Coal-fired unit flue gas recirculation dynamic optimization control method and control system thereof

Abstract

This application relates to the field of coal-fired power generation technology, and in particular to a dynamic optimization control method and control system for flue gas recirculation in coal-fired power units. The method includes: real-time acquisition of multi-dimensional parameters during unit operation; initial parameter setting based on the acquired multi-dimensional parameters; preprocessing the real-time acquired multi-dimensional parameters and constructing technical and economic indicators and evaluation indices to calculate the combustion stability index F1 and the economic index F2; dynamically adjusting and optimizing the flue gas recirculation parameters based on the combustion stability index F1 and the economic index F2; and closed-loop iterative optimization. The system includes a data acquisition system, a data processing and calculation system, and a dynamic control system that are sequentially connected to form a closed-loop control link. This enables real-time perception of the unit's operating status, coordinated protection of combustion stability and economy, self-optimization capability under varying operating conditions, and high-precision data processing, thereby promoting efficient and clean operation of coal-fired power units.

Description

Technical Field

[0001] This application relates to the field of coal-fired power generation technology, and in particular to a dynamic optimization control method and control system for flue gas recirculation in coal-fired units. Background Technology

[0002] Flue gas recirculation technology is a key means for coal-fired power units to reduce NOx emissions and optimize the combustion process. Its control effect directly affects the safety, economy and environmental protection of the unit operation.

[0003] Existing flue gas recirculation control systems for coal-fired power units mostly employ fixed parameters or simple closed-loop control modes, which have the following shortcomings: 1. The control logic is rigid and does not fully consider the impact of varying operating conditions on the combustion process, resulting in a mismatch between flue gas recirculation parameters and actual operating requirements; Second, the failure to consider the synergistic optimization of combustion stability and economy may lead to an imbalance where combustion efficiency is sacrificed in pursuit of emission reduction, or environmental protection indicators are relaxed in order to ensure stable operation. Third, insufficient data processing accuracy leads to biases in the basis for optimization decisions; Fourth, the control strategy lacks adaptability and is difficult to respond quickly and reach the optimal operating state under complex disturbance conditions. Summary of the Invention

[0004] The purpose of this application is to provide a dynamic optimization control method and control system for flue gas recirculation in coal-fired power units, which can sense the operating status of the unit in real time, coordinate and ensure combustion stability and economy, have the ability to self-optimize under changing operating conditions and high-precision data processing, so as to solve the above-mentioned defects of the existing technology and promote the efficient and clean operation of coal-fired power units.

[0005] Firstly, this application provides a dynamic optimization control method for flue gas recirculation in coal-fired power units, comprising: S10. Real-time acquisition of multi-dimensional parameters during unit operation; S20. Set initial parameters based on the multi-dimensional parameters collected in S10; S30. The multi-dimensional parameters collected in real time in S10 are preprocessed and technical and economic indicators are calculated and evaluation indices are constructed to calculate the combustion stability index F1 and the economic index F2. S40. Dynamically adjust and optimize flue gas recirculation parameters based on combustion stability index F1 and economic index F2: If F1 < preset minimum threshold F1min, the safety control mode is activated to reduce the flue gas recirculation ratio r and increase the secondary air volume until F1 ≥ F1min; if F1 ≥ preset minimum threshold F1min, the economic optimization mode is activated, and the optimal flue gas recirculation flow rate and recirculation ratio ropt are output through the fusion strategy of generalized predictive control and reinforcement learning. S50, repeat S10~S40 to form a closed-loop iterative optimization.

[0006] Furthermore, in S10, the collected multi-dimensional parameters include coal quality information and coal quantity fed into the furnace. S20 includes: S21. Based on the coal quality information and coal quantity collected in S10, calculate the theoretical air volume. Combined with the current load of the unit, set the initial excess air coefficient according to the historical best operating data and design data, calculate and output the minimum air volume Q1 required for combustion. S22. Retrieve the minimum furnace ventilation volume Qmin from the unit design parameters, compare the minimum air volume Q1 required for combustion with the minimum furnace ventilation volume Qmin, and determine the fan output air volume Qout: If Q1 > Qmin, then Qout = Q1; if Q1 ≤ Qmin, then Qout = Qmin. S23. Adjust the fan output based on the fan output air volume Qout, and adjust the recirculated flue gas volume according to the initial excess air coefficient and coal quality characteristics in a preset ratio, and set the initial flue gas recirculation ratio r0.

[0007] Furthermore, S30 includes: S31. Raw data preprocessing: Wavelet threshold denoising algorithm is used to denoise the multi-dimensional parameters collected in real time in S10, remove high-frequency interference signals, and identify and remove outliers through the 3σ criterion to avoid the impact of extreme data on subsequent calculations. S32. Calculate technical and economic indicators according to standards: Calculate the core technical and economic indicators of the boiler system, boiler auxiliary equipment, steam turbine system, and steam turbine auxiliary equipment online in real time, including the NOx emission concentration at the furnace outlet, boiler thermal efficiency, steam turbine heat consumption rate, plant power consumption rate, and unit power supply coal consumption. S33. Construction of evaluation index, including: S331. Data standardization: The modified Z-score standardization method is used to uniformly transform parameters of different dimensions to the [0,1] interval; S332. Based on the standardized data, construct the combustion stability index F1 and the economic index F2 to quantitatively evaluate the unit's operating status.

[0008] Furthermore, in S10, the multi-dimensional parameters collected also include furnace temperature, furnace pressure, burner nozzle flame detection intensity, and heated surface wall temperature. In step S332, the combustion stability index F1 is calculated by weighting the furnace temperature X1, furnace pressure X2, burner nozzle flame detection intensity X3, and heating surface wall temperature X4 collected in step S10. The calculation expression is as follows: F1 = k1×X1 + k2×X2 + k3×X3 + k4×X4, Where k1, k2, k3, and k4 are the weighting coefficients of furnace temperature, furnace pressure, burner nozzle flame detection intensity, and heated surface wall temperature, respectively, satisfying k1+k2+k3+k4=1; The economic index F2 is calculated by weighting the NOx emission concentration at the furnace outlet Y1, boiler thermal efficiency Y2, turbine heat consumption rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5 calculated in S32. The calculation expression is as follows: F2 = c1×Y1 + c2×Y2 + c3×Y3 + c4×Y4 + c5×Y5, Wherein, c1, c2, c3, c4, and c5 are the weighting coefficients of NOx emission concentration at the furnace outlet, boiler thermal efficiency, turbine heat consumption rate, plant power consumption rate, and unit power supply coal consumption, respectively, satisfying c1 + c2 + c3 + c4 + c5 = 1.

[0009] Furthermore, S332 also includes: The weight coefficients in the F1 and F2 calculation expressions are determined by calibration using the analytic hierarchy process combined with historical operating data of the unit, and each weight coefficient can be dynamically adjusted according to the unit's operating load range and standard requirements. Among them, the weighting coefficients k1 for furnace temperature and k3 for burner nozzle flame detection intensity in the F1 calculation expression shall account for no less than 60%; In the F2 calculation expression, the weighting coefficients c1 and c5 for NOx emission concentration at the furnace outlet and power supply coal consumption should account for no less than 60%.

[0010] Furthermore, in step S331, the parameters in the F1 and F2 calculation expressions are first modified and optimized using the modified Z-score standardization method: The positive safety index parameters are corrected, including furnace temperature X1, burner nozzle flame detection intensity X3 and boiler thermal efficiency Y2. The correction and standardization formula is: Z' = 0.5×(X - μ) / σ + 0.5; The negative safety index parameters are corrected, including furnace pressure X2, heating surface wall temperature X4, furnace outlet NOx emission concentration Y1, turbine heat consumption rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5. The standardized formula for furnace pressure X2 correction is: Z' = 0.5×(|μ| - |X|) / σ+ 0.5, The corrected standardization formulas for the heated surface wall temperature X4, furnace outlet NOx emission concentration Y1, turbine heat rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5 are as follows: Z' = 0.5×(μ - X) / σ + 0.5; Where X is the currently collected raw parameter value, μ is the historical operating average of the parameter, and σ is the historical operating standard deviation of the parameter.

[0011] Furthermore, in S40, a dynamic optimization of flue gas recirculation parameters is achieved through a collaborative decision-maker and a dual-algorithm fusion control strategy, including: Constraint setting: The combustion stability index F1 is used as a hard constraint. First, the minimum threshold F1min is preset. Construction of the comprehensive optimization objective function: Taking the optimal economic index F2 as the core objective, a comprehensive optimization objective function is constructed, the expression of which is: min(F2)=min(c1×Y1 + c2×Y2 + c3×Y3 + c4×Y4 + c5×Y5) stF1≥F1min; Dual-algorithm fusion control strategy: A control strategy that integrates coordinated control optimization techniques based on generalized predictive control with model self-driven optimization techniques based on reinforcement learning. The system uses a generalized predictive control algorithm to predict short-term operating conditions and output stable control commands. It uses a reinforcement learning algorithm to learn from long-term operating data and dynamically optimize control parameters, enabling the system to have self-optimizing characteristics under varying operating conditions and complex disturbances. Finally, the system integrates the output results of the two algorithms through a collaborative decision-maker to determine the optimal flue gas recirculation flow rate and recirculation ratio, and outputs them to the actuator.

[0012] Furthermore, the S40 also includes: Compare the recirculation ratio ropt with the current flue gas recirculation ratio rcurrent: If |ropt-rcurrent|>2%, then adjust according to the gradient, adjusting by 0.5%~1% each time, adjusting the frequency of the circulating fan inverter and the opening of the flue gas regulating damper, so that the recirculation ratio gradually approaches ropt; If |ropt-rcurrent|≤2%, then maintain the current parameters to avoid frequent adjustments that could cause system fluctuations.

[0013] Secondly, this application provides a dynamic optimization control system for flue gas recirculation in coal-fired power units, which applies the dynamic optimization control method for flue gas recirculation in coal-fired power units described in any of the preceding descriptions. The control system includes: a data acquisition system, a data processing and calculation system, and a dynamic control system that are sequentially connected to form a closed-loop control link. The data acquisition system is connected to high-precision sensors deployed at key locations of the unit to collect multi-dimensional parameters during unit operation in real time, providing raw data support for subsequent data processing and optimization control. The data processing and computing system includes a core processing module, a data analysis module, and a standard computing module, which are used to preprocess the raw data collected by the data acquisition system and to complete the calculation of technical and economic indicators and the construction of evaluation indices based on the preprocessed data. The dynamic control system includes a generalized predictive control module and a reinforcement learning algorithm module, which are used to achieve dynamic optimization of flue gas recirculation parameters through a collaborative decision-maker and a dual-algorithm fusion control strategy.

[0014] Furthermore, the key locations include at least one of the following: coal mill outlet, four corners of furnace, burner nozzle, heating surface tube wall, flue inlet and outlet, main steam pipeline, reheat steam pipeline, feedwater pipeline, high-pressure cylinder exhaust pipeline, medium-pressure cylinder exhaust pipeline, and low-pressure cylinder exhaust pipeline. The multi-dimensional parameters include at least one of the following: coal quality information, coal quantity, furnace oxygen content, furnace CO concentration, furnace temperature, furnace pressure, burner nozzle flame detection intensity, heating surface wall temperature, primary air volume, secondary air volume, recirculated flue gas volume, furnace outlet NOx emission concentration, flue gas temperature, flue gas oxygen content, fly ash carbon content, main steam temperature, main steam pressure, reheat steam temperature, reheat steam pressure, circulating fan power consumption, forced draft fan power consumption, primary air fan power consumption, induced draft fan power consumption, unit power generation, plant power consumption, and power supply. The high-precision sensor includes at least one of an online coal quality analyzer, a flow meter, a thermocouple and pressure transmitter, a flue gas analyzer, and an infrared gas analyzer.

[0015] Compared with existing technologies, the dynamic optimization control method and control system for flue gas recirculation in coal-fired power units provided in this application, through a data acquisition system, senses the unit's operating status in real time and collects multi-dimensional parameters during the unit's operation. It performs precise initial parameter setting, data processing, technical and economic indicator calculation, and evaluation index construction, calculating the combustion stability index F1 and the economic index F2. Then, it dynamically adjusts and optimizes the flue gas recirculation parameters, possessing scientific stability and economic evaluation and a collaborative dynamic control strategy. This enables dynamic optimization of flue gas recirculation parameters under varying operating conditions and complex disturbances, ensuring safe and stable boiler combustion. This approach aims to minimize NOx emissions and coal consumption for power generation, while improving the economic and environmental performance of the unit. It also addresses existing technologies that fail to adequately consider the impact of varying operating conditions on the combustion process, leading to a mismatch between flue gas recirculation parameters and actual operational requirements. Furthermore, it addresses the issues of insufficient consideration of the synergistic optimization of combustion stability and economy, which can result in an imbalance where combustion efficiency is sacrificed for emission reduction or environmental indicators are relaxed to ensure stable operation. Additionally, it addresses the problems of insufficient data processing accuracy leading to biased optimization decisions and a lack of adaptive control strategies that struggle to respond quickly and reach optimal operating conditions under complex disturbances. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the dynamic optimization control method for flue gas recirculation in coal-fired power units provided in the embodiments of this application. Figure 2 This is a schematic diagram of the structure of the dynamic optimization control system for flue gas recirculation in a coal-fired power unit provided in an embodiment of this application.

[0018] Figure label: 100 - Data acquisition system; 200 - Data processing and computing system; 300 - Dynamic control system. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0022] In the description of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this application is in use. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, the terms "first," "second," and "third," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0023] Furthermore, terms such as "horizontal," "vertical," and "sag" do not imply that components must be absolutely horizontal or suspended, but rather that they can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0024] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0025] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0026] like Figure 1 and Figure 2 As shown, this application provides a dynamic optimization control method for flue gas recirculation in coal-fired power units, and a dynamic optimization control system for flue gas recirculation in coal-fired power units that applies this control method.

[0027] like Figure 1 As shown in the embodiments of this application, the dynamic optimization control method for flue gas recirculation in coal-fired power units includes: Step S10: Real-time acquisition of multi-dimensional parameters during unit operation.

[0028] Among them, the multi-dimensional parameters can be acquired in real time through the data acquisition system; and the multi-dimensional parameters may specifically include at least one of the following: coal quality information, coal quantity, furnace oxygen content, furnace CO concentration, furnace temperature, furnace pressure, burner nozzle flame detection intensity, heating surface wall temperature, primary air volume, secondary air volume, recirculated flue gas volume, furnace outlet NOx emission concentration, flue gas temperature, flue gas oxygen content, fly ash carbon content, main steam temperature, main steam pressure, reheat steam temperature, reheat steam pressure, circulating fan power consumption, forced draft fan power consumption, primary air fan power consumption, induced draft fan power consumption, unit power generation, plant power consumption, and power supply.

[0029] Step S20: Set initial parameters based on the multi-dimensional parameters collected in step S10, including: Step S21: Based on the coal quality information and coal quantity collected in step S10, and the theoretical air volume calculated based on the coal quality analysis results, combined with the current load of the unit, the initial excess air coefficient is set according to the historical best operating data and design data, and the minimum air volume Q1 required for combustion is calculated and output. Step S22, Determining the fan output air volume Qout: Retrieve the minimum furnace ventilation volume Qmin from the unit's design parameters, which is the minimum ventilation threshold to ensure safe combustion in the furnace. Compare the minimum air volume Q1 required for combustion with the retrieved minimum furnace ventilation volume Qmin to determine the fan output air volume Qout. If Q1 > Qmin, then Qout = Q1, that is, Q1 is used as the initial output air volume Qou of the fan; If Q1≤Qmin, then Qout=Qmin, that is, Qmin is used as the initial output air volume of the blower Qout to ensure that the furnace ventilation volume meets the requirements for safe combustion.

[0030] Step S23, Initial Flue Gas Recirculation Parameter Matching: Based on the fan output air volume Qout determined in step S22, adjust the fan output to stabilize the actual ventilation volume at the fan output air volume Qout. At the same time, according to the initial excess air coefficient and coal quality characteristics, adjust the recirculated flue gas volume according to the preset ratio and set the initial flue gas recirculation ratio r0. According to r0 = recirculated flue gas volume / total flue gas volume, the initial value is determined according to the unit load. For example, when the load is >80% of the rated load, r0 = 10% ~ 20%; when the load is <50% of the rated load, r0 = 10% ~ 40%.

[0031] Step S30: Preprocess the multi-dimensional parameters collected in real time in step S10 and calculate and construct technical and economic indicators and evaluation indices to calculate the combustion stability index F1 and the economic index F2. This step can transmit the demonstration data of the multi-dimensional parameters collected in real time from the data acquisition system to the data processing and calculation system, and then the data processing and calculation system processes them, including: Step S31, Raw data preprocessing: The wavelet threshold denoising algorithm is used to denoise the multi-dimensional parameters collected in real time in step S10, remove high-frequency interference signals, and identify and remove outliers through the 3σ criterion to avoid the impact of extreme data on subsequent calculations.

[0032] Step S32: Calculate the technical and economic indicators according to the standard (DL / T 904-2015 standard "Calculation Method of Technical and Economic Indicators of Thermal Power Plants"): Calculate the core technical and economic indicators of the boiler system, boiler auxiliary equipment, turbine system, and turbine auxiliary equipment online in real time, including the NOx emission concentration at the furnace outlet, boiler thermal efficiency, turbine heat consumption rate, plant power consumption rate, and unit coal consumption for power supply.

[0033] Step S33, Evaluation Index Construction, including: Step S331, Data Standardization Processing: The modified Z-score standardization method is used to uniformly transform parameters of different dimensions to the [0,1] interval to ensure the rationality of the calculation and the integrity of the parameter contribution.

[0034] To avoid missing contributions due to a standardized value of 0 when the parameter equals the historical mean, the parameters in the subsequent F1 and F2 calculation expressions can be corrected and optimized using a modified Z-score standardization method. 1) The positive safety index parameters are revised. These parameters include the furnace temperature X1 and burner nozzle flame detection intensity X3 in the F1 calculation expression, and the boiler thermal efficiency Y2 in the F2 calculation expression. The revised standardized formula is as follows: Z' = 0.5×(X - μ) / σ + 0.5; Where X is the currently collected raw parameter value, μ is the historical operating mean of the parameter, and σ is the historical operating standard deviation of the parameter; Regarding the correction of positive safety indicators, the higher the value of furnace temperature X1 and the higher the value of burner nozzle flame detection intensity X3, the more stable the combustion; the higher the value of boiler thermal efficiency Y2, the better the economy. After correction, the standardized value range of all parameters is [0,1]. When the parameter is equal to the mean, i.e., Z'=0.5, it indicates a moderate stability contribution. The closer the parameter is to the optimal safety value, the closer Z' is to 1. The closer it is to the safety boundary, the closer Z' is to 0, ensuring that the contribution of parameters to the stability index under different operating conditions can be reasonably reflected.

[0035] 2) Correction of negative safety index parameters, which include furnace pressure X2 and heating surface wall temperature X4 in the F1 calculation expression, and furnace outlet NOx emission concentration Y1, turbine heat rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5 in the F2 calculation expression. The furnace pressure X2 characteristic is that the smaller the absolute value, the safer it is. The furnace pressure X2 is corrected by first taking the absolute value and then using a negative index. The standardized formula for furnace pressure X2 correction is: Z' = 0.5×(|μ| - |X|) / σ+ 0.5; The corrected standardization formulas for the heated surface wall temperature X4, furnace outlet NOx emission concentration Y1, turbine heat rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5 are as follows: Z' = 0.5×(μ - X) / σ + 0.5; Where X is the currently collected raw parameter value, μ is the historical operating mean of the parameter, and σ is the historical operating standard deviation of the parameter; The lower the values ​​of heating surface wall temperature, NOx emission concentration, turbine heat consumption rate, plant power consumption rate, and coal consumption for power supply, the better the economic performance. After correction, the standardized values ​​of all parameters are in the range of [0,1]. When the parameter is equal to the mean, i.e., Z'=0.5, it indicates a moderate economic contribution. The closer the parameter is to the economic optimum, the closer Z' is to 1, ensuring that the contribution of the parameters to the economic index under different operating conditions can be reasonably reflected.

[0036] Step S332: Based on the standardized data, construct the combustion stability index F1 and the economic index F2 to quantitatively evaluate the unit's operating status. 1) Combustion stability index F1: Calculated by weighting the furnace temperature X1, furnace pressure X2, burner nozzle flame detection intensity X3, and heated surface wall temperature X4 collected in step S10. The calculation expression is as follows: F1 = k1×X1 + k2×X2 + k3×X3 + k4×X4, Among them, k1, k2, k3, and k4 are the weighting coefficients of furnace temperature, furnace pressure, burner nozzle flame detection intensity, and heating surface wall temperature, respectively, satisfying k1+k2+k3+k4=1. Each weighting coefficient is determined by calibration using the analytic hierarchy process combined with historical operating data of the unit, and each weighting coefficient can be dynamically adjusted according to the unit's operating load range and standard requirements. It is recommended that the weighting coefficients k1 and k3 for furnace temperature and burner nozzle flame detection intensity account for no less than 60% to ensure the dominant role of core stability indicators.

[0037] 2) Economic Index F2: Calculated by weighting the NOx emission concentration at the furnace outlet (Y1), boiler thermal efficiency (Y2), turbine heat consumption rate (Y3), plant power consumption rate (Y4), and unit coal consumption for power supply (Y5) obtained in step S32. The calculation expression is as follows: F2 = c1×Y1 + c2×Y2 + c3×Y3 + c4×Y4 + c5×Y5, Wherein, c1, c2, c3, c4, and c5 are the weighting coefficients for NOx emission concentration at the furnace outlet, boiler thermal efficiency, turbine heat rate, plant power consumption rate, and unit coal consumption for power supply, respectively, satisfying c1 + c2 + c3 + c4 + c5 = 1; each weighting coefficient is determined by calibration using the analytic hierarchy process combined with historical operating data of the unit, and each weighting coefficient can be dynamically adjusted according to the unit's operating load range and standard requirements. Considering the dual needs of environmental protection and energy conservation, it is recommended that the weighting coefficient c1 for NOx emission concentration at the furnace outlet and the weighting coefficient c5 for unit coal consumption for power supply account for no less than 60%.

[0038] Step S40: Dynamically adjust and optimize flue gas recirculation parameters based on combustion stability index F1 and economic index F2. This includes receiving combustion stability index F1 and F2 data through a dynamic control system and making the following judgments and decisions: 1) If F1 < preset minimum threshold F1min: it is determined that the combustion stability does not meet the requirements, and the safety control mode is immediately activated to prioritize the combustion stability, reduce the flue gas recirculation ratio r by 1% to 2% each time, and increase the secondary air volume until F1 ≥ F1min; 2) If F1 ≥ preset minimum threshold F1min: the combustion state is determined to be safe, the economic optimization mode is started, the generalized predictive control algorithm is used to predict the operating condition change trend in the next 5 to 10 minutes, and the control parameters are optimized by combining the reinforcement learning algorithm with historical operating data, and the optimal flue gas recirculation flow rate and recirculation ratio ropt are output. 3) Compare the recirculation ratio ropt with the current flue gas recirculation ratio rcurrent: If |ropt-rcurrent|>2%, then adjust according to the gradient, adjusting by 0.5%~1% each time, adjusting the frequency of the circulating fan inverter and the opening of the flue gas regulating damper, so that the recirculation ratio gradually approaches ropt; If |ropt-rcurrent|≤2%, then maintain the current parameters to avoid frequent adjustments that could cause system fluctuations.

[0039] Step S50, repeat steps S10~S40 to form a closed-loop iterative optimization: complete data acquisition, processing, calculation and decision-making every 2~5 seconds to form a closed-loop iterative optimization link, track changes in unit operating conditions and parameter fluctuations in real time, dynamically correct flue gas recirculation parameters, and ensure that the unit always operates in the optimal range of safety, stability, economy and environmental protection.

[0040] It also includes a control strategy that integrates a collaborative decision-maker and dual algorithms to achieve dynamic optimization of flue gas recirculation parameters, including: 1) Constraint setting: The combustion stability index F1 is used as a hard constraint. First, the minimum threshold F1min is preset and determined according to the rated parameters of the unit and the safe operation boundary. For example, F1≥0.5, to ensure that all optimization operations are carried out within the safe operation range and to avoid safety hazards such as furnace flameout and overheating of heating surfaces. 2) Construction of the comprehensive optimization objective function: Taking the optimal economic index F2 as the core objective, a comprehensive optimization objective function is constructed, the expression of which is: min(F2)=min(c1×Y1 + c2×Y2 + c3×Y3 + c4×Y4 + c5×Y5) stF1≥F1min; 3) Dual-algorithm fusion control strategy: A control strategy that integrates coordinated control optimization technology based on generalized predictive control (GPC) and model self-optimization technology based on reinforcement learning (DQN): The GPC algorithm predicts short-term operating conditions and outputs stable control commands; the reinforcement learning algorithm learns from long-term operating data and dynamically optimizes control parameters, enabling the system to have self-optimizing characteristics under varying operating conditions and complex disturbances (such as coal quality fluctuations and sudden load changes); finally, the collaborative decision-maker integrates the output results of the two algorithms to determine the optimal flue gas recirculation flow rate and recirculation ratio, and outputs them to the actuators (such as the recirculating fan frequency converter, flue gas regulating damper, and blower controller).

[0041] Compared with existing technologies, the dynamic optimization control method for flue gas recirculation in coal-fired power units provided in this application uses a data acquisition system to perceive the unit's operating status in real time and collect multi-dimensional parameters during the unit's operation. It performs precise initial parameter setting, data processing, technical and economic indicator calculation, and evaluation index construction, calculating the combustion stability index F1 and the economic index F2. Then, it dynamically adjusts and optimizes the flue gas recirculation parameters, possessing scientific stability and economic evaluation and a collaborative dynamic control strategy. This enables dynamic optimization of flue gas recirculation parameters under varying operating conditions and complex disturbances, ensuring safe and stable boiler combustion. This technology aims to minimize NOx emissions and coal consumption for power generation, while improving the economic and environmental performance of the unit. It also addresses existing technologies that fail to adequately consider the impact of varying operating conditions on the combustion process, leading to a mismatch between flue gas recirculation parameters and actual operating requirements. Furthermore, it addresses the imbalance caused by neglecting the synergistic optimization of combustion stability and economy, which can result in sacrificing combustion efficiency for emission reduction or relaxing environmental protection standards to ensure stable operation. Additionally, it addresses issues such as insufficient data processing accuracy leading to biased optimization decisions and a lack of adaptive control strategies that struggle to respond quickly and reach optimal operating conditions under complex disturbances.

[0042] like Figure 1 and Figure 2 As shown in the embodiment of this application, a dynamic optimization control system for flue gas recirculation of a coal-fired unit is also provided. The aforementioned dynamic optimization control method for flue gas recirculation of a coal-fired unit is applied. The control system includes: a data acquisition system 100, a data processing and calculation system 200, and a dynamic control system 300 that are sequentially connected to form a closed-loop control link.

[0043] Specifically, the data acquisition system 100 is connected to high-precision sensors deployed at key locations of the unit to collect multi-dimensional parameters during unit operation in real time (as in step S10), providing raw data support for subsequent data processing and optimization control.

[0044] Furthermore, the key locations may specifically include at least one of the following: coal mill outlet, four corners of the furnace, burner nozzle, heating surface tube wall, flue inlet and outlet, main steam pipeline, reheat steam pipeline, feedwater pipeline, high-pressure cylinder exhaust pipeline, medium-pressure cylinder exhaust pipeline, and low-pressure cylinder exhaust pipeline.

[0045] Furthermore, the high-precision sensors may specifically include at least one of an online coal quality analyzer, a flow meter, a thermocouple and pressure transmitter, a flue gas analyzer, and an infrared gas analyzer. Specifically, among the multi-dimensional parameters acquired in real time during step S10: the coal quality information entering the furnace can be acquired through the online coal quality analyzer; the NOx emission concentration at the furnace outlet can be acquired through the infrared gas analyzer; flow parameters such as air volume and flue gas volume can be acquired through the flow meter; parameters such as temperature and pressure can be acquired through the thermocouple and pressure transmitter; and flue gas composition parameters can be acquired through the flue gas analyzer. Preferably, the sampling frequency of each high-precision sensor is set to 1Hz to ensure data real-time performance, and the acquisition frequency can be dynamically adjusted according to system requirements. Multiple sensors are calibrated periodically, including calibrating the flue gas analyzer and infrared gas analyzer using standard gases, calibrating the flow meter using known flow standards, and calibrating the temperature sensor using a temperature calibrator to ensure the accuracy of data acquisition.

[0046] The data processing and calculation system 200 is used to preprocess the raw data collected by the data acquisition system and to calculate technical and economic indicators and construct evaluation indices based on the preprocessed data. This data processing and calculation system 200 can use an industrial control computer (IPC) as its core processing module, equipped with a Python-based data analysis module to achieve wavelet threshold noise reduction and 3σ outlier removal. It can also have a built-in standard calculation module, conforming to the DL / T 904-2015 standard, to output indicators such as boiler thermal efficiency and coal consumption for power supply in real time. The system uses a modified Z-score standardization method to standardize the data and determines the weight coefficients of the combustion stability index, the economic index, and the minimum threshold F1 of the combustion stability index through the analytic hierarchy process. min .

[0047] Regarding the dynamic control system 300, its collaborative decision-maker adopts a PLC controller, integrating a generalized predictive control module and a reinforcement learning (DQN) algorithm module to implement a dual-algorithm fusion control strategy, thereby achieving dynamic optimization of flue gas recirculation parameters; the actuator uses a variable frequency speed-regulating fan and an electric regulating damper, with a control response time ≤1 second.

[0048] Compared with the prior art, the dynamic optimization control method and control system for flue gas recirculation of coal-fired units provided in this application have the following advantages: 1. Ensure safe and stable combustion: By constructing a combustion stability index and incorporating it as a hard constraint into the optimization loop, it ensures that all control operations are carried out within the safety boundary, effectively avoiding risks such as furnace flameout and overheating of heating surfaces under varying operating conditions and complex disturbances, and improving the reliability of unit operation; 2. Improve operational economy: Through precise parameter acquisition and processing, scientific construction of economic index, and combined with dual-algorithm fusion control strategy, dynamic optimization of flue gas recirculation parameters can be achieved, which can reduce power supply coal consumption by 3~5g / kWh and significantly improve the energy-saving efficiency of the unit. 3. Enhance environmental protection and emission reduction effects: By incorporating NOx emission concentration into the economic optimization target, and through precise control of flue gas recirculation parameters, the NOx emission concentration at the furnace outlet can be reduced by 10% to 20%, helping the unit meet stringent environmental emission standards. 4. Possesses strong adaptability and self-optimization capability: Integrating generalized predictive control and reinforcement learning algorithms, it can respond in real time to changing operating conditions such as coal quality fluctuations and load changes, automatically optimize control strategies without manual intervention, and is suitable for coal-fired units of different capacities and coal types.

[0049] The following is a specific embodiment using a 350MW coal-fired power unit as an example, with detailed illustrations: Step S10: Collect coal quality information for the furnace using an online coal quality analyzer: carbon content 50.57% (as received), hydrogen content 2.17% (as received), oxygen content 9.80% (as received), sulfur content 0.46% (as received), moisture content 7.63% (as received), ash content 28.90% (as received), net calorific value 17.64 MJ / kg (as received), and coal feed rate 45 t / h.

[0050] Step S20: Based on the coal quality analysis results, the theoretical air volume is 6.15 kg / kg. Combined with the current 20% rated load (70MW), the initial excess air coefficient is set to 1.52. The minimum air volume required to meet combustion is calculated as Q1 = 45 × 6.15 × 1.52 = 421 t / h. Determining the fan output air volume: The minimum ventilation volume of the furnace of this unit is Qmin=340t / h. Since Q1=421t / h>Qmin=340t / h, the initial output air volume of the fan is determined to be Qout=Q1=421t / h. Initial flue gas recirculation parameter matching: Adjust the output of the blower to stabilize the actual ventilation volume at 421t / h. Set the initial recirculation ratio r0=20% according to 20% load. Calculate the initial recirculated flue gas volume = total flue gas volume × 20%. Adjust the frequency of the recirculating fan inverter (initial frequency 20Hz) and the opening of the flue gas regulating damper (initial opening 40%) to stabilize the recirculated flue gas volume at the target value.

[0051] Step S30: The data acquisition system collects multi-dimensional parameters during the unit's operation in real time. The collected raw data is then transmitted to the data processing and calculation system. The data processing system performs noise reduction (wavelet threshold set to 0.05) and 3σ outlier removal (confidence interval 99.7%). Based on the DL / T 904-2015 standard, the core technical and economic indicators are calculated. The main output data are: furnace temperature 930℃, furnace pressure -60Pa, burner nozzle flame detection intensity 95%, heating surface wall temperature 549℃, furnace outlet NOx emission concentration 345mg / Nm³, boiler thermal efficiency 91.1%, turbine heat consumption rate 9715kJ / kWh, plant power consumption rate 9.03%, and unit power supply coal consumption 404.04g / kWh. The data was standardized using a modified Z-score standardization method. Based on the standardized data, the weighting coefficients for the combustion stability index were determined using the analytic hierarchy process (AHP) as k1=0.3, k2=0.2, k3=0.35, k4=0.15, and the weighting coefficients for the economy index were c1=0.25, c2=0.1, c3=0.1, c4=0.1, c5=0.45. The minimum threshold for the combustion stability index, F1min, was set to 0.5. Substituting these weighting coefficients, the combustion stability index F1 and the economy index F2 under the current operating conditions were calculated. The specific calculation process is as follows: Correcting the Z-score standardization formula: Positive indicators (furnace temperature X1, burner nozzle flame detection intensity X3, and boiler thermal efficiency Y2): Z'=0.5×(X-μ) / σ+0.5; Negative indicators (heated surface wall temperature X4, furnace outlet NOx emission concentration Y1, turbine heat consumption rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5): Z'=0.5×(μ-X) / σ+0.5; Negative index (furnace pressure X2): Z'=0.5×(|μ|-|X|) / σ+0.5; Where X is the currently collected original parameter value, μ is the historical operating mean of the unit for this parameter (based on the statistical data of normal operation of the unit under the same load in the past year), and σ is the historical operating standard deviation of the parameter; the standardized value range after correction is uniformly [0,1]. When X=μ, Z'=0.5 (medium contribution), to avoid missing contribution and eliminate dimensional differences. The historical mean μ, standard deviation σ, and standardized value of each parameter are calculated as follows: 1. Parameters related to the combustion stability index F1 (weights: k1=0.3, k2=0.2, k3=0.35, k4=0.15, summed to 1, F1min=0.6): Furnace temperature X1 = 930℃, positive safety indicators: historical mean μ1 = 920℃, standard deviation σ1 = 30℃, standardized value Z1' = 0.667; Furnace pressure X2 = -60Pa, a negative safety indicator; the smaller the absolute value, the safer it is: original absolute pressure |X2| = 60Pa, historical average absolute pressure |μ| = 70Pa, historical standard deviation of absolute pressure σ = 12Pa, standardized value Z2' = 0.917; Burner nozzle flame detection intensity X3=0.95, positive safety index: historical mean μ3=0.94, standard deviation σ3=0.05, standardized value Z3' =0.6; The wall temperature of the heated surface X4 = 549℃, and the negative safety index is: historical average μ4 = 550℃, standard deviation σ4 = 20℃, and standardized value Z4' = 0.525. Substituting the weighting coefficients, we calculate that F1=0.67≥F1min=0.6, indicating that the combustion state is safe and the calculation result meets the safety threshold requirements. 2. Parameters related to the economic index F2 (weights: c1=0.25, c2=0.1, c3=0.1, c4=0.1, c5=0.45, summed to 1): The NOx emission concentration at the furnace outlet is Y1=345mg / Nm³, and the negative economic indicators are: historical average μ5=350mg / Nm³, standard deviation σ5=40mg / Nm³, and standardized value Z5'=0.563. Boiler thermal efficiency Y2=91.1%, positive economic indicators: historical mean μ6=91%, standard deviation σ6=1.5%, standardized value Z6'=0.533; The turbine heat rate Y3 = 9715 kJ / kWh, negative economic indicators: historical mean μ7 = 9720 kJ / kWh, standard deviation σ7 = 100 kJ / kWh, standardized value Z7' = 0.525; Plant power consumption rate Y4=9.03%, negative economic indicators: historical mean μ8=9.02%, standard deviation σ8=0.3%, standardized value Z8'=0.483; The unit's coal consumption for power generation is Y5 = 404.04 g / kWh. The negative economic indicators are: historical average μ9 = 404.64 g / kWh, standard deviation σ9 = 10 g / kWh, and standardized value Z9' = 0.530. Substituting the weighting coefficients, we get F2=0.53. F2 is the comprehensive optimization objective. The higher the value after weighted summation, the better the economic efficiency.

[0052] Step S40: The dynamic control system starts the economic optimization mode. The generalized predictive control algorithm predicts that the load will stabilize at 70MW within the next 8 minutes and the coal quality will not fluctuate significantly. The reinforcement learning algorithm combines historical operating data to output the optimal flue gas recirculation ratio ropt=28%. The current recirculation ratio rcurrent=20%, |28% -20%|=8%. The frequency of the circulating fan inverter is adjusted to 35Hz according to the gradient, and the opening of the flue gas regulating damper is adjusted to 50%, so that the recirculation ratio is gradually increased to 30%.

[0053] Step S50: Repeat steps S10~S40, perform closed-loop iterative optimization, and collect and calculate the adjusted data in real time: furnace temperature: 925℃, furnace pressure: -60Pa, burner nozzle flame detection intensity: 95%, heating surface wall temperature: 552℃, furnace outlet NOx emission concentration: 306mg / Nm³, boiler thermal efficiency: 91.8%, turbine heat consumption rate: 9710kJ / kWh, plant power consumption rate: 8.95%, unit power supply coal consumption: 400.40g / kWh, the economic index F2 is optimized to 0.68; the combustion stability index F1=0.64≥0.5, the combustion state is stable; the system maintains the current parameters to achieve the optimal operating state.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A dynamic optimization control method for flue gas recirculation in a coal-fired power unit, characterized in that, include: S10. Real-time acquisition of multi-dimensional parameters during unit operation; S20. Set initial parameters based on the multi-dimensional parameters collected in S10; S30. The multi-dimensional parameters collected in real time in S10 are preprocessed and technical and economic indicators are calculated and evaluation indices are constructed to calculate the combustion stability index F1 and the economic index F2. S40. Dynamically adjust and optimize flue gas recirculation parameters based on combustion stability index F1 and economic index F2: If F1 < preset minimum threshold F1min, activate the safety control mode to reduce the flue gas recirculation ratio r and increase the secondary air volume until F1 ≥ F1min; If F1 ≥ preset minimum threshold F1min, start the economic optimization mode and output the optimal flue gas recirculation flow rate and recirculation ratio ropt through the fusion strategy of generalized predictive control and reinforcement learning. S50, repeat S10~S40 to form a closed-loop iterative optimization.

2. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to claim 1, characterized in that, In S10, the collected multi-dimensional parameters include coal quality information and coal quantity fed into the furnace. S20 includes: S21. Based on the coal quality information and coal quantity collected in S10, calculate the theoretical air volume. Combined with the current load of the unit, set the initial excess air coefficient according to the historical best operating data and design data, calculate and output the minimum air volume Q1 required for combustion. S22. Retrieve the minimum furnace ventilation volume Qmin from the unit design parameters, compare the minimum air volume Q1 required for combustion with the minimum furnace ventilation volume Qmin, and determine the fan output air volume Qout: If Q1 > Qmin, then Qout = Q1; if Q1 ≤ Qmin, then Qout = Qmin. S23. Adjust the fan output based on the fan output air volume Qout, and adjust the recirculated flue gas volume according to the initial excess air coefficient and coal quality characteristics in a preset ratio, and set the initial flue gas recirculation ratio r0.

3. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to claim 2, characterized in that, S30 includes: S31. Raw data preprocessing: Wavelet threshold denoising algorithm is used to denoise the multi-dimensional parameters collected in real time in S10, remove high-frequency interference signals, and identify and remove outliers through the 3σ criterion to avoid the impact of extreme data on subsequent calculations. S32. Calculate technical and economic indicators according to standards: Calculate the core technical and economic indicators of the boiler system, boiler auxiliary equipment, steam turbine system, and steam turbine auxiliary equipment online in real time, including the NOx emission concentration at the furnace outlet, boiler thermal efficiency, steam turbine heat consumption rate, plant power consumption rate, and unit power supply coal consumption. S33. Construction of evaluation index, including: S331. Data standardization: The modified Z-score standardization method is used to uniformly transform parameters of different dimensions to the [0,1] interval; S332. Based on the standardized data, construct the combustion stability index F1 and the economic index F2 to quantitatively evaluate the unit's operating status.

4. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to claim 3, characterized in that, In S10, the multi-dimensional parameters collected also include furnace temperature, furnace pressure, burner nozzle flame detection intensity, and heated surface wall temperature. In step S332, the combustion stability index F1 is calculated by weighting the furnace temperature X1, furnace pressure X2, burner nozzle flame detection intensity X3, and heating surface wall temperature X4 collected in step S10. The calculation expression is as follows: F1 = k1×X1 + k2×X2 + k3×X3 + k4×X4, Where k1, k2, k3, and k4 are the weighting coefficients of furnace temperature, furnace pressure, burner nozzle flame detection intensity, and heated surface wall temperature, respectively, satisfying k1+k2+k3+k4=1; The economic index F2 is calculated by weighting the NOx emission concentration at the furnace outlet Y1, boiler thermal efficiency Y2, turbine heat consumption rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5 calculated in S32. The calculation expression is as follows: F2 = c1×Y1 + c2×Y2 + c3×Y3 + c4×Y4 + c5×Y5, Wherein, c1, c2, c3, c4, and c5 are the weighting coefficients of NOx emission concentration at the furnace outlet, boiler thermal efficiency, turbine heat consumption rate, plant power consumption rate, and unit power supply coal consumption, respectively, satisfying c1 + c2 + c3 + c4 + c5 = 1.

5. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to claim 4, characterized in that, S332 also includes: The weight coefficients in the F1 and F2 calculation expressions are determined by calibration using the analytic hierarchy process combined with historical operating data of the unit, and each weight coefficient can be dynamically adjusted according to the unit's operating load range and standard requirements. Among them, the weighting coefficients k1 for furnace temperature and k3 for burner nozzle flame detection intensity in the F1 calculation expression shall account for no less than 60%; In the F2 calculation expression, the weighting coefficients c1 and c5 for NOx emission concentration at the furnace outlet and power supply coal consumption should account for no less than 60%.

6. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to any one of claims 3 to 5, characterized in that, In step S331, the parameters in the F1 and F2 calculation expressions are first modified and optimized using the modified Z-score standardization method. The positive safety index parameters are corrected, including furnace temperature X1, burner nozzle flame detection intensity X3 and boiler thermal efficiency Y2. The correction and standardization formula is: Z' = 0.5×(X - μ) / σ + 0.5; The negative safety index parameters are corrected, including furnace pressure X2, heating surface wall temperature X4, furnace outlet NOx emission concentration Y1, turbine heat consumption rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5. The standardized formula for furnace pressure X2 correction is: Z' = 0.5×(|μ| - |X|) / σ+ 0.5, The corrected standardization formulas for the heated surface wall temperature X4, furnace outlet NOx emission concentration Y1, turbine heat rate Y3, plant power consumption rate Y4, and unit coal consumption for power supply Y5 are as follows: Z' = 0.5×(μ - X) / σ + 0.5; Where X is the currently collected raw parameter value, μ is the historical operating average of the parameter, and σ is the historical operating standard deviation of the parameter.

7. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to claim 4, characterized in that, In S40, the method further includes a dynamic optimization of flue gas recirculation parameters through a collaborative decision-maker and a dual-algorithm fusion control strategy, including: Constraint setting: The combustion stability index F1 is used as a hard constraint. First, the minimum threshold F1min is preset. Construction of the comprehensive optimization objective function: Taking the optimal economic index F2 as the core objective, a comprehensive optimization objective function is constructed, the expression of which is: min(F2)=min(c1×Y1 + c2×Y2 + c3×Y3 + c4×Y4 + c5×Y5) stF1≥F1min; Dual-algorithm fusion control strategy: A control strategy that integrates coordinated control optimization techniques based on generalized predictive control with model self-driven optimization techniques based on reinforcement learning. The system uses a generalized predictive control algorithm to predict short-term operating conditions and output stable control commands. It uses a reinforcement learning algorithm to learn from long-term operating data and dynamically optimize control parameters, enabling the system to have self-optimizing characteristics under varying operating conditions and complex disturbances. Finally, the system integrates the output results of the two algorithms through a collaborative decision-maker to determine the optimal flue gas recirculation flow rate and recirculation ratio, and outputs them to the actuator.

8. The dynamic optimization control method for flue gas recirculation in coal-fired power units according to claim 1 or 7, characterized in that, S40 also includes: Compare the recirculation ratio ropt with the current flue gas recirculation ratio rcurrent: If |ropt-rcurrent|>2%, then adjust according to the gradient, adjusting by 0.5%~1% each time, adjusting the frequency of the circulating fan inverter and the opening of the flue gas regulating damper, so that the recirculation ratio gradually approaches ropt; If |ropt-rcurrent|≤2%, then maintain the current parameters to avoid frequent adjustments that could cause system fluctuations.

9. A dynamic optimization control system for flue gas recirculation in a coal-fired power unit, characterized in that, The dynamic optimization control method for flue gas recirculation of a coal-fired power unit according to any one of claims 1 to 8, wherein the control system comprises: a data acquisition system, a data processing and calculation system, and a dynamic control system that are sequentially connected in communication to form a closed-loop control link; The data acquisition system is connected to high-precision sensors deployed at key locations of the unit to collect multi-dimensional parameters during unit operation in real time, providing raw data support for subsequent data processing and optimization control. The data processing and computing system includes a core processing module, a data analysis module, and a standard computing module, which are used to preprocess the raw data collected by the data acquisition system and to complete the calculation of technical and economic indicators and the construction of evaluation indices based on the preprocessed data. The dynamic control system includes a generalized predictive control module and a reinforcement learning algorithm module, which are used to achieve dynamic optimization of flue gas recirculation parameters through a collaborative decision-maker and a dual-algorithm fusion control strategy.

10. The dynamic optimization control system for flue gas recirculation in coal-fired power units according to claim 9, characterized in that, The key locations include at least one of the following: coal mill outlet, four corners of furnace, burner nozzle, heating surface tube wall, flue inlet and outlet, main steam pipeline, reheat steam pipeline, feedwater pipeline, high-pressure cylinder exhaust pipeline, medium-pressure cylinder exhaust pipeline, and low-pressure cylinder exhaust pipeline. The multi-dimensional parameters include at least one of the following: coal quality information, coal quantity, furnace oxygen content, furnace CO concentration, furnace temperature, furnace pressure, burner nozzle flame detection intensity, heating surface wall temperature, primary air volume, secondary air volume, recirculated flue gas volume, furnace outlet NOx emission concentration, flue gas temperature, flue gas oxygen content, fly ash carbon content, main steam temperature, main steam pressure, reheat steam temperature, reheat steam pressure, circulating fan power consumption, forced draft fan power consumption, primary air fan power consumption, induced draft fan power consumption, unit power generation, plant power consumption, and power supply. The high-precision sensor includes at least one of an online coal quality analyzer, a flow meter, a thermocouple and pressure transmitter, a flue gas analyzer, and an infrared gas analyzer.

Images