injection amount for each zone, based on the concentration prediction value, include: Obtain the flue gas volumetric flow rate of each of the aforementioned zones; According to the NO x Based on the predicted concentration and the flue gas volumetric flow rate, the NO concentration for each zone is calculated. x Total amount; According to the NO x The target ammonia injection amount for each zone is obtained by multiplying the total amount by the preset ammonia-nitrogen molar ratio.

3. The method as described in claim 1, characterized in that, The NO x Concentration monitoring values ​​were used to determine the NO concentration in each of the aforementioned zones. x The steps for determining the non-uniformity coefficient include: According to the NO x Concentration monitoring values ​​were used to determine the NO concentration in the flue. x Average concentration; According to the NO x Concentration monitoring value and the NO x The first deviation of the average concentration is used to determine the NO concentration in each of the aforementioned zones. x Concentration standard deviation; According to the NO x Concentration standard deviation and the NO x The ratio of the average concentrations is used to obtain the NO concentration for each of the aforementioned zones. x Non-uniformity coefficient.

4. The method as described in claim 1, characterized in that, The NO x The steps for determining the corresponding ammonia injection amount for the zone based on the concentration monitoring value include: Determine the NO x The second deviation between the concentration monitoring value and the preset outlet concentration target value; The proportional adjustment amount of the partition is determined based on the second deviation and the preset proportional coefficient; The integral adjustment amount of the partition is determined based on the cumulative amount of the second deviation and the preset integral coefficient. The feedback ammonia injection amount for the zone is obtained by summing the proportional adjustment amount and the integral adjustment amount.

5. The method as described in claim 1, characterized in that, Based on the pre-built model, the current operating parameters of the flue gas denitrification device are input, and the NO values ​​of each zone in the flue gas duct are output. x Before the concentration prediction step, the following steps are also included: In each of the aforementioned zones, entry and exit measurement points are set up. The entry NO of each partition is obtained through the entry measurement point. x Concentration and flue gas flow parameters, and obtain the outlet NO of each zone through the outlet measuring point. x concentration; The entry NO x Concentration, the flue gas flow parameters, the outlet NO x The concentration and the amount of ammonia injected in each of the aforementioned zones are correlated to obtain the dataset used to construct the model.

6. The method as described in claim 1, characterized in that, The NO of each partition is dynamically obtained. x Concentration monitoring value, and according to the NO x Concentration monitoring values ​​were used to determine the NO concentration in each of the aforementioned zones. x Following the step of determining the non-uniformity coefficient, the following steps are also included: When the NO x When the non-uniformity coefficient is less than or equal to the preset threshold, the current ammonia injection amount and flue gas volume flow rate of each partition are obtained; According to the NO x Predicted concentration value, the NO x The concentration monitoring value, the current ammonia injection rate, and the flue gas volume flow rate are used to determine the theoretical ammonia slip and the theoretical total ammonia slip for each zone. Based on the actual total ammonia escape and the theoretical total ammonia escape, the theoretical ammonia escape is corrected using a proportional correction method to obtain the actual ammonia escape of each of the partitions; When the actual ammonia escape amount is greater than the preset zone escape threshold, the ammonia injection amount of the zone is reduced by a preset reduction ratio.

7. The method as described in claim 1, characterized in that, The method further includes: The operating monitoring parameters of the flue gas denitrification device are dynamically acquired, including the load reduction rate, main fuel trip status, furnace negative pressure fluctuation amplitude, flame loss signal, average flame intensity, flue gas oxygen content fluctuation amplitude, and furnace temperature drop rate. When the operation monitoring parameters meet the preset abnormal conditions, the current ammonia injection amount of each zone is adjusted according to the preset fixed molar ratio.

8. The method as described in claim 1, characterized in that, The method further includes: When the flue gas denitrification device meets the preset stable operating conditions, the valve opening of each zone is adjusted from the current opening to the first opening and maintained for a preset duration to obtain the first outlet NO. x Concentration change rate, wherein the first opening degree is greater than the current opening degree; Adjust the valve opening to the second opening and maintain it for the preset duration to obtain the second outlet NO. x Concentration change rate, wherein the second opening degree is less than the current opening degree; According to the first export NO x Concentration change rate and the second outlet NO x The concentration change rate is used to determine the actual flow coefficient of the valve in each of the partitions, and the correspondence between the valve opening and the ammonia injection amount is updated based on the actual flow coefficient of the valve. The correspondence is used in the step of coarsely adjusting the ammonia injection of each partition based on the target ammonia injection amount, and / or in the step of finely adjusting the ammonia injection of each partition based on the feedback ammonia injection amount.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the processor being capable of executing the steps of the zoned ammonia injection control method as described in any one of claims 1 to 8.

10. A storage medium, the storage medium being a computer-readable storage medium, the storage medium storing a computer program, the computer program being executed by a processor to implement the steps of the zoned ammonia injection control method as described in any one of claims 1 to 8.

Images