Denitration NOx control method and device based on intelligent control strategy and electronic equipment

Abstract

The invention relates to a denitration NOx control method and device based on an intelligent control strategy and electronic equipment, and relates to the technical field of power plant automatic control, and the denitration NOx control method comprises the steps that according to a monitoring instrument state signal of a reactor, process reaction parameters corresponding to the reactor are obtained; predicting a first NOx concentration corresponding to the reactor by using different concentration prediction algorithms according to the process reaction parameters; determining a second NOx concentration corresponding to the reactor based on the first NOx concentration and the process reaction parameters through cooperative operation of a concentration prediction model and a state compensator; and performing ammonia flow closed-loop regulation by using the first PID controller based on the second NOx concentration and the ammonia flow feed-forward control instruction, and generating an ammonia flow control instruction corresponding to the reactor. The first PID controller is combined with the second NOx concentration and the ammonia gas flow feed-forward instruction for closed-loop adjustment, the large lag characteristic of the denitration system is counteracted, and the ammonia spraying flow control precision is improved.

Description

Technical Field

[0001] This application relates to the field of power plant automation control technology, specifically to a denitrification NOx control method, device, and electronic equipment based on intelligent control strategies. Background Technology

[0002] During the fuel combustion stage of power plants, large amounts of nitrogen oxides (NOx) are typically produced. NOx not only causes a series of environmental problems such as acid rain, photochemical smog, and ozone pollution, but also poses serious harm to the human respiratory and cardiovascular systems. Effective control of NOx emissions from power plants is crucial. x Emissions are an essential requirement for implementing environmental protection policies and protecting the ecological environment.

[0003] Currently, the NH3 flow rate in the denitrification island is adjusted according to the PID proportional control, and the NO at the chimney inlet... x The pure delay of concentration is about 3 minutes, and the whole response process takes more than ten minutes, which is difficult to control stably. It has a large hysteresis characteristic and cannot guarantee the control quality of the denitrification system. Summary of the Invention

[0004] In view of this, this application provides a denitrification NOx control method, device, and electronic equipment based on an intelligent control strategy. The main objective is to solve the problem of NH3 flow rate regulation based on PID proportional control in denitrification islands and NOx control at the chimney inlet in related technologies. x The pure delay in concentration is about 3 minutes, and the whole response process takes more than ten minutes, making it difficult to control stably. It has a large hysteresis characteristic and is a technical problem that cannot guarantee the control quality of the denitrification system.

[0005] According to the first aspect of this application, a denitrification NOx control method based on an intelligent control strategy is provided, the method comprising: Based on the status signals of the reactor's monitoring instruments, obtain the corresponding process reaction parameters of the reactor; Based on the process reaction parameters, the method uses different concentrations to predict the first NO concentration in the reactor. x concentration; Through collaborative computation using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and process reaction parameters, determine the corresponding second NO in the reactor. x concentration; Using the first PID controller based on the second NO x The concentration and ammonia flow rate feedforward control commands are used to perform closed-loop regulation of ammonia flow rate, generating the corresponding ammonia flow rate control command for the reactor.

[0006] According to a second aspect of this application, a denitrification NOx control device based on an intelligent control strategy is provided, the device comprising: The acquisition module is used to acquire the corresponding process reaction parameters of the reactor based on the status signals of the reactor's monitoring instruments; The prediction module is used to predict the first NO concentration in the reactor based on the process reaction parameters obtained from different concentration prediction methods. x concentration; The determination module is used to perform collaborative calculations using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and process reaction parameters, determine the corresponding second NO in the reactor. x concentration; The generation module is used to utilize the first PID controller based on the second NO x The concentration and ammonia flow rate feedforward control commands are used to perform closed-loop regulation of ammonia flow rate, generating the corresponding ammonia flow rate control command for the reactor.

[0007] According to a third aspect of this application, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of the first aspect described above.

[0008] According to a fourth aspect of this application, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause a computer to perform the method of the first aspect described above.

[0009] The denitrification NOx control method, apparatus, and electronic equipment provided in this application, based on an intelligent control strategy, differs from related technologies in that it obtains the corresponding process reaction parameters of the reactor based on the status signals of the reactor's monitoring instruments; and uses different concentration prediction methods based on the process reaction parameters to predict the first NOx concentration in the reactor. x Concentration; through collaborative computation of a concentration prediction model and a state compensator, based on the first NO x Based on the concentration and process reaction parameters, determine the corresponding second NO in the reactor. x Concentration; using a first PID controller based on a second NO concentration. x Ammonia flow rate is closed-loop regulated using feedforward control commands based on concentration and flow rate, generating corresponding ammonia flow rate control commands for the reactor. In this way, this application can verify the validity of data through the status signals of the reactor's monitoring instruments, obtain valid process reaction parameters, ensure reliable control input data, avoid regulation errors caused by invalid data, and use different concentration prediction methods to predict the first NO... x Concentration, adaptable to various operating conditions, improves prediction coverage and fault tolerance. Through collaborative calculation between the concentration prediction model and the state compensator, based on the first NO xThe concentration was further corrected to correct for prediction biases caused by fluctuations in operating conditions, resulting in the second NO concentration. x The concentration effectively counteracts the large hysteresis characteristic of the denitrification system, and finally, the first PID controller is combined with the second NO concentration. x Closed-loop regulation of concentration and ammonia flow rate feedforward commands improves the accuracy of ammonia injection flow control and ensures the output NO content. x The concentration is stable and meets the standards, enabling intelligent and precise control of the denitrification system and improving the control quality of the denitrification system. Attached Figure Description

[0010] 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.

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0012] Figure 1 A schematic flowchart illustrating a denitrification NOx control method based on an intelligent control strategy provided in an embodiment of this application; Figure 2 An example flowchart provided for an embodiment of this application; Figure 3 This is a schematic diagram of a denitrification NOx control device based on an intelligent control strategy, provided as an embodiment of this application. Detailed Implementation

[0013] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0014] The following description, with reference to the accompanying drawings, describes a method, apparatus, and electronic device for NOx denitrification based on an intelligent control strategy, according to embodiments of this application.

[0015] This application provides a denitrification NOx control method, device, and electronic equipment based on an intelligent control strategy. The main objective is to solve the problem of NH3 flow rate regulation based on PID proportional control in denitrification islands and NOx control at the chimney inlet in related technologies. xThe pure delay of the concentration is about 3 minutes, and the full response process takes more than ten minutes. It is difficult to control stably, has the characteristic of large hysteresis, and cannot guarantee the control quality of the denitration system.

[0016] As Figure 1 shown, an embodiment of the present application provides a denitration NOx control method based on an intelligent control strategy, including: Step 101: Obtain the process reaction parameters corresponding to the reactor according to the monitoring instrument status signal of the reactor.

[0017] Among them, the reactor can be the core equipment of the denitration system, the Selective Catalytic Reduction (SCR) reactor. The flue gas and ammonia undergo a selective catalytic reduction reaction under the action of the catalyst in the reactor to achieve NO x removal.

[0018] In some embodiments, the monitoring instrument status signal can be the signal collected by various online monitoring instruments supporting the denitration reactor, such as the signal of the instrument's own operating status and the process parameter measurement signal collected by the instrument, to verify whether the instrument is normal, and then obtain the process reaction parameters collected by the normal instrument, eliminating invalid data in faults, drifts, and calibrations, making the process reaction parameters for concentration prediction true and effective, and avoiding the spray ammonia regulation error caused by the interference of invalid data on the control logic. Among them, the process reaction parameters can be the key operating parameters affecting the denitration reaction efficiency and reaction rate in the reactor, which can be divided into the flue gas side, equipment side, spray ammonia coordination side, etc., such as the SCR inlet NO x concentration, SCR outlet NO x concentration, total coal feeding amount, total air volume, compliance instruction, etc. On the equipment side, there are the pressure difference across the catalyst layer and catalyst activity. On the spray ammonia coordination side, there are ammonia gas flow rate (spray ammonia flow rate), ammonia-nitrogen molar ratio, spray ammonia grid pressure, etc.

[0019] Step 102: Use different concentration prediction algorithms to predict the first NO x concentration corresponding to the reactor according to the process reaction parameters.

[0020] Among them, the first NO x concentration can be the NO concentration at the reactor outlet predicted by different concentration prediction algorithms. x concentration.

[0021] In some embodiments, multiple different concentration prediction algorithms can be configured for the reactor, such as the mechanism model prediction algorithm, LSTM data-driven prediction algorithm, Automatic Generation Control (AGC) condition prediction algorithm, etc. Use multiple different concentration prediction algorithms to predict the corresponding first NO respectively based on the process reaction parameters at the current moment.x Concentration, to determine the first NO x Whether the concentration meets the emission requirements allows for advance ammonia injection adjustment using primary and secondary PID controllers and feedforward controllers, adapting to different chemical conditions, and using different algorithms to output multiple sets of first NO values. x Concentration can provide a multivariate benchmark for subsequent bias correction, avoiding prediction bias caused by the failure of a single algorithm and improving the reliability of predicted concentration.

[0022] Step 103: Through collaborative calculation using the concentration prediction model and the state compensator, based on the first NO... x Based on the concentration and process reaction parameters, determine the corresponding second NO in the reactor. x concentration.

[0023] In some embodiments, the concentration prediction model can be used to predict the first NO concentration at the current moment based on historical operating conditions. x NO concentration and process reaction parameters corresponding to the outlet NO of the denitrification reactor x Concentration, calculated using different algorithms, for the first NO x Predictions are made based on concentration. A state compensator can be used to receive the second NO from the concentration prediction model output. x Concentration, combined with the first NO x The concentration and real-time measured values ​​from the Continuous Emissions Monitoring System (CEMS) are used to calculate the deviation compensation coefficient and output NO. x Concentration compensation values ​​track changes in operating conditions in real time, compensating for the shortcomings of fixed parameters in concentration prediction models that make it difficult to adapt to dynamic operating conditions. This improves the adaptability to dynamic operating conditions and thus enhances prediction accuracy. Correspondingly, the second NO... x The concentration can be obtained by using a concentration prediction model and a state compensator for the first NO. x The reactor outlet NO concentration after deviation correction x The predicted concentration value, compared to the first NO x The concentration more closely matches real-time operating conditions, used to predict outlet NO x Changes in concentration can be addressed by adjusting the ammonia injection rate in advance to offset the significant lag in the system.

[0024] Step 104: Using the first PID controller based on the second NO x The concentration and ammonia flow rate feedforward control commands are used to perform closed-loop regulation of ammonia flow rate, generating the corresponding ammonia flow rate control command for the reactor.

[0025] In some embodiments, the first PID controller may refer to a core closed-loop controller used for regulating the ammonia injection flow rate in the denitrification process, with the second NO... xBased on the concentration, the ammonia injection flow rate is adjusted in a closed loop using ammonia flow rate feedforward control commands. This calculates the ammonia injection flow rate that meets the denitrification requirements, eliminates steady-state deviations, and dynamically generates corresponding ammonia flow rate control commands. This ensures that the ammonia injection flow rate stably tracks the required value and avoids flow fluctuations caused by disturbances. The ammonia flow rate feedforward control command can refer to a pre-adjusted ammonia injection flow rate command generated in advance based on real-time disturbance parameters. This forms a "feedforward + feedback" composite control with the closed-loop regulation, compensating for the lag in PID closed-loop regulation. Correspondingly, the ammonia flow rate control command can refer to the final ammonia injection adjustment command output by the first PID controller, directly driving the zone / main control valves of the ammonia injection system. This can include different valve opening adjustment ranges, thereby precisely controlling the ammonia injection flow rate and improving the quality of NOx concentration regulation. In this way, this embodiment uses the PID proportional element to quickly respond to deviations, the integral element to eliminate steady-state deviations, and the derivative element to predict deviation trends, dynamically correcting the ammonia injection adjustment amount, and finally generating the corresponding ammonia flow control commands for the reactor, such as total ammonia flow control commands and zoned ammonia flow control commands. This achieves a closed-loop design of the entire chain, including effective data screening, multi-algorithm prediction, and deviation compensation, which helps to achieve intelligent and precise control of the denitrification system and solves the problems of large lag and poor multi-condition capability of traditional denitrification control.

[0026] Compared with related technologies, this embodiment can verify the validity of data through the status signals of the monitoring instruments corresponding to the reactor, obtain valid process reaction parameters, ensure the reliability of control input data, avoid adjustment errors caused by invalid data, and use different concentration prediction methods to predict the first NO in parallel. x Concentration, adaptable to various operating conditions, improves prediction coverage and fault tolerance. Through collaborative calculation between the concentration prediction model and the state compensator, based on the first NO x The concentration was further corrected to correct for prediction biases caused by fluctuations in operating conditions, resulting in the second NO concentration. x The concentration effectively counteracts the large hysteresis characteristic of the denitrification system, and finally, the first PID controller is combined with the second NO concentration. x Closed-loop regulation of concentration and ammonia flow rate feedforward commands improves the accuracy of ammonia injection flow control and ensures the output NO content. x The concentration is stable and meets the standards, enabling intelligent and precise control of the denitrification system and improving the control quality of the denitrification system.

[0027] Based on the technical implementation described in the above embodiments, to further illustrate the specific implementation process of the method in this embodiment, step 102 may optionally include: using a denitrification system model to predict the first NOx concentration based on process reaction parameters and NOx concentration setpoints, wherein the denitrification system model is constructed based on an online self-learning neural network and is used to adjust the control parameters corresponding to the reactor; and / or, in the unit AGC operation mode, predicting the first NOx concentration based on a special ammonia-saving optimization algorithm according to process reaction parameters and NOx concentration setpoints; and / or, based on process reaction parameters and NOx concentration setpoints, performing online evaluation and intelligent selection of the measurement signals corresponding to the reactor to predict the first NOx concentration.

[0028] In some embodiments, the inputs to the denitrification system model may be process reaction parameters and NO. x Concentration setpoint, output first NO x The system can autonomously learn from changes in operating conditions and dynamically adjust control parameters to adapt to complex and ever-changing denitrification scenarios. The online self-learning neural network is a neural network algorithm with real-time learning capabilities. It does not require offline training and parameter fixing, and can collect real-time operating data of the denitrification system, autonomously correcting network weights and thresholds to adapt to fluctuations in operating conditions (such as changes in coal quality and catalyst decay). It dynamically adjusts reactor control parameters, such as the proportional gain Kp, integral time Ti, and derivative time Td of the PID controller, the ammonia injection zone flow weights, and the target value of the ammonia-nitrogen molar ratio, all dynamically optimized by the online self-learning neural network. For example, the denitrification system model is built based on an LSTM online self-learning neural network, with the input layer consisting of process reaction parameters and NO concentration. x Concentration setpoint, hidden layer has real-time learning module, output layer is the first NO x Concentration, and simultaneously output optimized control parameter values, real-time acquisition of process reaction parameters every 10 seconds, autonomously adjusting network weights, adapting to catalyst activity decay and coal quality fluctuations, and outputting the first NO concentration. x Optimized values ​​for concentration and control parameters are obtained by optimizing control parameters based on model prediction deviations, thereby improving the accuracy of subsequent adjustments. For example, under steady-state conditions, concentration prediction can be performed based on the denitrification system model, while simultaneously optimizing control parameters.

[0029] Specifically, based on the unit's historical denitrification operation data for the past 6 months, the process reaction parameters on the flue gas side, equipment side, and ammonia injection synergy side, as well as NO... x The concentration setpoint is the input feature, with the reactor outlet NO as the metric. x The measured values ​​are used as output labels to train an LSTM neural network model, which serves as the denitrification system model. During actual operation, this model takes real-time process reaction parameters as input and outputs the first NO value. x The concentration is adjusted to adapt to nonlinear operating conditions such as catalyst activity decay and coal quality fluctuations, thereby improving the system's generalization ability.

[0030] Optionally, a mathematical model can be constructed based on the chemical reaction mechanism and catalytic reaction kinetics of SCR denitrification, and the first NO can be calculated by inputting process reaction parameters. x Concentration reflects the denitrification reaction mechanism.

[0031] In some embodiments, load change rate compensation and inlet NO can be superimposed. x Sudden change compensation adapts to drastic parameter changes caused by rapid load fluctuations, constructing a special ammonia-saving optimization algorithm based on intelligent prediction in the AGC (Automatic Generation Control) operation mode. For example, in response to the unit entering AGC operation mode, if the load is monitored to drop from 660MW to 550MW at a rate of 6% of rated load / min, the inlet NO... x When the NOx concentration rapidly decreases from 260 mg / m³ to 215 mg / m³, a special ammonia-saving optimization algorithm can be used to construct ammonia-saving constraints based on the NOx concentration setpoint, and to predict the NOx concentration caused by load changes. x Trend of change, predicting the first NO based on load changes x Concentration makes the predicted values ​​more accurate and avoids excessive ammonia injection. Prediction process: Input real-time process reaction parameters (such as inlet NO). x 215 mg / m³, smoke temperature 348℃, smoke flow rate 1210 kN m³ / h, load decrease rate 6% / min) + NO x The setpoint is 30 mg / m³, and the algorithm incorporates load drop compensation to predict the inlet NO level after 5 minutes. x It will drop to 200 mg / m³, and considering the large hysteresis in denitrification, the first NO... x Concentration, in this way, can be determined in NO x Minimize ammonia injection volume while meeting standards, taking into account both environmental protection and ammonia conservation. Integrate operating condition prediction, deviation compensation, and ammonia conservation constraint logic to avoid excessive ammonia injection under AGC conditions as a fallback measure.

[0032] In some embodiments, online evaluation of measurement signals can be used for monitoring instruments associated with the reactor (such as NO). x The system performs real-time validity checks on signals collected by analyzers, temperature / flow transmitters, etc., to verify the presence of drift, faults, over-range, distortion, and other issues. Invalid signals are eliminated to ensure reliable predictive input. The system also performs intelligent selection and online evaluation of measurement signals, automatically selecting the optimal and most effective measurement signal as the predictive input based on the current process reaction parameters and corresponding operating conditions. If the core signal fails, the system automatically switches to redundant signals or calculated values ​​of related parameters to avoid prediction interruption due to the failure of a single signal.

[0033] For example, the online evaluation of the measurement signals verifies all associated measurement signals, and the evaluation results are as follows: EntranceNO xAnalyzer: Drift → Invalid; NO at the boiler furnace outlet x Analyzer, flue gas flow rate, unit load: Normal → Valid; Flue gas temperature, oxygen content, catalyst pressure difference in the reactor: Normal → Valid; Correspondingly, the inlet NO can be determined according to the above evaluation results x When the signal fails, the NO at the furnace outlet can be selected x and signals strongly correlated with flue gas flow rate and load, and the inlet NO can be calculated through the correlation model x concentration, and then the inlet NO after intelligent selection can be calculated x and the remaining valid process reaction parameters, as well as the NO x set value, to calculate the first NO x concentration, so as to intelligently switch the input when some signals fail and ensure the continuous and stable operation of the system.

[0034] Exemplarily, under the AGC condition, the reliability of the input of process reaction parameters can be ensured through online evaluation and intelligent selection of measurement signals, and the ammonia-saving type prediction can be realized by combining the special ammonia-saving optimization algorithm, the concentration prediction can be carried out by combining the denitration system model, and at the same time, the control parameters can be optimized. The concentration can be predicted through multiple algorithms, which can be adapted to various working conditions, meet the requirements of ammonia-saving rate and ammonia escape, the prediction is not interrupted when the signal fails, the fault tolerance of the system is improved, and a basis is provided for the NO x concentration correction and accurate ammonia injection.

[0035] In this way, different working conditions such as unit steady state, fluctuation, and AGC variable load can be adapted, and the prediction accuracy of the NO x concentration can be improved, and then the problem of exceeding the standard or ammonia escape caused by large lag can be solved.

[0036] Optionally, after predicting the first NO x concentration corresponding to the reactor according to the process reaction parameters using different concentration estimation algorithms, the method further includes: determining the ammonia injection partition weight of the ammonia injection equipment corresponding to the reactor according to the first NO x concentration.

[0037] Correspondingly, the ammonia injection equipment can be a selective catalytic reduction ammonia injection system supporting the reactor, such as a partitioned grid ammonia injection device, which can include multiple independent ammonia injection partitions, and the ammonia injection amount of each ammonia injection partition can be independently adjusted. Based on the first NO x concentration, the proportion of the ammonia injection amount of each ammonia injection partition in the total ammonia injection amount can be dynamically adjusted as the ammonia injection partition weight to achieve accurate ammonia injection, improve the denitration efficiency and avoid ammonia escape. Exemplarily, it can be divided into 3 independent ammonia injection partitions along the flue gas flow direction, such as partition 1, partition 2, and partition 3, and the ammonia injection partition weight can be allocated according to the concentration proportional distribution method according to the first NO xThe concentration determines the weight of each zone. The ammonia injection weight of each zone can also be calculated by combining the flue gas velocity correction coefficient and the catalyst activity correction coefficient. The total weight is 1. The flue gas velocity correction coefficient can be a weight correction coefficient determined based on the ratio of the zone flow velocity to the reference flow velocity, and the catalyst activity correction coefficient can be a weight correction coefficient determined based on the ratio of the reference activity to the zone activity.

[0038] Optionally, generating ammonia flow control instructions for the reactor may specifically include: generating ammonia flow control instructions for the ammonia injection equipment based on the ammonia injection zone weight.

[0039] In some embodiments, the ammonia flow control command may be an ammonia injection control signal issued by the denitrification system, and the ammonia flow rate may be the actual ammonia injection flow rate corresponding to the command, correlated with the denitrification efficiency, and used to compensate for the impact of ammonia injection on NO. x The influence of concentration; the ammonia injection weight can be sent to the ammonia injection equipment through the ammonia flow control command, and the ammonia injection in each zone can be controlled according to the weight, so as to avoid ammonia escape or denitrification failure caused by single concentration matching, and adapt to the automatic control scenario.

[0040] Optionally, step 103 may specifically include: performing collaborative calculations using a concentration prediction model and a state compensator, based on the first NO... x Concentration and process reaction parameters are used to determine the first NO. x NO concentration x Concentration compensation value; using a second PID controller based on NO x Concentration compensation value, first NO x Concentration and NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the corresponding second NO in the reactor. x concentration.

[0041] In some embodiments, the first NO can be predicted using a concentration prediction model. x The concentration trend is analyzed, and a basic predicted value is output. Then, a state compensator is used to correct the basic predicted value in real time for model prediction bias and process disturbances (such as flue gas composition fluctuations and catalyst activity decay), and the NO concentration is output. x Concentration compensation value; the second PID controller can NO x The concentration setpoint is the target, and the first NO is fused. x Concentration and NO x The concentration compensation value is adjusted in a closed loop to output the second NO. x The concentration of NO at the reactor outlet... x The concentration remained stable within the set range.

[0042] Optionally, by co-operating with a concentration prediction model and a state compensator, based on the first NO... xConcentration and process reaction parameters are used to determine the first NO. x NO concentration x The concentration compensation value may specifically include: a third NOx concentration output by a concentration prediction model based on process reaction parameters such as SCR inlet concentration, SCR outlet concentration, total coal feed rate, total air volume, and load command, as well as the first NOx concentration. This concentration prediction model is trained based on historical reaction parameters corresponding to historical operating conditions and is used to predict NOx concentrations for future periods. The state compensator determines the NOx concentration based on the third NOx concentration, the first NOx concentration, CEMS measurements, and the ammonia flow rate corresponding to the ammonia flow control command. x Concentration compensation value.

[0043] Among them, the CEMS measurement value can be the real-time NO measured by the continuous emission monitoring system for flue gas. x Concentration, used for calibration of the first NO x Concentration; the concentration prediction model can be a time-series prediction model (such as a GRU model) trained using historical reaction parameters from historical operating conditions as the training set. The model inputs process reaction parameters, such as SCR inlet concentration and first NO concentration. x Concentration, SCR outlet concentration, total coal feed, total air volume, load command, and output predicted NO for future time periods (e.g., the next 30 seconds or 5 minutes). x Concentration, as the third NO x Concentration, state compensator with third NO x Concentration, first NO x The NO concentration, CEMS measurement value, and ammonia flow rate are taken as inputs. After deviation correction and disturbance compensation calculations, the output is NO. x Concentration compensation value. Correspondingly, NO x If the concentration compensation value is positive, it indicates that future NO... x The concentration will increase, so ammonia injection needs to be increased in advance; if it is negative compensation, it indicates that NO will rise in the future. x The concentration will decrease, so ammonia spraying should be reduced to prevent ammonia escape.

[0044] Optionally, a second PID controller based on NO can be used. x Concentration compensation value, first NO x Concentration and NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the corresponding second NO in the reactor. x Concentration, specifically, may include: based on the first NO concentration via a phase compensation network. x Concentration and CEMS measurements are used to determine the phase compensation value; a second PID controller is then used to determine the phase compensation value and NO. x The concentration compensation value and the concentration setpoint are used to perform closed-loop concentration adjustment to determine the second NOx concentration.

[0045] In some embodiments, the first NO predicted based on process reaction data x Concentration and CEMS measurements are prone to phase deviation due to factors such as signal transmission delay and sensor response differences, leading to adjustment lag. A phase compensation network corrects this phase deviation of the signal corresponding to the concentration, outputting a phase compensation value. This phase compensation value is then fused with the NO value by a second PID controller. x Concentration compensation value and NO x A concentration setpoint is established, a closed-loop control circuit is constructed, the ammonia injection rate is fine-tuned, and the output ammonia injection rate control signal is controlled to ultimately reduce the NO concentration at the reactor outlet. x The concentration stabilizes within the set range. For example, the phase compensation network can employ a lead-lag phase compensation structure to correct the first NO concentration. x The phase difference between the concentration and the CEMS measurement can be calculated through correlation analysis of time series data, and the phase compensation value can be determined by the transfer function of the phase compensation network.

[0046] Optionally, when using the first PID controller based on the second NO x Before generating the corresponding ammonia flow control command for the reactor by performing closed-loop regulation of ammonia flow using the ammonia concentration and ammonia flow feedforward control commands, this embodiment of the method further includes: using an intelligent ammonia flow feedforward controller based on the first NO... x The concentration and process reaction parameters are used to generate ammonia flow feedforward control commands.

[0047] Among them, the intelligent feedforward controller for ammonia flow can be based on the first NO x Based on the concentration, SCR inlet / outlet concentration, total coal feed, total air volume, and load command, the system predicts the basic ammonia injection demand under the current operating conditions and generates feedforward control commands. When process parameters change abruptly, the feedforward controller can immediately generate ammonia injection commands based on the parameter changes without waiting for NO. x Adjust the concentration if it deviates, to avoid NO x Concentration exceeding limits or ammonia escape.

[0048] For example, such as Figure 2 As shown, input parameters may include: load command ( MWD ), total coal supply ( Fu Total air volume () Air SCR inlet concentration () NOx_in SCR outlet concentration () NOx_out ), NOx concentration CEMS measured value ( NOx_cems SCR inlet oxygen, SCR outlet oxygen, ammonia flow rate, and NOx concentration setpoints ( The input parameters, including instrument malfunction and calibration status signals, can be input into different calculation methods in the calculation module, such as the establishment of a denitrification system model and adjustment of control parameters based on online self-learning neural network technology, a special ammonia-saving optimization algorithm under the unit AGC operation mode based on intelligent prediction, online evaluation of various measurement signals, and intelligent selection. These methods are used to predict the first NOx concentration, output control model and parameter identification, adjust the weight allocation of model input parameters, and send the output information to the ammonia flow intelligent feedforward controller, NOx concentration GPC controller, NOx concentration prediction model, improved state variable control compensator, and phase compensation network for flow measurement and dynamic characteristic compensation intelligent control.

[0049] Specifically, the second PID controller can receive the NOx concentration setpoint, compensate for the NOx concentration through an improved state variable control compensator and a phase compensation network, output the adjusted NOx concentration, and input it to the NOx concentration GPC controller. The NOx concentration GPC controller then inputs the adjusted NOx concentration to the first PID controller. The first PID controller is used to receive the NOx concentration sent by the NOx concentration GPC controller and the ammonia flow command sent by the ammonia flow intelligent feedforward controller. ), to regulate ammonia flow and output ammonia flow command ( NH3_d ).

[0050] Compared with related technologies, this embodiment can utilize a denitrification system model, a special ammonia-saving optimization algorithm, and an online evaluation algorithm to predict the first NOx concentration based on process reaction parameters and NOx concentration setpoints, respectively. Then, through a concentration prediction model and a state compensator, the concentration is adjusted according to the first NOx concentration. x Concentration and process reaction parameters are used to determine the first NO. x NO concentration x Concentration compensation value, based on NO using a second PID controller. x Concentration compensation value, first NO x Concentration and NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the corresponding second NO in the reactor. x Concentration, using the first PID controller based on the second NO x Ammonia flow rate is closed-loop regulated using feedforward control commands for concentration and flow rate, generating corresponding ammonia flow rate control commands for the reactor. Adaptive Smith compensation technology is used to compensate for the large hysteresis characteristics of the denitrification process, and fuzzy control technology is used to adapt to the nonlinearity of the denitrification controlled process. Intelligent feedforward technology is used to eliminate the influence of various disturbances on NOx, and model reference adaptive technology is used to improve the adaptive capability of the denitrification control system under wide load operation of the unit, improve the automatic adjustment capability of denitrification NOx, and thus effectively improve the quality of NOx concentration regulation.

[0051] Based on the above Figure 1 The specific implementation of the method shown in this embodiment provides a denitrification NOx control device based on an intelligent control strategy, such as... Figure 3 As shown, the device includes: an acquisition module 31, a prediction module 32, a determination module 33, and a generation module 34; The acquisition module 31 is used to acquire the process reaction parameters corresponding to the reactor based on the status signals of the monitoring instruments of the reactor. Prediction module 32 is used to predict the first NO in the reactor based on the process reaction parameters according to different concentration prediction methods. x concentration; Module 33 is used to perform collaborative calculations using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and process reaction parameters, determine the corresponding second NO in the reactor. x concentration; Generation module 34 is used to utilize the first PID controller based on the second NO x The concentration and ammonia flow rate feedforward control commands are used to perform closed-loop regulation of ammonia flow rate, generating the corresponding ammonia flow rate control command for the reactor.

[0052] In some examples of this embodiment, the prediction module 32 is specifically configured to predict the first NOx concentration using a denitrification system model based on process reaction parameters and NOx concentration setpoints. The denitrification system model is constructed based on an online self-learning neural network and is used to adjust the control parameters corresponding to the reactor. Alternatively, in the unit's AGC operation mode, the first NOx concentration is predicted based on a special ammonia-saving optimization algorithm using process reaction parameters and NOx concentration setpoints. And / or, based on process reaction parameters and NOx concentration setpoints, the measurement signals corresponding to the reactor are evaluated online and intelligently selected to predict the first NOx concentration.

[0053] In some examples of this embodiment, the determining module 33 is specifically configured to determine based on the first NO. x Concentration, determine the ammonia injection zone weight of the ammonia injection equipment corresponding to the reactor.

[0054] In some examples of this embodiment, the generation module 34 is specifically configured to generate ammonia flow control instructions corresponding to the ammonia injection equipment based on the ammonia injection zone weight.

[0055] In some examples of this embodiment, the determining module 33 is specifically configured to perform collaborative calculations using a concentration prediction model and a state compensator, based on the first NO. x Concentration and process reaction parameters are used to determine the first NO. x NO concentration x Concentration compensation value; using a second PID controller based on NO xConcentration compensation value, first NO x Concentration and NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the corresponding second NO in the reactor. x concentration.

[0056] In some examples of this embodiment, the determining module 33 is specifically configured to output a third NOx concentration based on the SCR inlet concentration, SCR outlet concentration, total coal feed, total air volume, and load command in the process reaction parameters, as well as the first NOx concentration, using a concentration prediction model. The concentration prediction model is trained based on historical reaction parameters corresponding to historical operating conditions and is used to predict the NOx concentration for future time periods. The state compensator is used to determine the NOx concentration based on the third NOx concentration, the first NOx concentration, the CEMS measurement value, and the ammonia flow rate corresponding to the ammonia flow control command. x Concentration compensation value.

[0057] In some examples of this embodiment, the determining module 33 is specifically configured to determine the first NO through a phase compensation network. x Concentration and CEMS measurements are used to determine the phase compensation value; a second PID controller is then used to determine the phase compensation value and NO. x The concentration compensation value and the concentration setpoint are used to perform closed-loop concentration adjustment to determine the second NOx concentration.

[0058] In some examples of this embodiment, the generation module 34 is specifically configured to use an intelligent feedforward controller for ammonia flow based on a first NO x The concentration and process reaction parameters are used to generate ammonia flow feedforward control commands.

[0059] Based on the above, Figure 1 Accordingly, this embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. Figure 1 The method shown.

[0060] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0061] Based on the above, Figure 1 The method shown, and Figure 3To achieve the above objectives, the present application also provides an electronic device, comprising a storage medium and a processor; the storage medium for storing a computer program; and the processor for executing the computer program to implement the above-described virtual device embodiments. Figure 1 The method shown.

[0062] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0063] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0064] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0065] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented using hardware. The solution of this application can utilize a denitrification system model, a special ammonia-saving optimization algorithm, and an online evaluation algorithm to predict the first NOx concentration based on process reaction parameters and NOx concentration setpoints, respectively. Then, through a concentration prediction model and a state compensator, based on the first NOx concentration... x Concentration and process reaction parameters are used to determine the first NO. x NO concentration x Concentration compensation value, based on NO using a second PID controller. x Concentration compensation value, first NO x Concentration and NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the corresponding second NO in the reactor. x Concentration, using the first PID controller based on the second NO xAmmonia flow rate is closed-loop regulated using feedforward control commands for concentration and flow rate, generating corresponding ammonia flow rate control commands for the reactor. Adaptive Smith compensation technology is used to compensate for the large hysteresis characteristics of the denitrification process, and fuzzy control technology is used to adapt to the nonlinearity of the denitrification controlled process. Intelligent feedforward technology is used to eliminate the influence of various disturbances on NOx, and model reference adaptive technology is used to improve the adaptive capability of the denitrification control system under wide load operation of the unit, improve the automatic adjustment capability of denitrification NOx, and thus effectively improve the quality of NOx concentration regulation.

[0066] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0067] The above are merely specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A denitrification NOx control method based on an intelligent control strategy, characterized in that, include: Based on the status signals of the reactor's monitoring instruments, the corresponding process reaction parameters of the reactor are obtained; Based on the process reaction parameters, using different concentration prediction methods, the first NO corresponding to the reactor is predicted. x concentration; Through collaborative computation using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and the process reaction parameters, determine the second NO corresponding to the reactor. x concentration; Using the first PID controller based on the second NO x The concentration and ammonia flow rate feedforward control commands are used to perform closed-loop regulation of ammonia flow rate, generating the ammonia flow rate control command corresponding to the reactor.

2. The method according to claim 1, characterized in that, The method of predicting different concentrations of NO is based on the process reaction parameters to predict the first NO level corresponding to the reactor. x Concentration, including: The first NOx concentration is predicted using a denitrification system model based on the process reaction parameters and NOx concentration setpoints. This denitrification system model is constructed based on an online self-learning neural network and is used to adjust the control parameters corresponding to the reactor; and / or, In the unit's AGC operation mode, based on a special ammonia-saving optimization algorithm, the first NOx concentration is predicted according to the process reaction parameters and NOx concentration setpoint; and / or, Based on the process reaction parameters and NOx concentration setpoints, the measurement signals corresponding to the reactor are evaluated online and intelligently selected to predict the first NOx concentration.

3. The method according to claim 2, characterized in that, Based on the process reaction parameters obtained by the method of predicting different concentrations, the first NO corresponding to the reactor is predicted. x After concentration, the method further includes: According to the first NO x Concentration, determining the ammonia injection zone weight of the ammonia injection equipment corresponding to the reactor; Generate the ammonia flow control command corresponding to the reactor, including: Based on the ammonia injection zone weights, ammonia flow control commands corresponding to the ammonia injection equipment are generated.

4. The method according to claim 2, characterized in that, The method involves collaborative computation using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and the process reaction parameters, determine the second NO corresponding to the reactor. x Concentration, including: Through collaborative computation using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and the process reaction parameters, the first NO is determined. x NO concentration x Concentration compensation value; Using a second PID controller based on the NO x Concentration compensation value, the first NO x Concentration and the NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the second NO corresponding to the reactor. x concentration.

5. The method according to claim 4, characterized in that, The concentration prediction model and the state compensator work together to calculate based on the first NO x Based on the concentration and the process reaction parameters, the first NO is determined. x NO concentration x Concentration compensation values ​​include: The concentration prediction model outputs a third NOx concentration based on the SCR inlet concentration, SCR outlet concentration, total coal feed, total air volume and load command in the process reaction parameters, as well as the first NOx concentration. The concentration prediction model is trained based on historical reaction parameters corresponding to historical operating conditions and is used to predict the NOx concentration in future time periods. The state compensator determines the NOx concentration based on the third NOx concentration, the first NOx concentration, the CEMS measurement value, and the ammonia flow rate corresponding to the ammonia flow control command. x Concentration compensation value.

6. The method according to claim 5, characterized in that, The second PID controller is based on the NO x Concentration compensation value, the first NO x Concentration and the NO x The concentration setpoint is used for closed-loop concentration adjustment to determine the second NO corresponding to the reactor. x Concentration, including: Based on the first NO through the phase compensation network x The phase compensation value is determined by comparing the concentration with the CEMS measurement value. Using a second PID controller based on the phase compensation value and the NO x The concentration compensation value and the concentration set value are used to perform closed-loop concentration adjustment to determine the second NOx concentration.

7. The method according to claim 1, characterized in that, The first PID controller is used based on the second NO x Before generating the ammonia flow control command corresponding to the reactor by using the concentration and ammonia flow rate feedforward control commands for closed-loop regulation of ammonia flow rate, the method further includes: Based on the first NO gas flow intelligent feedforward controller, the ammonia flow rate is controlled. x The concentration and the process reaction parameters are used to generate the ammonia flow feedforward control command.

8. A denitrification NOx control device based on an intelligent control strategy, characterized in that, include: The acquisition module is used to acquire the process reaction parameters corresponding to the reactor based on the status signals of the reactor's monitoring instruments; The prediction module is used to predict the first NO concentration in the reactor based on the process reaction parameters obtained by the different concentration prediction methods. x concentration; The determination module is used to perform collaborative calculations using a concentration prediction model and a state compensator, based on the first NO... x Based on the concentration and the process reaction parameters, determine the second NO corresponding to the reactor. x concentration; The generation module is used to utilize the first PID controller based on the second NO x The concentration and ammonia flow rate feedforward control commands are used to perform closed-loop regulation of ammonia flow rate, generating the ammonia flow rate control command corresponding to the reactor.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

Images