A CFB boiler flue gas denitration automatic control method and system

Abstract

The application discloses a kind of CFB boiler flue gas denitration automatic control method, the method comprises the following steps: S1, real-time acquisition and record the reaction parameter in CFB boiler at different time as determination data;S2, the determination data is substituted into denitration control model and is output denitration control amount;S3, according to the denitration control amount to ammonia injection regulating valve is adjusted control actual ammonia injection amount;S4, measure actual denitration efficiency as basis to the denitration control model is optimized correction;The application also discloses a kind of CFB boiler flue gas denitration automatic control system.The application is modeled by using grey model to historical denitration big data, combined with Markov prediction model, can accurately predict data in future monitoring period, while cooperating with the denitration control model calculated accurately denitration control amount, to carry out accurate regulation control to ammonia injection amount, guarantee denitration efficiency quality.

Description

Technical Field

[0001] This invention relates to the field of flue gas denitrification technology, and more specifically, to an automatic control method and system for flue gas denitrification in CFB boilers. Background Technology

[0002] With increasingly stringent environmental standards, circulating fluidized bed (CFB) boilers, relying solely on their inherent characteristics, are no longer sufficient to meet flue gas emission requirements. Selective non-catalytic reduction (SNCR) denitrification technology, however, offers advantages such as low cost, small footprint, and simple facilities, making it suitable for denitrification retrofitting of older power plants and small to medium-sized units. The overall denitrification efficiency of SNCR denitrification systems in CFB boilers can reach 50%–70%, meeting current environmental standards. However, due to factors such as variations in coal quality, unstable operating conditions, and inadequate operation, various problems arise in actual operation, making it difficult to achieve emission standards.

[0003] Selective non-catalytic reduction (SNCR) is a technology that does not use a catalyst. A reducing agent (such as ammonia, liquid ammonia, or urea) is injected into the boiler furnace at a temperature range of 800–1100°C. The reducing agent rapidly decomposes into NH3 at high temperatures and reacts with NO in the flue gas. x The non-nitrification reaction (SNCR) is a process that produces harmless N2 and H2O through a reduction reaction. The SNCR system consists of a reducing agent storage tank, a multi-layer reducing agent injection device, and a corresponding control system. To ensure the best reduction effect with minimal ammonia injection, to ensure good mixing of the reducing agent with the flue gas, to reduce the probability of blockage and corrosion of the rotary air preheater, and to ensure the full reaction of the injected NH3, the reducing agent is injected at multiple points into the most effective parts of the furnace according to the temperature changes caused by boiler load variations.

[0004] However, precise and reasonable control of ammonia injection flow rate is a prerequisite for the efficient operation of the SNCR process. Because NH3 and NO... x The reaction is a process with a large delay and a large inertia, coupled with NO x The filtration-continuous sampling-condensation-secondary filtration-chemical analysis process of the online monitoring instrument CEMS also exhibits significant delays. In such systems with large delays, after the ammonia injection flow rate control valve is changed, there is always a long delay before it affects the NO₂. x As the monitored parameters change, traditional PID control often exhibits significant lag, overshoot, and even divergent characteristics. Furthermore, boiler flue gas NO... x The change in content is a very complex process, which is greatly affected by parameters such as unit load, air volume, coal quantity, coal type, and SOFA damper opening at the top of the low-NOx burner. These parameters all have different delay characteristics, and traditional on-site control methods such as manual control and PID automatic control are difficult to solve such industrial control problems.

[0005] However, the prior art has not yet proposed an effective solution to the problems. SUMMARY

[0006] In view of the problems in the prior art, the present application provides a CFB boiler flue gas denitration automatic control method and system to overcome the above technical problems existing in the prior art.

[0007] To this end, the application adopts the following specific technical solutions:

[0008] According to one aspect of the application, a CFB boiler flue gas denitration automatic control method is provided, which comprises the following steps:

[0009] S1, real-time collection and recording of reaction parameters in CFB boiler at different times as measurement data;

[0010] S2, substituting the measurement data into a denitration control model for prediction and output of a denitration control amount;

[0011] S3, adjusting and controlling the actual ammonia injection amount according to the denitration control amount;

[0012] S4, measuring the actual denitration efficiency as a basis for optimizing and correcting the denitration control model.

[0013] Further, the real-time collection and recording of reaction parameters in CFB boiler at different times as measurement data comprises the following steps:

[0014] S11, setting and real-time monitoring of the CFB boiler according to a monitoring period;

[0015] S12, measuring the bed temperature, air volume and coal supply amount parameters in the CFB boiler to determine the boiler load;

[0016] S13, measuring the concentration of NOx gas in the flue gas in the CFB boiler as the initial concentration; x

[0017] S14, obtaining the opening degree of the ammonia injection regulating valve in the current monitoring period;

[0018] S15, obtaining and recording the ammonia injection amount under the current monitoring period and the current opening degree as the initial ammonia injection amount;

[0019] S16, taking the boiler load, the initial concentration, the opening degree and the initial ammonia injection amount in the same monitoring period as the measurement data of a certain monitoring period.

[0020] Further, substituting the measurement data into a denitration control model for prediction and output of a denitration control amount comprises the following steps:​

[0021] S21, a grey Markov prediction model is constructed in combination with historical denitration concentration big data;

[0022] S22, an ammonia injection adjustment model is constructed in combination with a cerebellum model neural network of historical ammonia injection amount big data;

[0023] S23, a denitration control model meeting a denitration control strategy is constructed by combining the grey Markov prediction model and the denitration control neural network;

[0024] S24, the measured data are sequentially substituted into the denitration control model to output a denitration control amount.

[0025] Further, the grey Markov prediction model constructed in combination with historical denitration concentration big data comprises the following steps:

[0026] S211, a grey prediction model is constructed with historical denitration concentration big data as basic data;

[0027] S212, simulation data sets are input into the grey model for simulation prediction, and simulation prediction results are output;

[0028] S213, according to the simulation prediction results, the relative error between the true value and the predicted value is calculated, and the state interval is divided by the error range;

[0029] S214, the state transition probability matrix is formed by combining the probability of the current state being transferred to the next state after several steps;

[0030] S215, the state at the next time is predicted according to the current state and the state transition matrix.

[0031] Further, the ammonia injection adjustment model constructed in combination with the cerebellum model neural network of historical ammonia injection amount big data comprises the following steps:

[0032] S221, an initial model is constructed by using a Gaussian function as a base function of the cerebellum model neural network;

[0033] S222, historical ammonia injection amount big data is expanded and divided into a training set and a test set;

[0034] S223, the training set is substituted into the initial model for training to obtain an ammonia injection adjustment model;

[0035] S223, the test set is used to test, train and verify the ammonia injection adjustment model.

[0036] Further, the measured data are sequentially substituted into the denitration control model to output a denitration control amount, which comprises the following steps:

[0037] S241, the boiler load, the initial concentration and the opening degree in the measured data of the current monitoring period are substituted into the ammonia injection adjustment model to output results as the basic ammonia injection amount;

[0038] S242, the initial concentration in the measured data of the current monitoring period is substituted into the gray Markov prediction model to obtain a concentration prediction result of the next monitoring period;

[0039] S243, the difference between the concentration prediction result and the initial concentration in the current monitoring period is calculated, and the corresponding compensation ammonia injection amount is calculated in combination with the working condition parameters of the CFB boiler;

[0040] S244, the sum of the basic ammonia injection amount and the compensation ammonia injection amount is multiplied by a correction coefficient and a correction coefficient to obtain an actual ammonia injection amount of the next monitoring period, and the actual ammonia injection amount is used as a denitration control amount.

[0041] Further, the actual ammonia injection amount is adjusted and controlled by the ammonia injection adjustment valve using a cascade control strategy, the main parameter is the initial concentration, the secondary parameter is the opening degree, and the reference amount is the denitration control amount.

[0042] Further, the actual denitration efficiency is measured as the basis for optimizing and correcting the denitration control model, including the following:

[0043] S41, the concentration of NO x gas at the actual outlet is measured in real time to calculate the actual denitration efficiency;

[0044] S42, the actual denitration efficiency of the current monitoring period is substituted into the correction formula to calculate a correction coefficient of the next monitoring period;

[0045] S43, the correction coefficient is substituted into the denitration control model to calculate the denitration control amount of the next monitoring period;

[0046] S44, when the actual denitration efficiency of a plurality of consecutive monitoring periods is lower than a preset safety threshold, the system issues a warning, manually checks the denitration process, and modifies the correction coefficient.

[0047] Further, the correction formula is:

[0048]

[0049] In the formula, η represents the correction coefficient;

[0050] S represents the actual denitration efficiency of the current monitoring period;

[0051] S min represents the minimum value of the preset denitration efficiency standard;

[0052] S max This indicates the maximum value of the preset denitrification efficiency standard.

[0053] According to another aspect of the present invention, an automatic control system for flue gas denitrification of a CFB boiler is also provided, the system comprising: a CFB boiler body, a real-time monitoring unit, an automatic control unit, an ammonia injection regulating valve, and a monitoring and control center;

[0054] The CFB boiler body is used to provide a denitrification and reduction environment for flue gas.

[0055] The real-time monitoring unit is used to acquire boiler parameters and NO in flue gas in real time. x Gas concentration;

[0056] The automatic control unit is used to achieve adaptive adjustment and control of ammonia injection based on monitoring data;

[0057] The ammonia injection regulating valve is used to change the opening degree to control the amount of ammonia injected;

[0058] The monitoring and control center is used to realize remote monitoring and parameter adjustment input of the system.

[0059] The beneficial effects of this invention are as follows: By using a grey model to model historical denitrification big data and combining it with a Markov prediction model, it is possible to accurately predict data within future monitoring periods. Simultaneously, an adaptive denitrification control model calculates precise denitrification control quantities, thereby accurately adjusting and controlling the ammonia injection rate. This eliminates the impact of time lags such as monitoring signals and reaction delays on the reductive denitrification process of SNCR technology in the CFB boiler, thus ensuring denitrification efficiency and quality. Furthermore, based on real-time monitoring of operating parameters within the CFB boiler and the NO content of the flue gas entering and exiting the boiler... x The gas concentration is calculated in real time to determine the actual denitrification efficiency. This feedback correction command is then fed back to the denitrification control model to further improve the accuracy of automated control. It can adaptively and dynamically adjust according to the actual operating conditions, avoiding the impact of sudden changes in a certain parameter within the boiler on the efficiency and effect of reductive denitrification. By adding an automatic early warning system, it can promptly remind manual inspection and troubleshooting in the event of automatic control errors during long-term operation, minimizing system losses and hazards, and thus improving the safety and stability of the automated control system. In addition, the system's signal feedback command transmission efficiency is timely and fast, effectively solving the problems of signal lag and large inertia in traditional denitrification processes. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments 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 these drawings without creative effort.

[0061] Figure 1 This is a flowchart of an automatic control method for flue gas denitrification in a CFB boiler according to an embodiment of the present invention;

[0062] Figure 2 This is a system block diagram of an automatic control system for flue gas denitrification of a CFB boiler according to an embodiment of the present invention.

[0063] In the picture:

[0064] 1. CFB boiler body; 2. Real-time monitoring unit; 3. Automatic control unit; 4. Ammonia injection regulating valve; 5. Monitoring and control center. Detailed Implementation

[0065] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0066] According to an embodiment of the present invention, an automatic control method for flue gas denitrification in a CFB boiler is provided.

[0067] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, the automatic control method for CFB boiler flue gas denitrification according to an embodiment of the present invention includes the following steps:

[0068] S1. Real-time acquisition and recording of reaction parameters inside the CFB boiler at different times as measurement data;

[0069] The real-time acquisition and recording of reaction parameters inside the CFB boiler at different times as measurement data includes the following steps:

[0070] S11. Set and monitor the CFB boiler in real time according to the monitoring cycle;

[0071] S12. Measure the bed temperature, air volume and coal feed rate parameters in the CFB boiler to determine the boiler load;

[0072] S13. Measure the NO content in the flue gas inside the CFB boiler.x the concentration of the gas as an initial concentration;

[0073] S14, obtaining the opening and closing degree of the ammonia injection valve in the current monitoring period;

[0074] S15, obtaining and recording the amount of ammonia injection under the current monitoring period and the current opening and closing degree as the initial ammonia injection amount;

[0075] S16, taking the boiler load, the initial concentration, the opening and closing degree and the initial ammonia injection amount in the same monitoring period as the measurement data of a certain monitoring period.

[0076] S2, substituting the measurement data into the denitration control model for prediction and outputting the denitration control amount;

[0077] Wherein, substituting the measurement data into the denitration control model for prediction and outputting the denitration control amount includes the following steps:

[0078] S21, constructing a gray Markov prediction model in combination with historical denitration concentration big data;

[0079] The historical denitration concentration big data is selected from the process of denitration of CFB boiler using SNCR technology, and the NOx concentration at the inlet of the boiler is monitored. x The change data of the gas concentration is also arranged in sequence according to the period.

[0080] Markov prediction method is a probability prediction method about event occurrence. It is a prediction method for predicting the change status at each time (or period) in the future according to the current status of the event. The state of future things is predicted by the initial probability of different states of things and the transition probability of states.

[0081] Wherein, the gray Markov prediction model constructed in combination with the historical denitration concentration big data includes the following steps:

[0082] 0083.S211, constructing a gray prediction model based on historical denitration concentration big data as basic data;

[0083] 0084. Set the historical denitration concentration big data in a certain period of time as the original sequence, and the expression is: And the sequence is negative, and the first accumulation data sequence is obtained from the original sequence, that is, The first-order linear differential equation of the gray prediction model can be expressed as: Wherein a represents the development coefficient, and the size and sign of a reflect the development trend of the original sequence and the first accumulation data sequence, and b represents the gray action amount.

[0084] 0085. The values ​​of a and b can be obtained through the least squares method and the constructed matrix, and a grey prediction model can be constructed based on this.

[0085] S212. Input the simulation dataset into the gray model to perform simulation prediction, and output the simulation prediction results;

[0086] S213. Based on the simulation prediction results, calculate the relative error between the actual value and the predicted value, and divide the state interval by the error range;

[0087] The state interval can be represented as: E i =[L i H i ], where L i H represents the lower limit. i Indicates the upper limit.

[0088] S214. The combination of probabilities that the current state will suddenly transition to the next state after several steps (k steps) forms the state transition probability matrix.

[0089] Let sequence x t In state E i The state probability is p i (t)=p(x t =i), if sequence x t From state E i Transition to the next time state E j The transition probability is p ij Then p ij =p(x t+1 =j|x t Let $\mathbf{i}$ be the probability of transitioning from state $i$ at time $t$ to state $j$ at time $t+1$. If the value at time $t+1$ depends only on the value at time $t$ and the transition probability, then this time series is called a Markov chain. ij Let be the one-step transition probability of the Markov chain at time t.

[0090] State E i Transition to state E j The number of times is m ij State E i The number of times it appears is M i Then state E i Transfer to E j The probability is p ij The transition probability of k steps is called the probability of reaching state j from state i in k steps at time t. Therefore, the k-step transition probability matrix is ​​a matrix composed of the k-step transition probabilities.

[0091] S215, predicting the state of next time according to the current state and the state transition matrix. S22, constructing an ammonia injection adjustment model in combination with a history ammonia injection big data cerebellum model neural network;

[0092] The history ammonia injection big data includes specific ammonia injection amount and NOx x The initial concentration of the gas and the load of the CFB boiler and other parameters need to inject different doses of reducing agent for denitration in different boiler environments, so the parameters corresponding to different ammonia injection amounts are also different.

[0093] The constructing of the ammonia injection adjustment model in combination with the history ammonia injection big data cerebellum model neural network comprises the following steps:

[0094] S221, constructing an initial model by using a Gaussian function as a base function of the cerebellum model neural network;

[0095] The expression of the initial model is:

[0096]

[0097]

[0098] In the formula, N x represents the dimension of input x j , u k,j represents the center of the base function, and sigma k,j represents the variance of the base function, and k represents the input quantization technology determined by the mapping accuracy.

[0099] S222, expanding the history ammonia injection big data and dividing it into a training set and a test set;

[0100] S223, substituting the training set into the initial model to obtain an ammonia injection adjustment model;

[0101] S223, testing, training and verifying the ammonia injection adjustment model by using the test set.

[0102] S23, constructing a denitration control model meeting the denitration control strategy by combining the gray Markov prediction model and the denitration control neural network;

[0103] S24, substituting the measured data into the denitration control model in sequence to output a denitration control amount.

[0104] The substituting of the measured data into the denitration control model in sequence to output a denitration control amount comprises the following steps:

[0105] S241, the boiler load, the initial concentration and the opening degree in the determination data in the current monitoring period are substituted into the ammonia injection adjustment model to output results as the basic ammonia injection amount;

[0106] S242, the initial concentration in the determination data in the current monitoring period is substituted into the gray Markov prediction model to obtain a concentration prediction result in the next monitoring period;

[0107] S243, the difference between the concentration prediction result and the initial concentration in the current monitoring period is calculated, and the corresponding compensation ammonia injection amount is calculated in combination with the working condition parameters of the CFB boiler;

[0108] S244, the sum of the basic ammonia injection amount and the compensation ammonia injection amount is multiplied by a correction coefficient and a correction coefficient to obtain an actual ammonia injection amount in the next monitoring period, and the actual ammonia injection amount is used as a denitration control amount.

[0109] S3, the ammonia injection adjustment valve is adjusted and controlled according to the denitration control amount to control the actual ammonia injection amount;

[0110] According to the denitration control amount, the ammonia injection adjustment valve is adjusted and controlled to control the actual ammonia injection amount, and the main parameter is the initial concentration, the secondary parameter is the opening degree, and the reference amount is the denitration control amount.

[0111] S4, the actual denitration efficiency is measured as a basis for optimizing and correcting the denitration control model.

[0112] Wherein, the actual denitration efficiency is measured as a basis for optimizing and correcting the denitration control model, including the following:

[0113] S41, the concentration of NO x gas at the actual outlet is measured in real time to calculate the actual denitration efficiency;

[0114] S42, the actual denitration efficiency in the current monitoring period is substituted into the correction formula to calculate a correction coefficient in the next monitoring period;

[0115] 0116. Wherein, the correction formula is:

[0116]

[0117] In the formula, η represents the correction coefficient;

[0118] S represents the actual denitration efficiency in the current monitoring period;

[0119] S min represents the minimum value of the preset denitration efficiency standard;

[0120] S max represents the maximum value of the preset denitration efficiency standard.

[0121] S43, the correction coefficient is substituted into the denitration control model to calculate the denitration control amount of the next monitoring period;

[0122] S44, when the actual denitration efficiency of the continuous multiple monitoring periods is all lower than the preset safety threshold, the system issues a pre-warning warning, manually checks the denitration process, and modifies the correction coefficient.

[0123] When the actual denitration efficiency appears continuous deviation, at this time, there may be some abnormal parameters, so manual troubleshooting and maintenance are needed to ensure the effective operation of the system, and the main factors affecting the denitration efficiency of the SNCR flue gas denitration system are as follows:

[0124] 1) Boiler excess air coefficient: The excess air coefficient has a great influence on the generation and emission of NO x , the greater the excess air coefficient, the more NO x is generated, and appropriately reducing the excess air coefficient can greatly reduce the generation of NO x ; but the excess air coefficient is affected by many factors such as boiler load, furnace temperature, fuel combustion, etc.

[0125] 2) Unit load: The unit load plays a major role in the generation of NO x , the greater the load, the more fuel, the higher the NO x mass concentration; the regional control deviation (ACE) technology is used in the load distribution of the power grid, so that the load of the unit changes greatly, causing the boiler combustion system to be very unstable, the fuel quantity, air quantity, excess air coefficient, etc. change greatly, and the NO x emission mass concentration fluctuates sharply.

[0126] 3) Installation position of measuring point: The NO x analyzer is generally installed after the induced draft fan and at the inlet of the desulfurization tower. According to the actual test on site, when the unit load changes, the delay time of the corresponding change of the NO x measurement value is between 2-3min, which is quite large.

[0127] 4) Furnace temperature: High furnace temperature will intensify the thermal decomposition of NH3 oxidation and reduce the reducing agent, while increasing the ammonia escape amount; too low furnace temperature will reduce the reduction reaction speed of NH3.

[0128] According to another embodiment of the present application, as Figure 2 shown, there is also provided a CFB boiler flue gas denitration automatic control system, which comprises: a CFB boiler body 1, a real-time monitoring unit 2, an automatic control unit 3, an ammonia injection regulating valve 4, and a supervision and control center 5.

[0129] The CFB boiler body 1 is used for providing a denitration reduction environment of flue gas.

[0130] The real-time monitoring unit 2 is used for acquiring the boiler parameters and the NO x gas concentration in the flue gas in real time.

[0131] The automatic control unit 3 is used for realizing self-adaptive adjustment control of ammonia injection according to the monitoring data.

[0132] The ammonia injection adjustment valve 4 is used for controlling the ammonia injection amount by changing the opening degree.

[0133] The supervision control center 5 is used for realizing remote supervision and parameter adjustment input of the system.

[0134] According to the above technical scheme of the present application, the historical denitration big data is modeled by using the grey model, the future monitoring period data is accurately predicted by combining the Markov prediction model, and the accurate denitration control amount is calculated by cooperating with the self-adaptive denitration control model, so that the ammonia injection amount is accurately adjusted and controlled, the influence of the monitoring signal, the reaction delay and other time lags on the reduction denitration of the SNCR technology in the CFB boiler is eliminated, and the denitration efficiency and quality are ensured. x The real-time monitoring CFB boiler operation parameters and the NO The actual denitration efficiency is calculated in real time to input feedback correction instructions to the denitration control model, the accuracy of the automatic control is further improved, the self-adaptive dynamic adjustment is realized according to the actual operation condition, the reduction denitration efficiency and effect caused by the sudden change of the boiler parameters are avoided, the automatic control error is timely reminded for manual maintenance and troubleshooting in the long-term operation process, the loss and harm of the system are minimized, and the safety and stability of the automatic control system are improved. The signal feedback instruction transmission efficiency is timely and fast, and the problems of signal lag and large inertia in the traditional denitration process are effectively solved.

[0135] The above only describes the preferred embodiments of the present application and should not be used to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A CFB boiler flue gas denitration automatic control method, characterized in that, S1, real-time acquisition and recording of reaction parameters in the CFB boiler at different times as measurement data, including setting and real-time monitoring of the CFB boiler according to a monitoring period; measuring the bed temperature, air volume and coal supply amount in the CFB boiler to determine the boiler load; measuring the concentration of NOx gas in the flue gas in the CFB boiler as the initial concentration; obtaining the opening degree of the ammonia injection regulating valve in the current monitoring period; obtaining and recording the ammonia injection amount under the current monitoring period and the current opening degree as the initial ammonia injection amount; taking the boiler load, initial concentration, opening degree and initial ammonia injection amount in the same monitoring period as the measurement data of a certain monitoring period; S2, substituting the measurement data into the denitration control model for prediction and outputting the denitration control amount, including constructing a gray Markov prediction model combined with historical denitration concentration big data, including constructing a gray prediction model with historical denitration concentration big data as basic data, inputting a simulation data set into the gray model for simulation prediction, and outputting the simulation prediction result; according to the simulation prediction result, the relative error between the true value and the predicted value is calculated, and the state interval is divided through the error range; the probability of transferring the current state to the next state after several steps is combined to form a state transition probability matrix; the state at the next time is predicted according to the current state and the state transition matrix; combining the historical ammonia injection amount big data cerebellum model neural network to construct an ammonia injection adjustment model, including using a Gaussian function as the basis function of the cerebellum model neural network to construct an initial model, expanding and dividing the historical ammonia injection amount big data into a training set and a test set; substituting the training set into the initial model to obtain the ammonia injection adjustment model; testing, training and verifying the ammonia injection adjustment model by using the test set; constructing a denitration control model that meets the denitration control strategy by combining the gray Markov prediction model and the cerebellum model neural network; substituting the measurement data into the denitration control model in sequence to output the denitration control amount, including: substituting the boiler load, initial concentration and opening degree in the current monitoring period measurement data into the ammonia injection adjustment model to output the result as the basic ammonia injection amount; substituting the initial concentration in the current monitoring period measurement data into the gray Markov prediction model to obtain the concentration prediction result of the next monitoring period; calculating the difference between the concentration prediction result and the initial concentration in the current monitoring period, and combining the working condition parameters of the CFB boiler to calculate the corresponding compensation ammonia injection amount; multiplying the sum of the basic ammonia injection amount and the compensation ammonia injection amount by the correction coefficient and the correction coefficient to obtain the actual ammonia injection amount in the next monitoring period, and taking it as the denitration control amount; S3, adjusting and controlling the actual ammonia injection amount according to the denitration control amount; S4, measuring the actual denitration efficiency as the basis for optimizing and correcting the denitration control model.

2. The automatic control method for flue gas denitration of a CFB boiler according to claim 1, characterized in that, The ammonia injection regulating valve is adjusted and controlled according to the denitration control amount to obtain the actual ammonia injection amount, which adopts a cascade control strategy, the main parameter of which is the initial concentration, the secondary parameter is the opening degree, and the reference amount is the denitration control amount.

3. The automatic control method for flue gas denitration of a CFB boiler according to claim 1, characterized in that, The method comprises the following steps: measuring the actual de-nitration efficiency as a basis to optimize and correct the de-nitration control model, measuring the concentration of NOx gas at the actual outlet in real time, calculating the actual de-nitration efficiency, substituting the actual de-nitration efficiency of the current monitoring period into a correction formula to calculate a correction coefficient of the next monitoring period, substituting the correction coefficient into the de-nitration control model to calculate the de-nitration control amount of the next monitoring period, and issuing a pre-warning when the actual de-nitration efficiency of continuous multiple monitoring periods is lower than a preset safety threshold, manually checking the de-nitration process and modifying the correction coefficient.

4. The automatic control method for flue gas denitration of a CFB boiler according to claim 3, characterized in that, The correction formula is: , wherein η represents a correction coefficient; S represents an actual denitration efficiency of a current monitoring period; S min represents a minimum value of a preset denitration efficiency standard; S max represents the maximum value of the preset denitration efficiency standard.

5. A CFB boiler flue gas denitration automatic control system for implementing the CFB boiler flue gas denitration automatic control method of any one of claims 1-4, characterized in that, The system comprises a CFB boiler main body, a real-time monitoring unit, an automatic control unit, an ammonia injection regulating valve and a supervision control center. The CFB boiler main body is used to provide a de-nitration reduction environment for flue gas. The real-time monitoring unit is used to acquire boiler parameters and the concentration of NOx gas in flue gas in real time. The automatic control unit is used to realize self-adaptive adjustment control of ammonia injection according to monitoring data. The ammonia injection regulating valve is used to change the opening degree to control the ammonia injection amount. The supervision control center is used to realize remote supervision and parameter adjustment input of the system.

Images