Combustion control method for reducing boiler NOx generation amount based on intelligent combustion model

Abstract

The invention relates to a combustion control method for reducing boiler NOx generation amount based on an intelligent combustion model, which comprises the following steps: S1, basic data acquisition and intelligent combustion model construction, including full-dimension operation data acquisition and multi-algorithm fusion intelligent combustion model construction; s2, realizing real-time regulation and control based on dynamic combustion control of sensing data, including real-time monitoring and analysis of a temperature field and air-coal parameters, and generation and execution of a hierarchical regulation and control instruction of an intelligent combustion model; s3, according to the fed-back NOx concentration data, iteration and upgrading of the intelligent combustion model are controlled, including regulation and control effect feedback and NOx concentration verification, and model iteration and control strategy optimization; the method has the advantages that the uniformity of a temperature field is regulated and controlled, thermal NOX is inhibited from generating a main channel, the generation amount of the NOX is reduced, and meanwhile, the combustion stability is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of boiler combustion, in particular to a combustion control method for reducing the generation amount of NOx of a boiler based on an intelligent combustion model. BACKGROUND

[0002] The current control of the generation amount of NOx in the boiler combustion process has many shortcomings, whether in the front-end combustion optimization technology or in the back-end denitration process, and also faces problems such as insufficient detection and control accuracy, cost and working condition adaptation contradiction, etc., as follows:

[0003] I. Limitations of front-end low-nitrogen combustion technology;

[0004] 1. Control contradiction of air-coal ratio and combustion organization: like the mainstream low-nitrogen technology of staged combustion, it is necessary to create anoxic combustion environment by adjusting the amount of overfire air to suppress the generation of NOx, but it is difficult to accurately control the amount of overfire air. Too much input will easily lead to lack of air in the main combustion zone, causing incomplete combustion of pulverized coal and significantly reducing the thermal efficiency of the boiler; too little input will not achieve the expected denitration effect. And for tangentially fired boilers, the outlet of the furnace is prone to have large smoke temperature deviation due to residual rotation, and the impinging fired boiler has very high requirements for the adjustment of the primary and secondary air momentum ratio and the swirl intensity, and slight deviation will cause the flame to deviate, thereby causing local high temperature or oxygen enrichment, inducing a sharp increase in the generation of NOx.

[0005] 2. Insufficient control accuracy of single-pipe air-coal ratio: there is a problem of uneven flow distribution in multiple powder pipes of a coal mill, and existing detection and control devices cannot accurately match the air and coal quantities of single pipes. When the air-coal ratio of some powder pipes is unbalanced, it will not only cause incomplete combustion in the local area, but also form a local area of oxygen enrichment or high temperature, which becomes a local source of NOx generation.

[0006] 3. Poor adaptability to coal types and loads: most low-nitrogen combustion technologies are designed for specific coal types, and if the coal quality fluctuates (such as changes in volatile matter and nitrogen content), the original combustion parameters will no longer be suitable, which will cause a sharp increase in NOx emissions. In low-load conditions, the flame of tangentially fired boilers is prone to deviate, and oil guns are needed for combustion support, while impinging fired boilers are prone to unstable combustion, both of which will destroy the ideal conditions for low-nitrogen combustion, causing a significant decline in NOx control effect at low loads.

[0007] II. Shortcomings of the detection and control system;

[0008] 1. Incomplete temperature field and pollutant detection: current furnace temperature detection is mostly discrete point detection, which is difficult to obtain complete temperature field distribution data and cannot timely detect local high temperature areas. And the detection of flue gas NOx is mostly concentrated at the outlet of the furnace, which is difficult to timely capture the NOx generation dynamics in different areas of the furnace, resulting in a lag in control instructions.

[0009] 2. Control model adaptation is insufficient: In the practical application of existing model predictive control, adaptive fuzzy control and other technologies, the fuzzy rule base and the prediction model parameters are mostly set based on historical data. When new conditions occur in the boiler (such as extreme load, new type of coal combustion), the model is difficult to quickly adapt and adjust, and unreasonable control instructions are easily output, which cannot achieve dynamic and accurate control of NOx.

[0010] III. Balance between environmental protection and economy: Many low-nitrogen technologies sacrifice boiler thermal efficiency to reduce NOx emissions. For example, excessively reducing the combustion temperature can inhibit thermal NOx, but will increase the loss of incomplete combustion of fuel; increasing the amount of flue gas recirculation can dilute the oxygen concentration, but will reduce the heat exchange efficiency of the furnace. At the same time, whether it is front-end combustion equipment modification or rear-end denitration system installation, the initial investment is high, and for small and medium-sized enterprises, the initial investment pressure is large, and the operation and maintenance cost of some technologies is high, affecting the enthusiasm of enterprises to promote low-nitrogen technology. SUMMARY

[0011] In view of the above problems, the purpose of the present application is to provide a combustion control method for reducing the generation of NOx in a boiler based on a smart combustion model, which can reduce NOx while ensuring the combustion efficiency of the boiler, so as to overcome the shortcomings of the prior art.

[0012] The present application provides a combustion control method for reducing the generation of NOx in a boiler based on a smart combustion model, which comprises the following steps:

[0013] Step S1: basic data acquisition and smart combustion model construction, including: full-dimensional operation data acquisition and fusion of multiple algorithms for smart combustion model construction;

[0014] Step S2: dynamic combustion control based on perception data to achieve real-time regulation and control, including: real-time monitoring and analysis of temperature field and wind-coal parameters, and generation and execution of hierarchical regulation and control instructions of the smart combustion model;

[0015] Step S3: according to the feedback of NOx concentration data, control the iteration and upgrade of the smart combustion model, including: feedback of regulation and control effect and verification of NOx concentration, and model iteration and control strategy optimization.

[0016] As a preferred embodiment of the present application, step S1 further comprises the following steps:

[0017] Step S11: full-dimensional operation data acquisition and baseline establishment;

[0018] Step S111: A multi-parameter detection system is built, a temperature sensor is arranged at the combustion offset area of the boiler furnace (covering different heights, sections and burner outlets of the furnace), real-time detection of the temperature field is realized, a coal flow monitoring device is installed in each powder pipe of each coal mill, and the coal flow monitoring device is used to record the coal flow speed, flow and concentration data; the operation parameters of the boiler load, the furnace negative pressure and the NOx concentration are synchronously collected to form a data collection matrix;

[0019] Step S112: Historical data analysis and baseline determination, extract historical operation data under load conditions (low, medium and high load) in the boiler furnace, including the powder pipe air-coal ratio, temperature field distribution data and NOx emission data, determine the operation baseline under different conditions through statistical analysis, and determine the condition under which the current NOx generation is high (for example, the imbalance of the air-coal ratio of a certain powder pipe under a certain load, and the local temperature is too high).

[0020] As a preferred embodiment of the present application, the step S1 further comprises the following steps:

[0021] Step S12: Intelligent combustion model construction by fusing multiple algorithms;

[0022] Step S121: Intelligent combustion model construction, taking the MPC prediction model as the core framework, inputting the working condition parameter data composed of the boiler load and the coal mill operation station number, constructing the mapping relationship model of the NOx generation and the air-coal ratio and the temperature field distribution; embedding an adaptive combustion control algorithm module, establishing a combustion rule library (for example, "when the load increases by 10%, the single pipe coal quantity increases by no more than 8% and the air quantity increases by 12% simultaneously") for the time-varying characteristics (such as coal quality fluctuation and load change) in the boiler combustion process, and realizing the dynamic self-adaptive adjustment of the intelligent combustion model;

[0023] Step S122: The adaptive combustion control algorithm module is based on the diagonal matrix method, each powder pipe of each coal mill is taken as an independent control unit, an air-coal ratio optimization diagonal matrix is constructed, and a single pipe air-coal ratio optimal value (combined with historical optimal data and theoretical calculation, such as the air-coal ratio 1.2-1.4 corresponding to the coal with high volatile matter) is provided for data support for single pipe regulation and control.

[0024] As a preferred embodiment of the present application, the step S2 further comprises the following steps:

[0025] Step S21: Real-time monitoring and analysis of the temperature field and the air-coal parameters;

[0026] Step S211: real-time data synchronization and analysis, real-time data monitored by the pulverized coal flow monitoring device is transmitted to the boiler combustion control platform through the data acquisition system, and filtering algorithm is used to eliminate abnormal data (such as sensor instantaneous fluctuation); the visual system of the boiler combustion control platform is used to display the furnace temperature field distribution cloud picture, and the local high temperature area is directly identified, wherein the boiler combustion > 1500℃ will cause the rapid increase of thermal NOx;

[0027] Step S212: key parameter deviation diagnosis, comparison of real-time data and baseline data, and calculation of deviation between real-time data and baseline data by using adaptive combustion control algorithm module, wherein the deviation data includes wind-coal ratio deviation (difference between single tube wind-coal ratio and optimal value of diagonal matrix, if the deviation exceeds ± 10%, the regulation is triggered) and temperature field uniformity deviation (calculation of standard deviation of furnace cross section temperature, if the standard deviation > 80℃, it is judged that the temperature field is uneven).

[0028] As a preferred embodiment of the present application, the following steps are further included in step S2:

[0029] Step S22: model-based hierarchical regulation instruction generation and execution;

[0030] Step S221: primary regulation, single tube wind-coal ratio fine adjustment, for the pulverized coal pipe with wind-coal ratio deviation exceeding the standard, the intelligent combustion model outputs single tube air volume and coal quantity adjustment instruction based on the diagonal matrix method (for example, if the wind-coal ratio of a certain pulverized coal pipe is 5% lower, the instruction is to increase the air volume by 3% or reduce the coal quantity by 2%), which is executed by the coal mill inlet air damper and the coal feeder speed controller, so that the wind-coal ratio of each pulverized coal pipe is in the optimal interval, thereby reducing the generation of NOx caused by incomplete combustion of fuel and local oxygen enrichment from the source;

[0031] Step S222: secondary regulation, temperature field uniformity optimization, for the uneven temperature field area, the intelligent combustion model is used to analyze the wind-coal ratio of the pulverized coal pipe corresponding to the high temperature area and the operation state of the burner, and the cooperative adjustment instruction is output, if the local high temperature is caused by too high coal quantity of a certain group of burners, the corresponding pulverized coal quantity is adjusted by the adaptive combustion control algorithm module, and the air volume of the adjacent pulverized coal pipe is fine adjusted to guide the uniform distribution of flame center; if the overall temperature is too high, the primary air temperature is reduced or the secondary air ratio is adjusted under the premise of ensuring the combustion efficiency, so that the average temperature of the furnace is controlled in the low NOx combustion interval of 1350-1450℃.

[0032] As a preferred embodiment of the present application, the following steps are further included in step S3:

[0033] Step S31: regulation effect feedback and NOx concentration verification;

[0034] Step S311: Real-time monitoring of the control effect. After executing the command to adjust the combustion temperature, continuously monitor the target pulverized coal-air ratio, furnace temperature field distribution, and outlet NOx concentration data. The adaptive combustion control algorithm module calculates the NOx emission reduction before and after the control (e.g., a reduction of 15%-20% under low load conditions and a reduction of 10%-15% under high load conditions is considered satisfactory). At the same time, verify the boiler thermal efficiency (ensure that the thermal efficiency decrease does not exceed 1%) to avoid an imbalance between emission reduction and efficiency.

[0035] Step S312: Trace the cause of the deviation. If the NOx concentration does not meet the standard or the temperature field is still uneven after adjustment, trace the data chain to find the cause (such as sensor failure, sudden change in coal quality, or lag in model parameters), and re-execute the control process after targeted solutions.

[0036] As a preferred embodiment of the present invention, step S3 further includes the following step:

[0037] Step S32: Model iteration and control strategy optimization;

[0038] Step S321: Model parameter update. The operating condition data, adjustment instructions, and emission reduction effect data of each adjustment are included in the model training set. The rule base is updated through the rule optimization module controlled by the adaptive combustion control algorithm module. The mapping relationship parameters in the predictive control of the adaptive combustion control algorithm module are corrected to improve the model's adaptability to complex operating conditions.

[0039] Step S322: Upgrade the control strategy, regularly analyze the control effect under different operating conditions, optimize the optimal range of single-pipe air-coal ratio using the diagonal matrix of air-coal ratio optimization, and supplement the training data of the adaptive combustion control algorithm module for newly emerging operating conditions (such as new coal types and extreme loads) to form a closed-loop control system of monitoring-control-feedback-iteration, so as to achieve continuous and refined reduction of NOx generation.

[0040] As a preferred embodiment of the present invention, the pulverized coal flow monitoring device includes: an integrated sensor assembly and flange seats located on both sides of the integrated sensor assembly, wherein the inner diameter of the integrated sensor assembly is the same as the inner diameter of the pulverized coal pipeline; the integrated sensor assembly consists of a coiled shell and four sub-sensors, each sub-sensor being a quarter-circle in shape, and the four sub-sensors being evenly distributed and embedded in the inner wall of the coiled shell.

[0041] Each of the sub-sensors includes: three electrostatic ion intensity measuring rings, two coal powder flow rate measuring rings, and an insulating ring;

[0042] The two coal powder flow rate measuring rings are spaced apart between the three electrostatic ion strength measuring rings, and the insulating ring is placed between the adjacent electrostatic ion strength measuring ring and the coal powder flow rate measuring ring.

[0043] The two coal powder flow velocity measuring rings are used to measure the coal powder flow velocity. Each ring receives a through signal from coal powder flowing in the same direction. Due to the time difference between the coal powder passing through the two rings, the coal powder flow velocity is calculated using the following formula:

[0044]

[0045] T n =y t -x t

[0046] In the formula, V represents the pulverized coal flow rate; d represents the distance between the two pulverized coal flow rate measuring rings; T n This represents the time difference between the two coal powder particles passing through the two powder flow rate measuring rings; x t y represents the time it takes for pulverized coal to pass through the first pulverized coal flow rate measurement loop. t This indicates the time it takes for pulverized coal to pass through the second pulverized coal flow rate measuring loop;

[0047] Among them, the three electrostatic ion strength measuring rings calculate the coal powder concentration based on the coal powder flow frequency. The electrostatic ion strength is positively correlated with the coal powder concentration. The current coal powder concentration C = (Qt - Q0) * K can be determined in real time by measuring the change in electrostatic ion strength between the measured electrostatic ion strength and the standard coal powder concentration.

[0048] In the formula, C represents the coal powder concentration; Qt represents the electrostatic ion intensity measured in real time inside the coal powder pipeline; Q0 represents the standard electrostatic ion intensity inside the pipeline; and K represents the coefficient relating electrostatic ion intensity and concentration.

[0049] The formula for calculating the pulverized coal flow rate is as follows:

[0050] δ=C×V×S

[0051] In the formula, δ represents the pulverized coal flow rate; S represents the cross-sectional area of ​​the pipe.

[0052] As a preferred embodiment of the present invention, it further includes a signal amplification processor, which is connected to the boiler combustion control platform and installed on the outer wall of the integrated sensor assembly. The signal amplification processor is used to transmit the data measured by the four sub-sensors to the air-coal parameter measurement experimental module of the boiler combustion control platform in real time. The integrated sensor assembly includes a signal loading and staining ring and a signal balancing ring, which are respectively positioned at both ends. An insulating ring is provided between the signal loading and staining ring and the adjacent electrostatic ion strength measurement ring. The signal loading and staining ring is used to collect the signal uniformly and transmit it to the signal amplification processor. The signal balancing ring is used to balance and distinguish the signals between the three electrostatic ion strength measurement rings and the two coal powder flow rate measurement rings.

[0053] As a preferred embodiment of the present invention, the coal powder parameter measurement experimental module includes: a coal powder flow rate analysis module, a flow rate analysis module, a concentration analysis module, a coal powder particle fineness analysis module, and a calorific value analysis module. The coal powder flow rate analysis module is used to monitor the coal powder flow rate, the flow rate analysis module is used to monitor the coal powder flow rate, the concentration analysis module is used to monitor the coal powder concentration, the coal powder particle fineness analysis module uses the concentration analysis module, the coal powder flow rate analysis module, the flow rate analysis module, the concentration analysis module, and pipe size data to obtain the coal powder particle fineness, and the calorific value analysis module uses the concentration analysis module, the coal powder flow rate analysis module, the flow rate analysis module, and the concentration analysis module to obtain the calorific value of the mixed coal powder.

[0054] The advantages and positive effects of this invention are:

[0055] 1. This invention improves adaptability and anti-interference capability under complex working conditions, improves efficiency and accuracy of air-coal ratio optimization by using the diagonal matrix method, accurately controls the air-coal ratio in a single pipe, and eliminates local NOx generation sources.

[0056] 2. This invention achieves temperature field uniformity control, suppresses the main channel of thermal NOx formation, reduces NOx formation, and ensures combustion stability.

[0057] 3. This invention achieves online measurement of multiple parameters of primary air-coal powder flow state through an integrated sensor assembly. Because the four sub-sensors form a pipe-like structure with the same inner diameter as the measuring pipe, the inner wall of the integrated sensor assembly is smooth, minimizing erosion by coal powder. This allows it to withstand higher pressures and temperatures. Furthermore, the four sensors as a group provide more accurate measurements. This solves the current problem of not being able to monitor the airflow and coal powder quantity and calorific value of each burner during boiler combustion, and also addresses the issue of adjusting the most critical parameter within each burner—the air-coal ratio—to achieve calorific value balance.

[0058] 4. In this invention, the integrated sensor assembly's detection and subsequent pulverized coal flow control of the coal mills form an independent closed-loop control system. Whether the system is running or shut down, it does not affect the normal operation of the boiler system. It can accept signals transmitted from each coal mill, calculate and process data such as the pulverized coal flow rate, velocity, and distribution ratio measured in the output pipes of each coal mill, and adjust the valve opening of each pulverized coal regulating valve through PLC control software. This ensures that the air-coal concentration and velocity in the primary air-coal pipes of each coal mill are operated according to a predetermined input ratio. Furthermore, it can output corresponding data in real time to the boiler combustion control DCS system.

[0059] 5. The sensing electrode in the integrated sensor assembly of the present invention is embedded in the inner wall of the housing, which isolates the sensing electrode from the coal powder. It is a safe passive measurement, which is not affected by non-measurement factors. The measurement is independent of the coal powder flow mode. The full-section design can accurately capture and measure the flow state of coal powder in the entire primary air duct. Especially under the condition of severely wet coal, it can accurately measure the flow velocity, mass flow rate and concentration of primary air coal powder. The parameters are measured online, and the velocity, mass flow rate, particle size, calorific value and air-coal powder mixing concentration of coal powder in each primary air duct are measured in real time. Attached Figure Description

[0060] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0061] In the following description, numerous specific details are set forth for illustrative purposes and to provide a thorough understanding of one or more embodiments. However, it will be apparent that these embodiments may also be implemented without these specific details. In other instances, well-known structures and devices are shown in block diagram form for ease of description of one or more embodiments.

[0062] like Figure 1 As shown in the figure, an embodiment of the present invention provides a combustion control method for reducing NOx generation in boilers based on a smart combustion model, comprising the following steps:

[0063] Step S1: Basic data collection and intelligent combustion model construction, including: full-dimensional operation data collection and intelligent combustion model construction that integrates multiple algorithms;

[0064] Step S11: Conduct full-dimensional data collection and baseline establishment;

[0065] Step S111: Establish a multi-parameter detection system. Set up temperature sensors (covering different heights, cross-sections, and burner outlets of the furnace) in the combustion opposing zone of the boiler furnace to achieve real-time temperature field detection. Install pulverized coal flow monitoring devices in each pulverized coal pipe of each coal mill to record pulverized coal velocity, flow rate, and concentration data. Simultaneously collect operating parameters such as boiler load, furnace negative pressure, and NOx concentration to form a data acquisition matrix.

[0066] Step S112: Historical data review and baseline determination. Extract historical operating data under boiler furnace load conditions (low, medium, and high loads), including the air-coal ratio of each pulverized coal pipe, temperature field distribution data, and NOx emission data. Determine the operating baseline under different operating conditions through statistical analysis, and identify the operating conditions with high current NOx generation (such as an imbalance in the air-coal ratio of a specific pulverized coal pipe or excessively high local temperature under a certain load).

[0067] Step S12: Construction of a smart combustion model integrating multiple algorithms;

[0068] Step S121: Building a smart combustion model. Using the MPC prediction model as the core framework, inputting operating parameter data consisting of boiler load and the number of coal mills in operation, constructing a mapping relationship model between NOx generation and air-coal ratio and temperature field distribution; embedding an adaptive combustion control algorithm module, establishing a combustion rule library (e.g., "when the load increases by 10%, the increase in coal volume per pipe shall not exceed 8% and the air volume shall increase synchronously by 12%) for time-varying characteristics during boiler combustion (such as coal quality fluctuations and load changes), to achieve dynamic adaptive adjustment of the smart combustion model;

[0069] Step S122: The adaptive combustion control algorithm module is based on the diagonal matrix method. Each coal mill's pulverized coal pipe is treated as an independent control unit. An optimized diagonal matrix of air-coal ratio is constructed, corresponding to the optimal value of the air-coal ratio for a single pipe (combining historical optimal data and theoretical calculations, such as an air-coal ratio of 1.2-1.4 for coal types with high volatile matter content), providing data support for single-pipe regulation.

[0070] Step S2: Real-time regulation is achieved through dynamic combustion control based on sensing data, including: real-time monitoring and analysis of temperature field and air-fuel parameters, as well as generation and execution of hierarchical regulation commands from the intelligent combustion model;

[0071] Step S21: Real-time monitoring and analysis of temperature field and air-fuel parameters;

[0072] Step S211: Real-time data synchronization and analysis. The real-time data monitored by the pulverized coal flow monitoring device is transmitted to the boiler combustion control platform through the data acquisition system. Abnormal data (such as instantaneous fluctuations of the sensor) is removed by the filtering algorithm. The visualization system of the boiler combustion control platform is used to display the furnace temperature field distribution cloud map and intuitively identify local high temperature areas. Among them, boiler combustion >1500℃ will lead to a surge in thermal NOx.

[0073] Step S212: Diagnosis of key parameter deviations. Compare real-time data with baseline data and use the adaptive combustion control algorithm module to calculate the deviation between real-time data and baseline data. The deviation data includes air-coal ratio deviation (the difference between the air-coal ratio of a single pipe and the optimal value of the diagonal matrix; if the deviation exceeds ±10%, control is triggered) and temperature field uniformity deviation (calculate the standard deviation of furnace cross-section temperature; if the standard deviation is >80℃, it is determined to be a temperature field non-uniformity).

[0074] Step S22: Generation and execution of model-based hierarchical control instructions;

[0075] Step S221: First-level control, fine adjustment of air-coal ratio in single pipe. For pulverized coal pipes with excessive air-coal ratio deviation, the intelligent combustion model outputs single pipe air volume and coal quantity adjustment instructions based on the diagonal matrix method (e.g., if the air-coal ratio of a certain pulverized coal pipe is 5% lower, the instruction is to increase its air volume by 3% or decrease its coal quantity by 2%). The adjustment is executed through the pulverizer inlet air regulating damper and the coal feeder speed controller to ensure that the air-coal ratio of each pulverized coal pipe is in the optimal range, thereby reducing NOx generation caused by incomplete combustion of fuel and local oxygen enrichment from the source.

[0076] Step S222: Secondary control and temperature field uniformity optimization. For areas with uneven temperature fields, the intelligent combustion model is used to analyze the coal-air-air ratio and burner operating status corresponding to high-temperature areas, and output coordinated adjustment commands. If the local high temperature is caused by excessive coal quantity in a certain group of burners, the coal quantity in the corresponding coal-air-air-coal ratio is adjusted through the adaptive combustion control algorithm module, and the air quantity in adjacent coal-air-air-coal ...

[0077] Step S3: Based on the feedback NOx concentration data, control the iteration and upgrading of the intelligent combustion model, including: feedback on the regulation effect and verification of NOx concentration, as well as model iteration and optimization of control strategy.

[0078] Step S31: Feedback on the regulation effect and verification of NOx concentration;

[0079] Step S311: Real-time monitoring of the control effect. After executing the command to adjust the combustion temperature, continuously monitor the target pulverized coal-air ratio, furnace temperature field distribution, and outlet NOx concentration data. The adaptive combustion control algorithm module calculates the NOx emission reduction before and after the control (e.g., a reduction of 15%-20% under low load conditions and a reduction of 10%-15% under high load conditions is considered satisfactory). At the same time, verify the boiler thermal efficiency (ensure that the thermal efficiency decrease does not exceed 1%) to avoid an imbalance between emission reduction and efficiency.

[0080] Step S312: Trace the cause of the deviation. If the NOx concentration does not meet the standard or the temperature field is still uneven after adjustment, trace the data chain to find the cause (such as sensor failure, sudden change in coal quality, or lag in model parameters), and re-execute the control process after targeted solutions.

[0081] Step S32: Model iteration and control strategy optimization;

[0082] Step S321: Model parameter update. The operating condition data, adjustment instructions, and emission reduction effect data of each adjustment are included in the model training set. The rule base is updated through the rule optimization module controlled by the adaptive combustion control algorithm module. The mapping relationship parameters in the predictive control of the adaptive combustion control algorithm module are corrected to improve the model's adaptability to complex operating conditions.

[0083] Step S322: Upgrade the control strategy, regularly analyze the control effect under different operating conditions, optimize the optimal range of single-pipe air-coal ratio using the diagonal matrix of air-coal ratio optimization, and supplement the training data of the adaptive combustion control algorithm module for newly emerging operating conditions (such as new coal types and extreme loads) to form a closed-loop control system of monitoring-control-feedback-iteration, so as to achieve continuous and refined reduction of NOx generation.

[0084] Furthermore, the pulverized coal flow monitoring device in this embodiment includes: an integrated sensor assembly and flange seats located on both sides of the integrated sensor assembly. The inner diameter of the integrated sensor assembly is the same as the inner diameter of the pulverized coal pipeline. The integrated sensor assembly consists of a coiled shell and four sub-sensors. Each sub-sensor is shaped like a quarter-circle, and the four sub-sensors are evenly distributed and embedded in the inner wall of the coiled shell.

[0085] Each of the sub-sensors includes: three electrostatic ion intensity measuring rings, two coal powder flow rate measuring rings, and an insulating ring;

[0086] The two coal powder flow rate measuring rings are spaced apart between the three electrostatic ion strength measuring rings, and the insulating ring is placed between the adjacent electrostatic ion strength measuring ring and the coal powder flow rate measuring ring.

[0087] The two coal powder flow velocity measuring rings are used to measure the coal powder flow velocity. Each ring receives a through signal from coal powder flowing in the same direction. Due to the time difference between the coal powder passing through the two rings, the coal powder flow velocity is calculated using the following formula:

[0088]

[0089] T n =y t -x t

[0090] In the formula, V represents the pulverized coal flow rate; d represents the distance between the two pulverized coal flow rate measuring rings; T n This represents the time difference between the two coal powder particles passing through the two powder flow rate measuring rings; x t y represents the time it takes for pulverized coal to pass through the first pulverized coal flow rate measurement loop. t This indicates the time it takes for pulverized coal to pass through the second pulverized coal flow rate measuring loop;

[0091] Among them, the three electrostatic ion strength measuring rings calculate the coal powder concentration based on the coal powder flow frequency. The electrostatic ion strength is positively correlated with the coal powder concentration. The current coal powder concentration C = (Qt - Q0) * K can be determined in real time by measuring the change in electrostatic ion strength between the measured electrostatic ion strength and the standard coal powder concentration.

[0092] In the formula, C represents the coal powder concentration; Qt represents the electrostatic ion intensity measured in real time inside the coal powder pipeline; Q0 represents the standard electrostatic ion intensity inside the pipeline; and K represents the coefficient relating electrostatic ion intensity and concentration.

[0093] The formula for calculating the pulverized coal flow rate is as follows:

[0094] δ=C×V×S

[0095] In the formula, δ represents the pulverized coal flow rate; S represents the cross-sectional area of ​​the pipe.

[0096] Furthermore, in this embodiment, the signal amplification processor is connected to the boiler combustion control platform. The signal amplification processor is installed on the outer pipe wall of the integrated sensor assembly. The signal amplification processor is used to transmit the data measured by the four sub-sensors to the air-coal parameter measurement experimental module of the boiler combustion control platform in real time. The integrated sensor assembly includes: a signal loading and staining ring and a signal balancing ring. The signal loading and staining ring and the signal balancing ring are respectively set at both ends. An insulating ring is set between the signal loading and staining ring and the adjacent electrostatic ion strength measurement ring. The signal loading and staining ring is used to collect the signal uniformly and transmit it to the signal amplification processor. The signal balancing ring is used to balance and distinguish the signals between the three electrostatic ion strength measurement rings and the two coal powder flow rate measurement rings.

[0097] Furthermore, the coal powder parameter measurement experimental module in this embodiment includes: a coal powder flow rate analysis module, a flow rate analysis module, a concentration analysis module, a coal powder particle fineness analysis module, and a calorific value analysis module. The coal powder flow rate analysis module is used to monitor the coal powder flow rate, the flow rate analysis module is used to monitor the coal powder flow rate, the concentration analysis module is used to monitor the coal powder concentration, the coal powder particle fineness analysis module uses the concentration analysis module, the coal powder flow rate analysis module, the flow rate analysis module, the concentration analysis module, and the pipe size data to obtain the coal powder particle fineness, and the calorific value analysis module uses the concentration analysis module, the coal powder flow rate analysis module, the flow rate analysis module, and the concentration analysis module to obtain the calorific value of the mixed coal powder.

[0098] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A combustion control method for reducing the amount of NOx generated by a boiler based on a smart combustion model, characterized by, Comprise the following steps: Step S1: basic data acquisition and intelligent combustion model construction, comprising: full-dimensional operation data acquisition and fusion of multi-algorithm intelligent combustion model construction; Step S2: based on the dynamic combustion control of perception data Real-time regulation, including: temperature field and wind coal parameter real-time monitoring and analysis, and intelligent combustion model hierarchical regulation instruction generation and execution; Step S3: according to the feedback of NOx concentration data, control the iteration and upgrade of intelligent combustion model, including: feedback of regulation effect and verification of NOX concentration, and model iteration and control strategy optimization.

2. The combustion control method for reducing the amount of NOx generated by a boiler based on a smart combustion model according to claim 1, characterized by, In step S1 also includes the following steps: Step S11: full-dimensional operation data acquisition and baseline establishment; Step S111: build multi-parameter detection system, set temperature sensor in the combustion hedge area of boiler furnace, realize real-time detection of temperature field; Install coal flow monitoring device in each root powder pipe of each coal mill, record coal flow velocity, flow, concentration data with coal flow monitoring device; Synchronous acquisition of boiler load, furnace negative pressure, NOx concentration operation parameters, form data acquisition matrix; Step S112: historical data analysis and baseline determination, extract historical operation data under load condition in boiler furnace, including each powder pipe wind coal ratio, temperature field distribution data, NOx emission data, determine the operation baseline under different conditions through statistical analysis, and clarify the current NOx generation condition of high.

3. The combustion control method for reducing the NOx generation amount of a boiler based on the intelligent combustion model according to claim 2, characterized by, In step S1 also includes the following steps: Step S12: fusion of multi-algorithm intelligent combustion model construction; Step S121: intelligent combustion model construction, taking MPC prediction model as core framework, inputting working condition parameter data composed of boiler load and coal mill operation station, constructing mapping relationship model of NOx generation and wind coal ratio, temperature field distribution; Embed adaptive combustion control algorithm module, establish combustion rule library according to time-varying characteristics in boiler combustion process, realize dynamic self-adaptive adjustment of intelligent combustion model; Step S122: adaptive combustion control algorithm module based on diagonal matrix method, each root powder pipe of each coal mill is regarded as independent control unit, wind coal ratio optimization diagonal matrix is constructed, corresponding to single pipe wind coal ratio optimal value, providing data support for single pipe regulation.

4. The combustion control method for reducing the NOx generation amount of a boiler based on a smart combustion model according to claim 1, characterized by, In step S2 also includes the following steps: Step S21: real-time monitoring and analysis of temperature field and wind coal parameters; Step S211: real-time data synchronization and analysis, transmit real-time data monitored by coal flow monitoring device to boiler combustion control platform through data acquisition system, adopt filtering algorithm to eliminate abnormal data; Use the visualization system of boiler combustion control platform to display furnace temperature field distribution cloud picture, intuitively identify local high temperature area, wherein, boiler combustion > 1500℃ will lead to thermal NOx surge; Step S212: key parameter deviation diagnosis, compare real-time data with baseline data, calculate the deviation of real-time data and baseline data by using adaptive combustion control algorithm module, wherein, deviation data includes wind coal ratio deviation and temperature field uniformity deviation.

5. The combustion control method for reducing the NOx generation amount of a boiler based on the intelligent combustion model according to claim 4, characterized by, In step S2 also includes the following steps: Step S22: hierarchical regulation instruction generation and execution based on model; Step S221: primary regulation, fine adjustment of single-tube air-coal ratio, for the powder tube with air-coal ratio deviation exceeding the standard, the intelligent combustion model outputs single-tube air volume and coal quantity adjustment instructions based on the diagonal matrix method, and the adjustment is executed through the inlet air damper of the coal mill and the coal feeder speed controller, so that the air-coal ratio of each powder tube is in the optimal interval, and the incomplete combustion of fuel and the generation of NOx caused by local oxygen enrichment are reduced from the source; Step S222: secondary regulation, temperature field uniformity optimization, for the temperature field uneven area, the intelligent combustion model is used to analyze the corresponding powder tube air-coal ratio and the burner operation state of the high temperature area, and the coordinated adjustment instruction is output. If the local high temperature is caused by too high coal quantity of a group of burners, the corresponding powder tube coal quantity is controlled and adjusted through the self-adaptive combustion control algorithm module, and the air quantity of the adjacent powder tube is fine-tuned to guide the uniform distribution of the flame center. If the overall temperature is too high, the primary air temperature or the secondary air ratio is adjusted to control the average temperature of the furnace to be in the low-NOx combustion interval of 1350-1450℃ under the premise of ensuring the combustion efficiency.

6. The combustion control method for reducing the NOx generation amount of a boiler based on a smart combustion model according to claim 1, characterized by, The following steps are further included in step S3: Step S31: regulation effect feedback and NOx concentration verification; Step S311: real-time monitoring of regulation effect, after executing the adjustment combustion temperature instruction, the target powder tube air-coal ratio, the furnace temperature field distribution and the outlet NOx concentration data are continuously monitored, the self-adaptive combustion control algorithm module calculates the NOx emission reduction amplitude before and after the regulation, and the boiler thermal efficiency is verified to avoid the imbalance between emission reduction and efficiency; Step S312: deviation reason tracing, if the NOx concentration does not meet the standard or the temperature field is still uneven after the regulation, the data chain is traced to find out the reason and the regulation process is executed again after solving the problem.

7. The combustion control method for reducing the NOx generation amount of a boiler based on the intelligent combustion model according to claim 6, characterized by, The following steps are further included in step S3: Step S32: model iteration and control strategy optimization; Step S321: model parameter updating, the working condition data, adjustment instruction and emission reduction effect data of each regulation are included in the model training set, the rule library is updated through the rule optimization module of the self-adaptive combustion control algorithm module, and the mapping relationship parameters in the prediction control of the self-adaptive combustion control algorithm module are corrected; Step S322: control strategy upgrade, the regulation effect under different working conditions is regularly counted, the optimal value interval of single-tube air-coal ratio is optimized by air-coal ratio optimization diagonal matrix, the training data of the self-adaptive combustion control algorithm module is supplemented for the newly appeared working condition, a closed-loop control system of monitoring-regulation-feedback-iteration is formed, and the continuous fine reduction of NOx generation is realized.

8. The combustion control method for reducing the NOx production amount of a boiler based on the intelligent combustion model according to claim 2, characterized by, The coal powder flow monitoring device comprises an integrated sensor assembly and flange disc seats located on both sides of the integrated sensor assembly, the inner diameter of the integrated sensor assembly is the same as the inner diameter of the coal powder pipeline, and the integrated sensor assembly is composed of a coiled pipe shell and four sub-sensors. Each sub-sensor is in the shape of a quarter ring and is evenly inlaid in the inner wall of the coiled pipe shell. Each sub-sensor comprises three electrostatic ion intensity measuring rings, two coal powder flow velocity measuring rings, and an insulating ring. Two coal flow velocity measurement rings are arranged between three electrostatic ion intensity measurement rings, and the insulating rings are arranged between adjacent electrostatic ion intensity measurement rings and coal flow velocity measurement rings. Two coal flow velocity measurement rings are used to measure the coal flow velocity, and the through signals generated by the coal in the same flow direction are received by the two coal flow velocity measurement rings respectively. Since the coal passes between the two coal flow velocity measurement rings, there is a time difference, and the coal flow velocity calculation formula is: T n = y t - x t where V represents the coal flow velocity; d represents the distance between the two coal flow velocity measurement loops; T n represents the time difference between the coal particle passing through the two coal flow velocity measurement loops; x t represents the time of the coal passing through the first coal flow velocity measurement loop, y t represents the time of the coal passing through the second coal flow velocity measurement loop; Three electrostatic ion intensity measurement rings are used to calculate the coal concentration by the coal flow frequency. The intensity of the electrostatic ion is positively correlated with the concentration of the coal. The current coal concentration C=(Qt-Q0)*K can be measured in real time by the change of the standard coal concentration electrostatic ion intensity and the real-time measured electrostatic ion intensity. In the formula, C represents the coal concentration; Qt represents the real-time measured electrostatic ion intensity in the coal pipeline; Q0 represents the standard electrostatic ion intensity in the pipeline; and K represents the relationship coefficient between the electrostatic ion intensity and the concentration. The coal flow calculation formula is: δ=C×V×S In the formula, δ represents the coal flow; and S represents the cross-sectional area of the pipeline.

9. The combustion control method for reducing the NOx production amount of a boiler based on the intelligent combustion model according to claim 2, characterized by, The signal amplification processor is connected with the boiler combustion control platform, and is installed on the external pipe wall of the integrated sensor assembly. The signal amplification processor is used to transmit the data measured by the four sub-sensors to the wind-powder parameter measurement experiment module of the boiler combustion control platform in real time. The integrated sensor assembly includes a signal loading dyeing ring and a signal balancing ring, which are arranged at both end positions respectively. Insulating rings are arranged between the signal loading dyeing ring and the signal balancing ring and the adjacent electrostatic ion intensity measurement rings. The signal loading dyeing ring is used to uniformly collect signals and transmit them to the signal amplification processor. The signal balancing ring is used to balance and distinguish the signals between the three electrostatic ion intensity measurement rings and the two coal flow velocity measurement rings.

10. The pulverized coal flow monitoring device for a deeply peak-shaving coal-fired boiler according to claim 9, characterized in that, The wind-powder parameter measurement experiment module includes a coal flow velocity analysis module, a flow analysis module, a concentration analysis module, a coal particle fineness analysis module and a calorific value analysis module. The coal flow velocity analysis module is used to monitor the coal flow velocity. The flow analysis module is used to monitor the coal flow. The concentration analysis module is used to monitor the coal concentration. The coal particle fineness analysis module uses the concentration analysis module, the coal flow velocity analysis module, the flow analysis module, the concentration analysis module and the pipeline size data to obtain the coal particle fineness. The calorific value analysis module uses the concentration analysis module, the coal flow velocity analysis module, the flow analysis module and the concentration analysis module to obtain the calorific value of the mixed coal.

Images