Thermal power generating unit deep peak regulation optimization method

Abstract

The invention relates to the technical field of thermal power generation automatic control, and discloses a thermal power generating unit deep peak regulation optimization method, which constructs a system comprising an auxiliary machine adaptive transformation module, a three-level control architecture module, a load mapping module, a collaborative regulation module, a combustion environmental protection module, a data sensing module, a state monitoring module and a self-adaptive optimization module. A dynamic load-working condition prediction model is established through an LSTM neural network, the fuel quantity is pre-controlled, a pressure correction value is adopted to accurately restore the main steam pressure, the air-fuel ratio is dynamically adjusted, an SCR system is linked to stabilize the flue gas temperature and control NOx emission, data are collected in real time by deploying a 100 Hz high-frequency sensor network, an error compensation instruction is triggered, and the accuracy and the reliability of the system are improved. And evaluating the water erosion risk of the last-stage blade of the steam turbine based on the water erosion risk index, and finally outputting a PID self-tuning parameter, an air-fuel ratio set value, an ammonia injection amount correction instruction and a water erosion risk threshold adjustment parameter, thereby realizing dynamic iterative optimization of system parameters.

Description

Technical Field

[0001] This invention relates to the field of automatic control and energy conservation and environmental protection technology for thermal power generation, specifically a method for deep peak shaving optimization of thermal power units. Background Technology

[0002] With the increasing proportion of renewable energy power generation, the demand for deep peak-shaving capabilities of thermal power units is becoming increasingly urgent. However, thermal power units face a series of severe challenges when operating under low-load conditions: First, the design operating conditions of traditional auxiliary systems (such as coal mills and fans) deviate from the low-load zone, posing safety risks such as unstable combustion and fan surge; second, the huge thermal inertia of the boiler leads to sluggish load response, making it difficult for conventional PID control to achieve rapid and accurate tracking of load and pressure, resulting in large fluctuations in main steam pressure and affecting the stability and economy of the unit; third, the decrease in furnace temperature under low load leads to changes in the initial NOx formation, making it difficult to put the SCR system into operation, and exacerbating the difficulty of environmental emission control; in addition, the turbine's last-stage blades operate in the wet steam zone for a long time, significantly increasing the risk of water erosion damage, seriously threatening equipment safety and unit lifespan. Existing technologies mostly focus on the local optimization of single systems or parameters, lacking a global solution that includes equipment hardware modification, intelligent feedforward prediction, cross-system collaborative control, and adaptive equipment health status, making it difficult to simultaneously achieve safety, stability, economy, and environmental protection in deep peak-shaving of the unit. Summary of the Invention

[0003] (a) Technical problems to be solved

[0004] To address the shortcomings of existing technologies, this invention provides a method for optimizing deep peak shaving in thermal power units. This method has the advantages of improving peak shaving depth and speed, enhancing operational stability, ensuring environmental compliance, and extending equipment life. It solves the comprehensive problems of unstable combustion, large fluctuations in main steam pressure, delayed load response, difficulties in environmental control, and high risk of water erosion damage to the last stage blades of turbines in existing thermal power units operating at low loads.

[0005] (II) Technical Solution

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for deep peak shaving optimization of thermal power units, comprising the following steps:

[0007] Step 1: Establish a deep peak shaving and optimization system for thermal power units. The system includes an auxiliary equipment adaptation and modification module, a three-level control architecture module, a load mapping module, a coordinated regulation module, a combustion environmental protection module, a data sensing module, a status monitoring module, and an adaptive optimization module.

[0008] Step 2: The auxiliary equipment adaptation and modification module is used to perform low-load performance modification on the coal mill, fan, and turbine last-stage blades;

[0009] Step 3: Build a three-level control architecture module to construct a three-level framework of unit-system-equipment and establish a cross-system coupling model;

[0010] Step 4: The load mapping module constructs a dynamic load-condition prediction model based on an LSTM neural network and integrates a load pre-control fuel quantity calculation unit to execute load pre-control fuel quantity calculation. Calculate and generate pre-controlled fuel quantity commands;

[0011] Step 5: The coordinated control module synchronously acquires the main steam set pressure and measured pressure data, and calculates the pressure correction value. To accurately correct the main steam pressure deviation and generate a pressure deviation adjustment strategy;

[0012] Step Six: The combustion environmental protection module dynamically receives the pre-controlled fuel quantity command output by the load mapping module, dynamically adjusts the air-fuel ratio, and links with the SCR system to stabilize the flue gas temperature and control NO. x emission;

[0013] Step 7: The data sensing module collects real-time monitoring data of the thermal system and environmental indicators by deploying a high-frequency sensor network, and calculates compensation command values ​​based on the collected data. This triggers an error compensation command;

[0014] Step 8: The condition monitoring module collects auxiliary machine operation data and equipment health status data through the industrial fieldbus, analyzes equipment vibration and stress, and calculates the water erosion risk index. Assess the risk of water erosion on the last stage blades of a steam turbine;

[0015] Step 9: The adaptive optimization module, based on the analysis results output by the data sensing module and the condition monitoring module, adaptively adjusts the deep peak shaving optimization system of the thermal power unit, and outputs the correction coefficient optimization value to the load mapping module, the PID self-tuning parameter to the collaborative adjustment module, the air-fuel ratio setpoint and ammonia injection correction command to the combustion environmental protection module, and the water erosion risk threshold adjustment parameter to the condition monitoring module, thereby realizing the dynamic iterative optimization of system parameters.

[0016] Preferably, the data sensing module includes a thermal system monitoring unit and an environmental protection indicator monitoring unit; the status monitoring module includes an auxiliary machine operation monitoring unit and an equipment health monitoring unit.

[0017] Preferably, the auxiliary machine operation monitoring unit collects auxiliary machine operation monitoring data online via an industrial fieldbus, including coal mill outlet air pressure, coal mill inlet temperature, fan bearing temperature, coal mill current, fan speed, and inverter frequency data.

[0018] Preferably, the thermal system monitoring unit collects thermal system monitoring data, including main steam pressure, steam humidity, main steam temperature, reheat steam temperature, feedwater flow rate, and flue gas temperature data, by installing high-temperature pressure transmitters, capacitive humidity sensors, and armored thermocouples installed in the main steam pipeline, turbine inlet chamber, and economizer inlet flue.

[0019] Preferably, the environmental protection indicator monitoring unit uses laser scattering dust collectors, zirconia oxygen analyzers, and chemiluminescence NO analyzers installed at the inlet and outlet of the denitrification reactor, the inlet of the desulfurization tower, and the inlet of the chimney. x The analyzer collects environmental indicator monitoring data, including particulate matter concentration, flue gas oxygen content, and NO. x Data on concentration, SO2 concentration, flue gas temperature, and ammonia injection rate.

[0020] Preferably, the equipment health monitoring unit collects equipment health monitoring data online through the TSI online monitoring system, including rotor eccentricity, lubricating oil temperature, axial displacement, bearing vibration, metal wall temperature, alternating stress, and blade water erosion signals.

[0021] Preferably, the load mapping module constructs a dynamic load-condition prediction model based on an LSTM neural network and calculates the load pre-control fuel quantity. The calculation formula is as follows: In the formula, Indicates the pre-controlled fuel quantity. Indicates the fuel quantity at rated load. Indicates the target load. Indicates the rated load. This represents the combustion field temperature correction factor. This represents the coal powder concentration correction factor.

[0022] Preferably, the coordinated adjustment module calculates the pressure correction value. The formula for accurately correcting main steam pressure deviation is as follows: In the formula, This indicates the pressure correction value. This indicates the setting of pressure. Indicates the measured pressure. , , This represents the PID self-tuning parameters.

[0023] Preferably, the data sensing module calculates the compensation command value. The error compensation command is triggered, and its calculation formula is as follows: In the formula, Indicates the compensation instruction value. For steady-state deviation, Indicates the integral compensation coefficient. This represents the proportional compensation coefficient.

[0024] Preferably, the condition monitoring module collects auxiliary machine operation data and equipment health status data through an industrial fieldbus, and calculates the water erosion risk index. The formula for assessing the water erosion risk of the last-stage blades of a steam turbine is as follows: In the formula, Indicates the water erosion risk index. This represents the water erosion sensitivity coefficient of the material. Indicates the density of wet vapor. This indicates the linear velocity at the blade tip. Indicates the cumulative running time under low load. Indicates the coefficient of humidity non-uniformity. This indicates the yield strength of the blade material.

[0025] Compared with the prior art, the present invention provides a deep peak shaving optimization method for thermal power units, which has the following beneficial effects:

[0026] 1. This invention uses an auxiliary equipment adaptation and modification module to modify the low-load performance of coal mills, fans, and turbine last-stage blades, providing a solid equipment foundation for the safe, stable, efficient, and environmentally friendly operation of the entire deep peak shaving and optimization system under low load. It can achieve the beneficial effects of widening the stable operating load range of the unit, ensuring the reliability of key equipment, and reducing operational risks from the source.

[0027] 2. This invention pre-controls fuel quantity by calculating load. Using this as the core control benchmark, it is incorporated into the coordinated adjustment module and the combustion environmental protection module to control the advance amount of fuel supply and coordinate the air-fuel ratio and ammonia injection quantity. This effectively compensates for the adjustment lag caused by boiler thermal inertia, avoids combustion instability caused by fuel-load mismatch under low load, and achieves shorter load response time, improved combustion efficiency, and reduced NO. x The beneficial effect of reducing initial emission concentrations.

[0028] 3. This invention calculates the pressure correction value. Using this as the basis for precise control of the main steam pressure, it is incorporated into the cross-system coupling model of the three-level control architecture module, synchronously triggering boiler fuel adjustment and turbine valve action to achieve real-time compensation for main steam pressure deviation, strictly controlling pressure fluctuations within ±0.2MPa, and achieving the beneficial effects of improving the stability of the thermal system under low load, enhancing the safety of turbine operation, and reducing power generation efficiency loss.

[0029] 4. This invention calculates the compensation command value. This signal is used as a dynamic error compensation signal for fuel and air volume. It is then substituted into the combustion environmental protection module and the load mapping module to perform secondary correction on the pre-controlled fuel quantity and air-fuel ratio. This accurately offsets the adjustment error caused by steady-state deviation, achieving the beneficial effects of improved fuel supply accuracy, reduced deviation in flue gas oxygen content control, and further enhanced combustion stability.

[0030] 5. This invention calculates the water erosion risk index. This is used as the safety boundary criterion for the last stage blades of the steam turbine. It is then substituted into the adaptive optimization module and the three-level control architecture module to dynamically adjust the low-load operating range and water erosion risk threshold, triggering the blade protection strategy. This achieves the beneficial effects of early warning of blade water erosion risk, extending blade service life, avoiding unplanned shutdowns caused by water erosion under low load, and ensuring the safety boundary of deep peak shaving of the unit. Attached Figure Description

[0031] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] Please see Figure 1 A method for deep peak shaving optimization of thermal power units includes the following steps:

[0034] Step 1: Establish a deep peak shaving and optimization system for thermal power units. The system includes an auxiliary equipment adaptation and modification module, a three-level control architecture module, a load mapping module, a coordinated regulation module, a combustion environmental protection module, a data sensing module, a status monitoring module, and an adaptive optimization module.

[0035] Step 2: The auxiliary equipment adaptation and modification module performs low-load performance modifications on the coal mill, fan, and turbine last-stage blades, specifically as follows:

[0036] (1) Coal mill: Add a dynamic rotary separator and optimize the profile of the grinding components to improve the fineness and uniformity of coal powder under low load. By improving the fineness of coal powder, it can ensure stable ignition and complete combustion when the coal powder is fed at low concentration, thus solving the problem of unstable combustion under low load.

[0037] (2) Fan: It adopts high-efficiency wide-load three-dimensional flow blades and is equipped with a wide-range variable frequency speed control system to expand the high-efficiency working range of the fan. Through variable frequency speed control, it accurately matches the air volume demand under low load conditions, effectively avoids the huge energy loss caused by the throttling of the inlet guide vanes, prevents the fan from entering the stall or surge zone, and ensures safe operation.

[0038] (3) Last stage blades of steam turbine: The top of the steam inlet edge of the last stage blades is subjected to water-resistant and corrosion-resistant Stellite alloy welding and surface laser hardening treatment. At the same time, the blade profile is optimized to improve the flow field of wet steam under low load. This can greatly improve the blades' resistance to water erosion, directly address the risk of water erosion damage to the blades caused by increased exhaust humidity under low load, and extend the service life of core components.

[0039] Through the above targeted hardware modifications, a solid equipment foundation has been provided for the safe, stable, efficient and environmentally friendly operation of the entire deep peak shaving and optimization system under low load conditions. This can achieve the beneficial effects of widening the stable operating load range of the unit, ensuring the reliability of key equipment and reducing operational risks from the source.

[0040] Step 3: The three-level control architecture module and its cross-system coupling model are the central nervous system and command system of the entire optimization system. They integrate all independent modules into an organic whole, so that the potential for hardware transformation can be fully utilized through intelligent software coordination, ultimately achieving the comprehensive effect of improving peak shaving depth, ensuring operational safety and reducing operating costs.

[0041] Step 4: The load mapping module constructs a dynamic load-condition prediction model based on an LSTM neural network and integrates a load pre-control fuel quantity calculation unit to execute load pre-control fuel quantity calculation. Calculate and generate pre-controlled fuel quantity commands;

[0042] Step 5: The coordinated control module synchronously acquires the main steam set pressure and measured pressure data (thermal system monitoring unit), and calculates the pressure correction value. To accurately correct the main steam pressure deviation and generate a pressure deviation adjustment strategy;

[0043] Step Six: The combustion environmental protection module dynamically receives the pre-controlled fuel quantity command output by the load mapping module, dynamically adjusts the air-fuel ratio, and links with the SCR system to stabilize the flue gas temperature and control NO. x Emissions are controlled to ensure that the unit meets environmental protection standards under low load conditions;

[0044] Step 7: The data sensing module collects real-time monitoring data of the thermal system (such as main steam pressure and temperature) and environmental indicators (such as NOx concentration) by deploying a 100Hz high-frequency sensor network, and calculates the compensation command value based on the collected data. This triggers error compensation commands, providing timely data support for upper-level modules;

[0045] Step 8: The condition monitoring module collects auxiliary machine operating data and equipment health status data through an industrial fieldbus (such as Profibus), analyzes equipment vibration and stress, and calculates the water erosion risk index. Assess the water erosion risk of the last stage blades of the steam turbine to provide equipment safety boundaries for system optimization;

[0046] Step 9: The adaptive optimization module, based on the analysis results output by the data sensing module and the condition monitoring module, adaptively adjusts the deep peak shaving optimization system of the thermal power unit, and outputs the correction coefficient optimization value to the load mapping module, the PID self-tuning parameter to the collaborative adjustment module, the air-fuel ratio setpoint and ammonia injection correction command to the combustion environmental protection module, and the water erosion risk threshold adjustment parameter to the condition monitoring module, thereby realizing the dynamic iterative optimization of system parameters.

[0047] The advantages are: through the collaborative cooperation of the above-mentioned load mapping module, coordinated regulation module, combustion environmental protection module, data sensing module, status monitoring module and adaptive optimization module, load forecasting, pressure regulation, combustion optimization, data compensation, equipment monitoring and system self-adaptation are achieved, and comprehensive optimization of deep peak shaving of thermal power units is carried out, which can achieve the beneficial effects of improving peak shaving capacity, ensuring stable operation, reducing environmental emissions and extending equipment life.

[0048] The data sensing module includes a thermal system monitoring unit and an environmental protection indicator monitoring unit; the status monitoring module includes an auxiliary machine operation monitoring unit and an equipment health monitoring unit.

[0049] The auxiliary machine operation monitoring unit collects auxiliary machine operation monitoring data online via the industrial fieldbus (Profibus-DP), including coal mill outlet air pressure, coal mill inlet temperature, fan bearing temperature, coal mill current, fan speed, and frequency converter frequency data.

[0050] The thermal system monitoring unit collects thermal system monitoring data, including main steam pressure, steam humidity, main steam temperature, reheat steam temperature, feedwater flow rate, and flue gas temperature, by installing high-temperature pressure transmitters, capacitive humidity sensors, and armored thermocouples in the main steam pipeline, turbine inlet chamber, and economizer inlet flue.

[0051] The environmental protection indicator monitoring unit uses laser scattering dust analyzers, zirconia oxygen analyzers, and chemiluminescence NO analyzers installed at the inlet and outlet of the denitrification reactor, the inlet of the desulfurization tower, and the inlet of the chimney. x The analyzer collects environmental indicator monitoring data, including particulate matter concentration, flue gas oxygen content, and NO. x Data on concentration, SO2 concentration, flue gas temperature, and ammonia injection rate.

[0052] The equipment health monitoring unit collects equipment health monitoring data online through the TSI (Turbine Monitoring Instrument) online monitoring system, including rotor eccentricity, lubricating oil temperature, axial displacement, bearing vibration, metal wall temperature, alternating stress, and blade water erosion signals.

[0053] The advantages are: through the predictive-correction closed-loop collaborative cooperation of the load mapping module and the coordinated regulation module, combined with the high-frequency data support of the data sensing module and the dynamic adaptation of the combustion environmental protection module, a full-process optimization system of load prediction, fuel pre-control, pressure fine adjustment and environmental protection adaptation is constructed. This achieves precise matching of fuel supply, main steam pressure and environmental protection indicators under low load conditions, and achieves the beneficial effects of "improved load response speed, enhanced operational stability, compliance with environmental emission standards and significant reduction in energy consumption" during deep peak shaving of the unit.

[0054] The load mapping module constructs a dynamic load-condition prediction model based on an LSTM neural network and calculates the pre-controlled fuel quantity for load conditions. The calculation formula is as follows: In the formula, This indicates the pre-controlled fuel quantity (t / h). This indicates the fuel quantity at rated load (t / h). Indicates the target load (MW). Indicates the rated load (MW). This represents the combustion field temperature correction factor (based on combustion environmental monitoring unit data, with a value of 0.95-1.05). This represents the coal powder concentration correction coefficient (based on combustion environmental monitoring unit data, with a value of 0.98-1.02); it uploads the pre-controlled fuel quantity command to the three-level control architecture module to provide a basis for coordinated regulation.

[0055] The advantage is that the fuel quantity can be pre-controlled by calculating the load. Using this as the core control benchmark, it is incorporated into the coordinated adjustment module and the combustion environmental protection module to control the advance amount of fuel supply and coordinate the air-fuel ratio and ammonia injection quantity. This effectively compensates for the adjustment lag caused by boiler thermal inertia, avoids combustion instability caused by fuel-load mismatch under low load, and achieves shorter load response time, improved combustion efficiency, and reduced NO. x The beneficial effect of reducing initial emission concentrations.

[0056] The coordinated adjustment module calculates the pressure correction value. The formula for accurately correcting main steam pressure deviation is as follows: In the formula, This indicates the pressure correction value (MPa). Indicates the set pressure (MPa). This indicates the measured pressure (MPa, from the thermal system monitoring unit). , , Indicates PID self-tuning parameters ( , , The values ​​are in the range of 0.3-0.8, 0.01-0.05, and 0.1-0.3 respectively. Based on the pre-controlled fuel quantity and pressure correction value, the boiler fuel adjustment and turbine control valve operation commands are issued simultaneously to compensate for the boiler thermal inertia and control the main steam pressure fluctuation within ±0.2MPa.

[0057] The advantage is that it calculates the pressure correction value. Using this as the basis for precise control of the main steam pressure, it is incorporated into the cross-system coupling model of the three-level control architecture module, synchronously triggering boiler fuel adjustment and turbine valve action to achieve real-time compensation for main steam pressure deviation, strictly controlling pressure fluctuations within ±0.2MPa, and achieving the beneficial effects of improving the stability of the thermal system under low load, enhancing the safety of turbine operation, and reducing power generation efficiency loss.

[0058] The data sensing module calculates the compensation command value. The error compensation command is triggered, and its calculation formula is as follows: In the formula, Indicates the compensation instruction value. For steady-state deviation (when | Triggered at 5% (|>5%) This represents the integral compensation coefficient (0.02-0.08). This represents the proportional compensation coefficient (0.5-1.2), used for precise compensation of fuel supply and air volume regulation.

[0059] The advantage is that it calculates the compensation command value. This signal is used as a dynamic error compensation signal for fuel and air volume. It is then substituted into the combustion environmental protection module and the load mapping module to make secondary corrections to the pre-controlled fuel quantity and air-fuel ratio, accurately offsetting the adjustment error caused by steady-state deviation (rapid response when |e(t)|>5%). This achieves the beneficial effects of improved fuel supply accuracy, reduced deviation in flue gas oxygen content control, and further enhanced combustion stability.

[0060] The condition monitoring module collects auxiliary machine operation data and equipment health status data through an industrial fieldbus (such as Profibus) and calculates the water erosion risk index. The formula for assessing the water erosion risk of the last-stage blades of a steam turbine is as follows: In the formula, The water erosion risk index (dimensionless) is used to quantify the degree of water erosion damage to blades; a higher value indicates a higher risk. The water erosion sensitivity coefficient (1 / MPa) is determined by the characteristics of the blade material and reflects the material's resistance to water erosion. This indicates the density of wet vapor (kg / m³), obtained through a humidity monitoring unit. Higher density indicates higher water droplet impact energy. The blade tip linear velocity (m / s) is calculated from a speed sensor and the blade geometry. This indicates the cumulative operating time under low load (h), which represents the duration of unit operation in low load areas prone to water erosion. The coefficient representing the non-uniformity of humidity distribution (dimensionless) is based on simulation data of the final-stage channel flow field and ranges from 1.2 to 1.8. The yield strength (MPa) of the blade material is provided by the material property parameter library.

[0061] The advantage is that it allows for the calculation of water erosion risk index. This is used as the safety boundary criterion for the last stage blades of the steam turbine. It is then substituted into the adaptive optimization module and the three-level control architecture module to dynamically adjust the low-load operating range and water erosion risk threshold, triggering blade protection strategies (such as load fine-tuning and steam humidity control). This achieves the beneficial effects of early warning of blade water erosion risk, extending blade service life, avoiding unplanned shutdowns caused by water erosion under low load, and ensuring the safety boundary of deep peak shaving of the unit.

[0062] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for deep peak regulation optimization of a thermal power unit, characterized in that, The method comprises the following steps: Step one, establishing a deep peak regulation optimization system for thermal power generating units, which is provided with an auxiliary machine adaptation and modification module, a three-level control architecture module, a load mapping module, a coordinated regulation module, a combustion and environmental protection module, a data sensing module, a state monitoring module and a self-adaptive optimization module; Step two, the auxiliary machine adaptation and modification module performs low-load performance modification on the coal mill, the fan and the last-stage blades of the steam turbine; Step three, the three-level control architecture module builds a three-level framework of the unit, the system and the equipment and establishes a cross-system coupling model; Step four, the load mapping module constructs a dynamic load-working condition prediction model based on an LSTM neural network, integrates a load pre-control fuel quantity calculation unit, and executes load pre-control fuel quantity calculation, and generates a pre-control fuel quantity instruction; Step five, the coordinated regulation module synchronously acquires the main steam set pressure and measured pressure data, calculates the pressure correction value to accurately repair the main steam pressure deviation, and generates a pressure deviation adjustment strategy; Step six, the combustion environmental protection module dynamically receives the pre-control fuel quantity instruction output by the load mapping module, dynamically adjusts the air-fuel ratio, and links the SCR system to stabilize the flue gas temperature and control NO x emissions; Step seven, the data sensing module collects real-time monitoring data of the heat supply system and environmental protection indicators by deploying a high-frequency sensor network, calculates a compensation instruction value according to the collected data, and triggers an error compensation instruction . Step eight, the state monitoring module collects auxiliary machine operation data and equipment health state data through the industrial field bus, analyzes equipment vibration and stress, and calculates water erosion risk index Evaluate the water erosion risk of the last stage blade of the steam turbine; Step nine, the self-adaptive optimization module performs self-adaptive regulation on the deep peak regulation optimization system for thermal power generating units based on the analysis results output by the data sensing module and the state monitoring module, and outputs a correction coefficient optimization value to the load mapping module, PID self-tuning parameters to the coordinated regulation module, an air-fuel ratio set value and an ammonia injection amount correction instruction to the combustion and environmental protection module, and a water erosion risk threshold adjustment parameter to the state monitoring module, so as to realize dynamic iterative optimization of system parameters.

2. The method of claim 1, wherein: The data sensing module comprises a thermal system monitoring unit and an environmental protection index monitoring unit; the state monitoring module comprises an auxiliary machine operation monitoring unit and an equipment health monitoring unit.

3. The method of claim 2, wherein: The auxiliary machine operation monitoring unit collects auxiliary machine operation monitoring data online through an industrial field bus, including mill outlet air pressure, mill inlet temperature, fan bearing temperature, mill current, fan speed and frequency data of the frequency converter.

4. The method of claim 2, wherein: The thermal system monitoring unit collects thermal system monitoring data through high-temperature pressure transmitters, capacitive humidity sensors and armored thermocouples installed on the main steam pipeline, the steam turbine admission chamber and the coal economizer inlet flue, including main steam pressure, steam humidity, main steam temperature, reheat steam temperature, feedwater flow and flue gas temperature data.

5. The method of claim 2, wherein: The environmental protection index monitoring unit is provided with a laser scattering type smoke dust instrument, a zirconium oxide oxygen analyzer and a chemiluminescence NO x The analyzer collects environmental protection index monitoring data, including smoke concentration, flue gas oxygen content, NO x Concentration, SO2 concentration, flue gas temperature and ammonia injection amount data.

6. The method of claim 2, wherein: The equipment health monitoring unit collects equipment health monitoring data online through a TSI online monitoring system, including rotor eccentricity value, lubricating oil temperature, axial displacement, bearing vibration, metal wall temperature, alternating stress and blade water erosion signal.

7. The method of Claim 1, wherein: The load mapping module constructs a dynamic load-working condition prediction model based on an LSTM neural network, and calculates a pre-control fuel amount of the load The calculation formula is: In the formula, represents a pre-control fuel amount, represents a rated load fuel amount, represents a target load, represents a rated load, represents a combustion field temperature correction coefficient, represents a pulverized coal concentration correction coefficient.

8. The method of Claim 1, wherein: The synergic regulation module accurately repairs the main steam pressure deviation by calculating a pressure correction value The calculation formula is: In the formula, represents a pressure correction value, represents a set pressure, represents a measured pressure, , , represents a PID self-tuning parameter.

9. The method of Claim 1, wherein: The data-aware module calculates a compensation instruction value The trigger error compensation instruction is calculated by the formula: In the formula, represents a compensation instruction value, is a steady-state deviation, represents an integral compensation coefficient, represents a proportional compensation coefficient.

10. The method of claim 1, wherein: The state monitoring module collects auxiliary machine operation data and equipment health state data through an industrial field bus, and calculates a water erosion risk index to evaluate the water erosion risk of the last-stage blade of the steam turbine, and the calculation formula is: in the formula, represents the water erosion risk index, represents the material water erosion sensitivity coefficient, represents the wet steam density, represents the blade tip line speed, represents the low load cumulative running time, represents the humidity distribution unevenness coefficient, represents the yield strength of the blade material.

Images