Ammonia injection self-adaptive control method and system for SCR denitration large lag

Abstract

The application discloses a kind of facing SCR denitration big hysteresis spray ammonia self-adapting control method and system, method includes: collection NO x Concentration value is added to sliding observation window, transient protection condition is checked, transient protection action is executed when meeting transient protection condition, after valve opening degree is adjusted based on preset independent protection amplitude and valve control instruction is generated;Judge whether the time interval from the last valve adjustment action reaches the preset adjustment period, generate valve control instruction to maintain the current valve opening degree when not reaching;Stability analysis is carried out to NO x Concentration data in sliding observation window, corresponding valve control instruction is generated;After outputting valve control instruction to spray ammonia regulating valve actuator, enter next control cycle.The application is applied to flue gas purification field, can significantly reduce valve action frequency under the premise of guaranteeing denitration efficiency and emission standard requirement, simplify system structure, reduce engineering implementation and maintenance cost.

Description

Technical Field

[0001] This invention relates to the fields of flue gas purification and environmental protection automatic control, and industrial Internet of Things technology, specifically a method and system for adaptive ammonia injection control for large hysteresis in SCR denitrification. Background Technology

[0002] Selective catalytic reduction (SCR) denitrification technology is used to remove nitrogen oxides (NOx) from industrial flue gas. x The mainstream technology for NO emission control is based on injecting a reducing agent (such as ammonia) into the flue gas duct, which, under the action of a catalyst, reduces NO. x It is reduced to harmless nitrogen and water. Precise control of the ammonia injection rate directly determines the denitrification efficiency, ammonia slip rate, and operating economy, and is a key aspect of SCR system operation.

[0003] The SCR denitrification process exhibits a significant time lag: from the ammonia injection valve actuation to the outlet NO... x The concentration produces measurable changes, typically with a pure time lag of over 60 seconds, and the system exhibits second-order inertial characteristics, with a time constant ranging from 60 to 180 seconds. This large time lag poses a significant challenge to traditional continuous regulation control strategies (such as PID control), making it difficult to achieve a balance between response speed and system stability.

[0004] Currently, extensive research has been conducted on the ammonia injection control problem in SCR denitrification systems, resulting in the following main technical solutions: The first category is a feedforward-feedback composite ammonia injection control scheme based on predictive models. Some studies construct multi-dimensional dynamic predictive models by integrating Kalman filtering and neural networks to predict outlet NO. x Concentration and generate feedforward correction; simultaneously based on the outlet NO x The deviation between the actual concentration and the setpoint generates a feedback correction, and the final ammonia injection flow rate is obtained by combining the baseline flow rate, feedforward, and feedback correction. This type of scheme also includes a self-learning optimization module to dynamically correct the ammonia-nitrogen molar ratio, and is mainly used for ammonia injection flow control in low-temperature SCR denitrification systems. However, this type of scheme relies on complex prediction models, requires a large amount of historical data for training, and has high computational overhead for online model updates; prediction accuracy decreases when operating conditions change drastically or data distribution shifts, which may lead to control deviations; the system is highly complex, and on-site deployment and maintenance are difficult. Other studies use different concentration prediction methods to predict reactor NO. x The concentration of NO is determined through collaborative computation between a concentration prediction model and a state compensator. xThe concentration is then controlled using a PID controller combined with ammonia flow feedforward control commands for closed-loop regulation, primarily applied to the automated control of SCR denitrification in power plants. This scheme is essentially still a PID continuous regulation architecture, with frequent valve actions; multi-model collaboration increases system complexity; and the design of the state compensator relies on accurate system model parameters.

[0005] The second type is a precision ammonia injection control scheme based on spatial zoning. This type of scheme divides the flue into several spatial zones, each with an independent ammonia injection unit; it predicts the NO injection inlet based on feedforward information from unit operating parameters. x Adjust the ammonia injection flow rate according to the concentration changes, based on the NO concentration in each area. x Fine-tuning the flow rate of each ammonia injection unit is suitable for ammonia injection control in SCR denitrification reactors with large cross-section flues. However, this type of solution requires the installation of multiple ammonia injection units and zoned NOx control. x The monitoring equipment requires significant hardware investment and upgrades; the system structure is complex, and multi-loop coordinated control is difficult; it is not economically viable for small and medium-sized SCR systems.

[0006] The third category is adaptive control schemes based on Smith predictor compensation. This type of scheme uses recursive least squares to identify dynamic characteristics online, designs an adaptive fuzzy PID strategy based on Smith predictor compensation, and combines it with combustion load feedforward compensation for composite control. It is mainly used for flue gas system airflow regulation. However, this scheme requires continuous application of excitation signals for online identification, which may interfere with normal production; furthermore, it is still a PID continuous regulation architecture, which cannot fundamentally solve the mechanical wear problem caused by frequent valve operations; and the rule design relies on expert experience, limiting its versatility.

[0007] Furthermore, in industrial practice, many small and medium-sized SCR denitrification systems still employ an open-loop control method with a fixed valve opening: operators set a fixed ammonia injection valve opening based on experience, and the system operates continuously at this fixed opening. While this method is simple and reliable, requiring no complex controller, it suffers from issues with the inlet NO... x NO concentration fluctuations at the outlet x The concentration fluctuated significantly, failing to meet increasingly stringent environmental emission requirements.

[0008] In summary, the mainstream directions of existing SCR ammonia injection control technologies focus on predictive models, feedforward-feedback composite control, and spatial partitioning control. Their common characteristic is the use of continuous adjustment mechanisms (based on PID controllers), resulting in frequent valve actuation. Existing technologies generally suffer from the following common drawbacks: Frequent valve actuation and severe mechanical wear: Existing solutions generally adopt PID continuous regulation control, in which the controller outputs a regulation signal in each control cycle, causing the ammonia injection regulating valve to actuate frequently, accelerating the wear of the valve actuator, increasing maintenance costs, and affecting the long-term operational reliability of the system.

[0009] Complex model dependency and difficult engineering deployment: Schemes based on prediction models (neural networks, Kalman filters, etc.) require a large amount of historical data for training, and online model updates consume a lot of computing resources. The on-site deployment and subsequent maintenance are difficult to achieve, making it difficult to promote and apply them in small and medium-sized SCR systems.

[0010] Lack of a dedicated protection mechanism for instantaneous exceedances: The control logic of existing solutions mainly focuses on eliminating steady-state deviations, when the inlet NO... x Drastic fluctuations in NO concentration led to export x When the instantaneous and significant exceedance occurs, the response speed of conventional PID control is limited by the large lag characteristic of the system, making it difficult to implement effective intervention before the harm of exceeding the limit occurs.

[0011] High hardware modification costs: The spatial partitioning-based solution requires the addition of multiple ammonia injection units and partition monitoring devices, which involves large-scale hardware modifications to the existing SCR system, resulting in high investment costs. Furthermore, multi-loop coordinated control increases the system complexity.

[0012] Large hysteresis characteristics lead to insufficient adaptability: The pure hysteresis time (more than 60 seconds) of the SCR denitrification process makes the proportional and integral actions of PID controllers prone to overshoot and oscillation, making parameter tuning difficult and making it hard to achieve a good balance between response speed and stability.

[0013] Therefore, there is an urgent need in this field to develop an SCR denitrification ammonia injection control technology that can overcome the above-mentioned defects of existing technologies, significantly reduce valve operation frequency, simplify system structure, reduce engineering implementation and maintenance costs, and have the ability to respond quickly to instantaneous exceedances of standards, while ensuring denitrification efficiency and emission compliance requirements. Summary of the Invention

[0014] To address the significant time lag in the SCR denitrification process and the common problems in existing ammonia injection control methods, such as frequent valve actuation, high system complexity, and lack of transient protection mechanisms, this invention provides an adaptive ammonia injection control method and system for SCR denitrification with significant time lag. It employs a low-frequency hybrid control strategy that uses a fixed baseline value as the normal state, stability deviation as the adjustment trigger condition, and transient exceedance as the emergency protection condition. This strategy significantly reduces valve actuation frequency while ensuring denitrification efficiency and emission compliance. Furthermore, it does not rely on complex prediction models or spatial partitioning hardware, simplifying the system structure and reducing engineering implementation and maintenance costs. It also possesses rapid response capabilities for transient exceedances.

[0015] Specifically, the present invention aims to solve the following technical problems: (1) How to ensure export NO x Under the premise of accurate concentration control, the operating frequency of the ammonia injection regulating valve is reduced by an order of magnitude, thereby significantly reducing the mechanical wear of the valve actuator and extending the equipment life. (2) How to achieve effective adaptive control of the SCR denitrification process without relying on complex prediction models (such as neural networks, Kalman filters, etc.) and complex hardware modifications (such as spatial partitioning ammonia injection), thereby reducing the threshold for system deployment and maintenance; (3) How to export NO x The system enables rapid and accurate identification and response to instantaneous concentration exceedances, establishing an independent transient protection channel to prevent the escalation of hazards caused by the system's large hysteresis characteristics. (4) How to make the controller's adjustment range adaptively scale according to the deviation of the system's long-term operating state, so as to avoid over- or under-adjustment of the fixed adjustment range under different disturbances.

[0016] To achieve the above objectives, this invention provides an adaptive ammonia injection control method for large hysteresis in SCR denitrification, comprising the following steps: Step 1: Real-time collection of NO at the outlet of the SCR denitrification reactor x The concentration value is added to a sliding observation window of a preset length, and the NO value is updated. x Long-term mean error of concentration; Step 2, based on the currently collected NO x The concentration value is used to check the transient protection conditions to determine whether the transient protection conditions are met. If so, execute the transient protection action, adjust the valve opening based on the preset independent protection amplitude and generate a valve control command, then proceed to step 5; Otherwise, proceed to step 3; Step 3: Determine whether the time interval since the last valve adjustment has reached the preset adjustment cycle. If so, proceed to step 4; Otherwise, after generating a valve control command to maintain the current valve opening, proceed to step 5; Step 4, for NO within the sliding observation window x Stability analysis was performed on the concentration data, and a stability status determination was conducted. If the state is determined to be stable, after generating a valve control command to maintain the current valve opening, proceed to step 5; If the condition is determined to be continuously high, the adaptive adjustment range is calculated based on the long-term error mean, and a valve control command to increase the valve opening is generated accordingly before proceeding to step 5. If the condition is determined to be a persistently low state, the adaptive adjustment range is calculated based on the long-term error mean, and a valve control command to reduce the valve opening is generated accordingly before proceeding to step 5. Step 5: After outputting the valve control command to the ammonia injection regulating valve actuator, return to step 1 to enter the next control cycle.

[0017] To achieve the above objectives, the present invention also provides an adaptive ammonia injection control system for large hysteresis in SCR denitrification, employing the above-described method, wherein the adaptive ammonia injection control system includes: Export NO x The online concentration monitoring module is used to collect NO concentration data at the outlet of the SCR denitrification reactor in real time. x Concentration data; The time reference and timing module is used to provide a unified time reference for the system and record the occurrence time of each normal valve adjustment and transient protection action; The transient protection module is used to perform transient protection condition checks and trigger independent transient protection actions through a combined determination of two conditions. The adjustment trigger judgment module is used to determine whether the time interval since the last valve adjustment action has reached the preset adjustment cycle; The stability analysis module is used to analyze NO within the sliding observation window. x Statistical analysis of concentration data is performed to determine the system's operating status; The adaptive adjustment range calculation module is used to calculate the adaptive adjustment range of the valve opening based on the long-term error average. The baseline maintenance management module is used to maintain the current valve opening unchanged when the system is stable or the adjustment cycle has not been reached; The valve opening command output module is used to summarize various control commands and output the valve opening target value after limiting to the ammonia injection regulating valve actuator.

[0018] Compared with the prior art, the present invention has the following beneficial technical effects: 1. This invention adopts a low-frequency hybrid control strategy based on a fixed base value. The system maintains a fixed valve opening for most of the time, and only triggers a limited number of adjustments when the stability condition is not met and the adjustment cycle is reached. Compared with PID-type continuous control schemes, the valve action frequency can be reduced by more than an order of magnitude, which significantly reduces the mechanical wear of the actuator, extends the service life of the valve and actuator, and reduces maintenance costs and the risk of unplanned downtime due to valve failure. 2. The core logic of this invention is based on statistical analysis and threshold judgment. It does not require training complex prediction models such as neural networks and Kalman filters, does not rely on a large amount of historical data and online computing resources, and all control parameters have clear physical meanings. Field engineers can directly tune them according to system characteristics. The deployment and maintenance threshold is low, making it particularly suitable for low-cost upgrades and transformations of existing SCR systems. 3. This invention sets up an independent transient protection channel, employing a dual-condition joint determination of "duration + growth trend," which can detect NO at the outlet. x It responds quickly when the concentration exceeds the standard, and effectively distinguishes between "persistent exceedance" and "brief random spike" through a dual-condition design to avoid false triggering. This mechanism is independent of the normal adjustment logic, and the protection range can be set separately, providing dual protection for environmental emission compliance. 4. This invention dynamically adjusts the adjustment range based on the long-term error accumulation of the SCR system. Under large deviation conditions, it automatically increases the adjustment amount to accelerate the return to the target value, and under small deviation conditions, it automatically decreases the adjustment amount to avoid over-adjustment and oscillation. This achieves a positive match between the control strength and the degree of deviation, without the need for manual intervention to switch control parameters. 5. This invention, through the design principle that "the adjustment cycle must be greater than the system's dynamic response time," ensures that the effect of each adjustment is fully reflected in the export NO. x The next judgment is made only after the concentration has increased, fundamentally avoiding the oscillation problems caused by integral saturation and phase lag that are common in PID controllers in systems with large time delays. The structural characteristics of the control strategy (discontinuous feedback regulation) are inherently compatible with the large time delay characteristics of the SCR system. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0020] Figure 1 This is a flowchart of the adaptive ammonia injection control method for large hysteresis in SCR denitrification in an embodiment of the present invention; Figure 2 This is a block diagram of the adaptive ammonia injection control system for large hysteresis in SCR denitrification in an embodiment of the present invention.

[0021] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0024] Example 1 The core controlled variable of the SCR denitrification system is the outlet NO. x The concentration is manipulated by the ammonia injection valve opening. The main disturbance in the SCR denitrification system comes from the inlet NO concentration. x Fluctuations in NO concentration (caused by changes in upstream process conditions). Due to the significant pure time lag (typically over 120 seconds) in the SCR denitrification reactor and flue gas sampling pipeline, and the second-order inertial characteristic of the system's dynamic response, the effect of valve action requires a considerable amount of time to be reflected in the outlet NO. x This is reflected in the concentration. Based on this, this embodiment discloses an adaptive ammonia injection control method for SCR denitrification with large hysteresis. It abandons the traditional continuous PID control paradigm and instead adopts a low-frequency hybrid control strategy of "fixed base value operation + periodic stability judgment triggering intermittent adjustment + transient exceedance dual-condition special protection". The basic concept of this control strategy is: for the vast majority of the time, the SCR denitrification system operates at a preset fixed valve opening (base value), only when NO is detected at the outlet... x A limited-amplitude valve adjustment is triggered only when the concentration deviates persistently and significantly; simultaneously, an independent transient protection channel is established, which triggers an adjustment when the outlet NO... x When the concentration momentarily exceeds the standard, a rapid response mechanism independent of normal adjustments is used for intervention.

[0025] refer to Figure 1 The adaptive ammonia injection control method for large hysteresis in SCR denitrification in this embodiment specifically includes the following steps: Step 1: Real-time collection of NO at the outlet of the SCR denitrification reactor x The concentration value is added to a sliding observation window of a preset length, and the NO value is updated. x Long-term mean error of concentration; Step 2, based on the currently collected NO x The concentration value is used to check the transient protection conditions to determine whether the transient protection conditions are met. If so, execute the transient protection action, adjust the valve opening based on the preset independent protection amplitude and generate a valve control command, then proceed to step 5; Otherwise, proceed to step 3; Step 3: Determine whether the time interval since the last valve adjustment has reached the preset adjustment cycle. If so, proceed to step 4; Otherwise, after generating a valve control command to maintain the current valve opening, proceed to step 5; Step 4, for NO within the sliding observation window x Stability analysis was performed on the concentration data, and a stability status determination was conducted. If the state is determined to be stable, after generating a valve control command to maintain the current valve opening, proceed to step 5; If the condition is determined to be continuously high, the adaptive adjustment range is calculated based on the long-term error mean. After generating the corresponding valve control command to increase the valve opening, step 5 is performed. If the condition is determined to be a persistently low state, the adaptive adjustment range is calculated based on the long-term error average. After generating the corresponding valve control command to reduce the valve opening, step 5 is performed. Step 5: After outputting the valve control command to the ammonia injection regulating valve actuator, return to step 1 to enter the next control cycle.

[0026] In the specific implementation of step 1, the sampling period can be set to 1 second or 2 seconds, etc., and the sliding observation window will contain the historical NO of the most recent few seconds. x Concentration data must be obtained, and the length of the sliding observation window must be designed to cover the dynamic response time scale of the SCR denitrification system to ensure that the stability analysis in step 4 is statistically significant.

[0027] In the specific implementation of step 2, the transient protection condition check adopts a dual-condition joint judgment. If either condition is met, it is determined that the transient protection condition is met: Condition 1 (Duration Condition): The currently collected NO x The concentration value exceeds the preset transient protection threshold, and the duration of the exceeding state reaches the preset transient protection response time; this condition ensures that the transient exceeding is not a brief spike noise, but a continuous abnormal operating condition.

[0028] Condition 2 (Deteriorating Trend Condition): The currently collected NO x The concentration value exceeds the preset transient protection threshold, and within the trend analysis window, NO x The concentration values ​​show a clear upward trend; for example, the increase in the final value of the trend analysis window compared to the initial value exceeds the preset value, and the currently collected NO concentration... x If the concentration value exceeds the preset transient protection threshold by a preset increment, it is determined to be NO.x The concentration value shows a clear upward trend. This condition is used to capture rapidly deteriorating exceedance conditions, triggering protection even if the duration has not yet met condition one. The trend analysis window can be set either as an equivalent to the sliding observation window or as a separate time window independent of it. Its length can be flexibly adjusted according to the dynamic response characteristics of the SCR denitrification system and the sensitivity requirements of transient protection. Setting an independent trend analysis window provides a more suitable time scale for judging concentration change trends, avoiding trend misjudgment or response lag caused by forcibly binding the window length to the stability analysis window.

[0029] If any of the above conditions are met, transient protection will be triggered: the valve opening will be increased by a preset independent protection level (usually smaller than the normal adjustment level, such as 1% of the valve opening) to suppress outlet NO in a rapid response manner. x The concentration further increases. When transient protection is triggered, the over-limit timing and over-limit recording sequence are reset. In practical applications, to avoid excessively frequent protection actions, a minimum interval between transient protection actions can also be set.

[0030] In the specific implementation of step 3, the adjustment cycle is longer than the dynamic response time of the SCR denitrification system, thereby ensuring that the effect of the previous adjustment is fully reflected in the outlet NO. x In terms of concentration.

[0031] In the specific implementation of the stability analysis in step 4, the exit NO within the current sliding observation window is first extracted. x The system collects concentration data and calculates the mean and standard deviation. If the standard deviation is less than or equal to a preset stable standard deviation threshold, the SCR denitrification system is considered to be in a stable state, and no valve opening adjustment is required. If the standard deviation is greater than the preset stable standard deviation threshold, the mean and NO concentration are calculated. x Deviation from the target concentration value: If the positive deviation exceeds the adjustment threshold, the SCR denitrification system is determined to be in a state of continuous high concentration; if the negative deviation exceeds the adjustment threshold, the SCR denitrification system is determined to be in a state of continuous low concentration. This analysis mechanism effectively distinguishes between "true systematic deviation" and "random measurement noise," thereby avoiding invalid valve adjustments caused by random sensor fluctuations. It reduces the frequency of ammonia injection regulating valve operations at the source, while eliminating unnecessary adjustment actions that cause system overshoot and oscillation, ensuring the stable operation of the SCR denitrification system and meeting the outlet NO0 requirement. x While meeting the requirements for concentration control accuracy, the system's operational reliability and economy were also taken into account.

[0032] When the SCR denitrification system is determined to be in a state of continuous high or continuous low error, the adaptive adjustment range for this adjustment is calculated based on the long-term error average of the SCR denitrification system. The specific implementation process is as follows: The adaptive adjustment range is obtained by multiplying the preset basic adjustment range by the adaptive coefficient. When the absolute value of the long-term error mean is greater than the first preset threshold, the adaptive coefficient is set to a value greater than 1. For example, when the absolute value of the long-term error mean is greater than 1.5 times the adjustment threshold, the adaptive coefficient is set to 1.5. When the absolute value of the long-term error mean is less than the second preset threshold, the adaptive coefficient is set to a value less than 1. For example, when the absolute value of the long-term error mean is less than 0.5 times the adjustment threshold, the adaptive coefficient is set to 0.7. When the absolute value of the long-term error mean is between the second preset threshold and the first preset threshold, the adaptive coefficient is set to 1. For example, when the absolute value of the long-term error mean is greater than or equal to 0.5 times the adjustment threshold and less than or equal to 1.5 times the adjustment threshold, the adaptive coefficient is set to 1.

[0033] Once the adaptive adjustment range is calculated, the valve opening can be adjusted incrementally in either a positive (increase) or negative (decrease) direction based on the direction determined by stability analysis, and the time and direction count of this adjustment are updated and recorded. Through dynamic adjustment of the adaptive adjustment range, precise matching of ammonia injection adjustment force under different deviation conditions is achieved: when the system has a persistent large deviation, the adjustment range is amplified to quickly correct the deviation and avoid excessive emissions caused by long-term deviation from the target value; when the system has only a small steady-state deviation, the adjustment range is reduced to avoid system oscillations caused by over-adjustment. This ensures control accuracy while further reducing the valve's operating frequency and wear risk, significantly improving the control stability and adaptability of the large hysteresis process in SCR denitrification.

[0034] It is important to note that the final calculated adaptive adjustment range is constrained by the preset upper and lower limits of the adjustment range. That is, when the calculated adaptive adjustment range is lower than the lower limit, the lower limit is directly used as the corresponding adjustment range. When the calculated adaptive adjustment range is higher than the upper limit, the upper limit is directly used as the corresponding adjustment range, so as to ensure that the adjustment action is neither too weak (ineffective) nor too aggressive (causing oscillation).

[0035] As a specific implementation method, the key technical parameters, their meanings, and value ranges involved in the ammonia injection adaptive control method for large hysteresis in SCR denitrification in this embodiment are shown in Table 1 below. The parameter values ​​in Table 1 are all exemplary descriptions, and can be adjusted according to the dynamic characteristics, emission standards, and operating conditions of the specific SCR system in actual applications.

[0036] Table 1

[0037] The following further elaborates on the ammonia injection adaptive control method for large lag in SCR denitration in this embodiment in combination with specific examples.

[0038] The application scenario in the example is set as follows: an SCR denitration system supporting an industrial waste gas incinerator, with a flue gas treatment volume of approximately 100,000 Nm³ / h, and the inlet NO x concentration fluctuates periodically within the range of 150 ppm to 450 ppm (the fluctuation period is about 10 minutes, affected by intermittent discharging of upstream processes and double flue duct commutation), and the emission standard requires that the outlet NO x concentration does not exceed 50 ppm.

[0039] The system configuration is set as follows: the SCR denitration reactor is filled with vanadium-based honeycomb catalysts, and the ammonia injection reducing agent is an aqueous ammonia solution with a concentration of 20%. The ammonia injection regulating valve is a pneumatic regulating valve, and the valve opening adjustment range is 15% to 95%. The outlet NO x concentration is collected in real time through the CEMS online monitoring system, and the sampling period is 1 second. The control system is deployed on the DCS platform.

[0040] Control parameter settings: the target outlet NO x concentration is 45 ppm, the basic valve opening is 60%, the adjustment threshold is ±4 ppm, the adjustment period is 600 seconds, the basic adjustment amplitude is 2.0%, the transient protection threshold is 50 ppm, the transient protection response time is 240 seconds, the transient protection amplitude is 1.0%, the stable standard deviation threshold is 1.2 ppm, the lower limit of the adjustment amplitude is 1.0%, the upper limit of the adjustment amplitude is 4.0%, the observation window length is 120 seconds, and the trend analysis window length is 180 seconds.

[0041] Under the above example conditions, the control effects of using a fixed valve, traditional PID control, and the method of this embodiment are compared as shown in Table 2 below.

[0042] Table 2

[0043] As can be seen from Table 2, under the condition of large fluctuations in the inlet NO x concentration, the method of this embodiment can effectively suppress the fluctuation amplitude of the outlet NO x concentration, and its standard deviation improvement effect is better than that of traditional PID control, while the number of valve actions is only less than one-tenth of that of PID control. When the outlet NO x concentration shows a large instantaneous overshoot (such as exceeding 50 ppm), the transient protection mechanism can respond and apply a correction action within 4 minutes, effectively preventing the continuous deterioration of the overshoot condition.

[0044] Example 2 Based on the ammonia injection adaptive control method for large lag in SCR denitrification in Example 1, this example discloses an ammonia injection adaptive control system for large lag in SCR denitrification, referencing... Figure 2 The ammonia injection adaptive control system includes an outlet NO x The system includes an online concentration monitoring module, a time base and timing module, a transient protection module, an adjustment trigger judgment module, a stability analysis module, an adaptive adjustment range calculation module, a base value maintenance management module, and a valve opening command output module. Specifically: Export NO x The online concentration monitoring module is used to collect NO concentration data at the outlet of the SCR denitrification reactor in real time. x Concentration data, as feedback input to the control system, typically has a sampling period of 1 second, consistent with the sampling frequency of the DCS system.

[0045] The time reference and timing module is used to provide a unified time reference for the system, record the occurrence time of each normal valve adjustment and transient protection action, and provide a time basis for judging the adjustment cycle and calculating the duration of transient protection.

[0046] The stability analysis module is used to calculate the outlet NO within a specified time window (e.g., the most recent 2 minutes). x The mean and standard deviation of the concentration are used to determine whether the system is currently in a "stable", "continuously high" or "continuously low" state based on the deviation of the standard deviation and the mean. This module is the core judgment link to distinguish between "true deviation of the system" and "random noise fluctuation".

[0047] The transient protection module is independent of the normal adjustment logic and is specifically designed to monitor NO at the outlet. x Whether the concentration instantaneously exceeds the standard. This module uses a dual-condition joint judgment mechanism: condition one is the duration of the exceedance (e.g., continuous exceedance for more than 4 minutes), and condition two is the growth trend during the exceedance period (e.g., the export NO concentration in the most recent period). x (The concentration shows a significant upward trend and exceeds the preset increase). When any of the conditions is met and there is a sufficient interval since the last transient protection action, a fast protection response independent of normal adjustment is triggered.

[0048] The adaptive adjustment range calculation module dynamically adjusts the valve opening range based on the error information accumulated during long-term system operation. When the long-term deviation is large, it automatically increases the adjustment range to accelerate correction; when the long-term deviation is small, it automatically decreases the adjustment range to avoid over-adjustment. The adjustment range is constrained by preset upper and lower limits.

[0049] The adjustment trigger judgment module uses the status output of the comprehensive stability analysis module, the time interval since the last adjustment, and the adaptive adjustment range to determine whether valve adjustment needs to be performed at the current moment. When both conditions are met simultaneously, namely "the time since the last adjustment has reached the adjustment cycle" and "stability analysis determines that the state is unstable", an adjustment action is triggered.

[0050] The base value maintenance management module is used to maintain the current valve opening (initial value is a preset fixed base value) when the system is in a stable state or before the adjustment cycle is reached, without issuing any adjustment commands, so as to achieve the lowest frequency of valve operation.

[0051] The valve opening command output module is used to summarize the normal adjustment command, transient protection command and base value holding status, output the final valve opening target value, and send it to the ammonia injection control valve actuator after amplitude limiting.

[0052] In this embodiment, the export NO x The specific working processes and principles of the online concentration monitoring module, time base and timing module, transient protection module, adjustment trigger judgment module, stability analysis module, adaptive calculation module for adjustment range, base value maintenance management module, and valve opening command output module are the same as those in Example 1, and therefore will not be repeated in this example. Each unit module can be implemented entirely or partially through software, hardware, or a combination thereof. Each unit module can be embedded in the processor of a computer device in hardware form or independent of it, or it can be stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each of the above unit modules.

[0053] The above description is only a preferred embodiment of the present invention and does not limit the scope of protection of the present invention. All equivalent structural transformations made under the inventive concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the scope of protection of the present invention.

Claims

1. An adaptive ammonia injection control method for large hysteresis in SCR denitrification, characterized in that, Includes the following steps: Step 1: Real-time collection of NO at the outlet of the SCR denitrification reactor x The concentration value is added to a sliding observation window of a preset length, and the NO value is updated. x Long-term mean error of concentration; Step 2, based on the currently collected NO x The concentration value is used to check the transient protection conditions to determine whether the transient protection conditions are met. If so, execute the transient protection action, adjust the valve opening based on the preset independent protection amplitude and generate a valve control command, then proceed to step 5; Otherwise, proceed to step 3; Step 3: Determine whether the time interval since the last valve adjustment has reached the preset adjustment cycle. If so, proceed to step 4; Otherwise, after generating a valve control command to maintain the current valve opening, proceed to step 5; Step 4, for NO within the sliding observation window x Stability analysis was performed on the concentration data, and a stability status determination was conducted. If the state is determined to be stable, after generating a valve control command to maintain the current valve opening, proceed to step 5; If the condition is determined to be continuously high, the adaptive adjustment range is calculated based on the long-term error mean, and a valve control command to increase the valve opening is generated accordingly before proceeding to step 5. If the condition is determined to be a persistently low state, the adaptive adjustment range is calculated based on the long-term error mean, and a valve control command to reduce the valve opening is generated accordingly before proceeding to step 5. Step 5: After outputting the valve control command to the ammonia injection regulating valve actuator, return to step 1 to enter the next control cycle.

2. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 1, characterized in that, In step 2, the transient protection condition check adopts a dual-condition joint judgment; if either condition is met, it is determined that the transient protection condition is met. Condition 1: The currently collected NO x The concentration value exceeds the preset transient protection threshold, and the duration of the exceeding state reaches the preset transient protection response time; Condition 2: The currently collected NO x The concentration value exceeds the preset transient protection threshold, and within the trend analysis window, NO x The concentration values ​​showed a clear upward trend.

3. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 2, characterized in that, The final value of the trend analysis window increases by more than a preset value compared to the initial value, and the currently collected NO... x If the concentration value exceeds the preset transient protection threshold by a preset increment, it is determined to be NO. x The concentration values ​​showed a clear upward trend.

4. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 1, 2, or 3, characterized in that, In step 3, the adjustment cycle is greater than the dynamic response time of the SCR denitrification system.

5. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 1, 2, or 3, characterized in that, In step 4, the process of determining the stability state is as follows: Calculate NO within the sliding observation window x Standard deviation and mean of concentration values; If the standard deviation is less than or equal to a preset stable standard deviation threshold, it is determined to be in a stable state; If the standard deviation is greater than a preset stable standard deviation threshold, calculate the mean and NO. x Deviation of concentration target value: If the deviation is positive and exceeds the adjustment threshold, it is determined to be a continuously high state; if the deviation is negative and exceeds the adjustment threshold, it is determined to be a continuously low state.

6. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 1, 2, or 3, characterized in that, In step 4, the process of calculating the adaptive adjustment amplitude based on the long-term error mean is as follows: The adaptive adjustment range is obtained by multiplying the preset basic adjustment range by the adaptive coefficient; When the absolute value of the long-term error mean is greater than the first preset threshold, the adaptive coefficient takes a value greater than 1. When the absolute value of the long-term error mean is less than the second preset threshold, the adaptive coefficient takes a value less than 1. When the absolute value of the long-term error mean is between the second preset threshold and the first preset threshold, the adaptive coefficient is 1.

7. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 6, characterized in that, In step 4, the adaptive adjustment range is constrained by a preset upper limit and lower limit of the adjustment range.

8. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 7, characterized in that, In step 2, the independent protection amplitude is less than the preset basic adjustment amplitude.

9. The adaptive ammonia injection control method for large hysteresis in SCR denitrification according to claim 1, 2, or 3, characterized in that, In step 1, the length of the sliding observation window covers the dynamic response time scale of the SCR denitrification system.

10. An adaptive ammonia injection control system for large hysteresis in SCR denitrification, characterized in that, The ammonia injection adaptive control system, using any one of claims 1 to 9, comprises: Export NO x The online concentration monitoring module is used to collect NO concentration data at the outlet of the SCR denitrification reactor in real time. x Concentration data; The time reference and timing module is used to provide a unified time reference for the system and record the occurrence time of each normal valve adjustment and transient protection action; The transient protection module is used to perform transient protection condition checks and trigger independent transient protection actions through a combined determination of two conditions. The adjustment trigger judgment module is used to determine whether the time interval since the last valve adjustment action has reached the preset adjustment cycle; The stability analysis module is used to analyze NO within the sliding observation window. x Statistical analysis of concentration data is performed to determine the system's operating status; The adaptive adjustment range calculation module is used to calculate the adaptive adjustment range of the valve opening based on the long-term error average. The base value maintenance management module is used to maintain the current valve opening unchanged when the system is stable or the adjustment cycle has not been reached; The valve opening command output module is used to summarize various control commands and output the valve opening target value after limiting to the ammonia injection regulating valve actuator.

Images