Method and system for on-line real-time monitoring and control of high-temperature corrosion of pulverized coal boiler in power plant

Abstract

The application discloses a power station pulverized coal boiler high-temperature corrosion on-line real-time monitoring and control method, which comprises the following steps: collecting boiler operation core parameters, furnace combustion zone, near-wall zone flue gas and heating surface tube wall temperature data in real time; constructing a high-temperature corrosion prediction model based on the collected data, combining with boiler design parameters, coal quality characteristics and laboratory high-temperature corrosion test data; dynamically calculating and predicting the corrosion rate of each region of the heating surface and identifying high-risk regions; according to the prediction results, synchronously implementing combustion condition optimization and corrosion inhibitor addition; and simultaneously monitoring the heating surface tube wall state in real time. The measured data and the model prediction results are assimilated and fused through Kalman filtering algorithm to form a closed-loop control system. The application further discloses a power station pulverized coal boiler high-temperature corrosion on-line real-time monitoring and control system. The power station pulverized coal boiler high-temperature corrosion on-line real-time monitoring and control method and system provided by the application solve the problems of high-temperature corrosion prevention and control lag and passive regulation and control in the prior art.

Description

Technical Field

[0001] This invention belongs to the field of power plant boiler technology, specifically relating to a method for online real-time monitoring and control of high-temperature corrosion in power plant pulverized coal boilers. This invention also relates to a control system for online real-time monitoring and control of high-temperature corrosion in power plant pulverized coal boilers. Background Technology

[0002] Pulverized coal boilers in power plants are core equipment for power production. Their heating surface tube walls (water-cooled walls, superheaters, reheaters, and other core components) are exposed to high-temperature and highly corrosive flue gas environments for a long time. High-temperature corrosion has become a key pain point restricting the safe and stable operation of boilers. High-temperature corrosion is mainly manifested as thinning of the tube wall, local pitting corrosion, and even tube rupture. It not only causes unplanned boiler shutdowns and increases equipment maintenance costs, but also reduces boiler thermal efficiency and increases coal consumption. In severe cases, it can even lead to safety production accidents and have a great impact on the stable power supply of the power system. At present, the traditional treatment methods for high-temperature corrosion of pulverized coal boilers in power plants are mainly divided into three categories, but all of them have obvious limitations and cannot fundamentally solve the high-temperature corrosion problem: (1) By adjusting the air distribution method and controlling the excess air coefficient, the combustion conditions in the furnace are improved in an attempt to reduce the generation of SO3 and HCl. However, in actual operation, factors such as coal quality fluctuations, real-time changes in boiler load, and uneven flow field in the furnace are difficult to control precisely, making it impossible to ensure that the boiler is always in the best combustion state and to completely eliminate the causes of high-temperature corrosion. In addition, traditional combustion adjustment relies solely on macroscopic flue gas components and cannot detect the chemical activity of intermediate combustion products in the near-wall zone in real time, resulting in insufficient ability to eliminate the boundary layer of local reducing atmosphere. (2) Spray a corrosion-resistant coating on the surface of the heated tube wall or directly use corrosion-resistant alloy materials to manufacture the heated tube. Although this scheme can improve corrosion resistance to a certain extent, the corrosion-resistant coating is prone to peeling off under the action of high-temperature flue gas scouring and thermal stress, resulting in high maintenance costs in the later stage; the price of corrosion-resistant alloy materials is expensive, which greatly increases the manufacturing cost of the boiler and is not suitable for large-scale promotion and application. (3) Add magnesium-based, calcium-based and other corrosion inhibitors into the furnace, which can generate harmless substances by reacting with SO3 and HCl, thereby reducing their corrosion on the tube wall. However, the amount and location of the inhibitor addition are affected by various factors such as the flue gas flow field, temperature distribution and coal quality characteristics in the furnace. The existing addition method lacks precision, and the inhibitor particles are mostly dispersed with the mainstream of flue gas, making it difficult to effectively accumulate in the near-wall corrosion area. The inhibitor utilization efficiency is low, the anti-corrosion effect is limited, and it is still impossible to completely eliminate high-temperature corrosion. Summary of the Invention

[0003] The first objective of this invention is to provide an online real-time monitoring and control method for high-temperature corrosion in pulverized coal boilers of power plants, thereby solving the problems of lagging high-temperature corrosion prevention and control and passive regulation in the existing technology.

[0004] The second objective of this invention is to provide an online real-time monitoring and control system for high-temperature corrosion in pulverized coal boilers of power plants.

[0005] The first technical solution adopted in this invention is an online real-time monitoring and control method for high-temperature corrosion of pulverized coal boilers in power plants. This method includes real-time acquisition of core boiler operating parameters, flue gas temperature in the combustion zone and near-wall zone, and tube wall temperature data of the heated surface. Based on the acquired data, combined with boiler design parameters, coal quality characteristics, and laboratory high-temperature corrosion test data, a high-temperature corrosion prediction model is constructed. The corrosion rate of each area of ​​the heated surface is dynamically calculated and predicted, and high-risk areas are identified. Combustion condition optimization and corrosion inhibitor addition are implemented simultaneously based on the prediction results. At the same time, the tube wall condition of the heated surface is monitored in real time. The measured data and model prediction results are assimilated and fused using a Kalman filter algorithm to form a closed-loop control system.

[0006] The first technical solution of this invention is also characterized in that, Specifically, the following steps are included: Step 1: Using sensors and detection equipment, collect the core operating parameters of the boiler in real time at various key monitoring points of the boiler. The core parameters include flue gas temperature, flue gas composition, heating surface tube wall temperature, boiler load, and excess air coefficient. Also, collect the OH* and CH* chemiluminescence radiation intensity of the flame in the combustion zone of the furnace and the online data of HCl concentration in the flue gas near the wall to characterize the local combustion equivalence ratio and corrosive atmosphere activity. Step 2: Based on the collected real-time operating data, combined with boiler design parameters, coal quality characteristics data, and environmental high-temperature corrosion test data, a high-temperature corrosion prediction model is established. The high-temperature corrosion prediction model dynamically calculates and predicts the high-temperature corrosion rate of the pipe wall in different areas of the heating surface according to the real-time monitoring parameters, and identifies high-risk corrosion areas. The high-temperature corrosion prediction model is a hybrid model that integrates physical mechanism-driven and data-driven approaches. The input features of the data-driven sub-model include the standard deviation of pipe wall temperature fluctuation and the pipe wall temperature change rate extracted from the time series data of the heating surface pipe wall temperature, reflecting the accelerating effect of thermal fatigue on corrosion. Step 3: Based on the corrosion rate and high-risk area information output by the high-temperature corrosion prediction model, send control commands to the boiler control system in real time to dynamically adjust the boiler's air distribution mode, air volume distribution in each air chamber, and excess air coefficient, optimize the combustion conditions in the furnace, and reduce the generation of corrosive media; the control commands also include online adjustment of the intensity of secondary air swirl in each layer, and eliminate the local reducing atmosphere boundary layer by changing the turbulent mixing intensity in the near-wall area. Step 4: Based on the prediction results of the corrosion prediction model, combined with the characteristics of flue gas flow field and temperature distribution in the furnace, control the amount, location and rate of corrosion inhibitor addition to ensure that the inhibitor accurately reaches the high-risk corrosion area; during the addition process, the corrosion inhibitor powder is subjected to triboelectric charging to make the particles carry unipolar charge, and the water-cooled wall tube bank is used as a grounding electrode to form a DC electric field in the furnace to guide the charged inhibitor particles to migrate and deposit in a directional manner towards the high-risk corrosion area near the wall. Step 5: Use non-contact online detection equipment to monitor the thickness change and local pitting corrosion of the heated pipe wall in real time, and cross-validate the pipe wall status data with the corrosion prediction model results; when the pipe wall thickness change exceeds the threshold or abnormal corrosion occurs, issue an early warning signal and link up to strengthen prevention and control measures; cross-validation includes using the Kalman filter algorithm to assimilate the predicted corrosion rate output by the corrosion prediction model with the ultrasonic thickness measurement data to obtain the fused corrosion status estimate.

[0007] In step 1, the flue gas temperature measurement range is 0–1200℃; the flue gas composition includes CO, H2S, O2, SO2, and HCl; the heating surface tube wall temperature measurement range is 0–800℃; the boiler load is 0–120% of the rated load; key monitoring points include the furnace and the near-wall area of ​​the heating surface; the OH* and CH* chemiluminescence radiation intensity is collected by a photomultiplier tube array with a filter, and the sampling frequency is not less than 10Hz.

[0008] Step 2 includes boiler design parameters such as heating surface material, structural dimensions, and internal flow field characteristics; coal quality data includes industrial analysis and elemental analysis data of the coal.

[0009] In step 3, the corrosive media include SO3 and HCl; optimizing the combustion conditions in the furnace includes eliminating local reducing atmosphere; the intensity of the secondary air swirl is adjusted online by an adjustable blade cyclone separator, with the swirl number adjustable in the range of 0.4 to 1.2.

[0010] In step 4, the amount, location, and rate of corrosion inhibitor addition are dynamically adjusted based on the real-time prediction results of high-risk corrosion areas to ensure that the inhibitor reacts fully with the corrosive medium. The triboelectric electrostatic charging uses a tubular triboelectric electrostatic charger to achieve a charge-to-mass ratio of 0.5–1.5 μC / g for the inhibitor particles.

[0011] In step 5, the non-contact online detection equipment is an ultrasonic thickness gauge; the early warning signals include audible and visual warnings and system pop-up warnings; enhanced prevention and control measures include increasing the amount of inhibitor added and optimizing local ventilation.

[0012] The second technical solution adopted in this invention is an online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants, using the aforementioned monitoring and control method, including: The data acquisition module is used to collect the core parameters of the boiler during operation in real time and transmit the data to the data processing module. The data processing module cleans, filters, analyzes, and models the collected data to achieve dynamic prediction of high-temperature corrosion rate and identification of high-risk corrosion areas, and outputs the prediction results. The data processing module is equipped with a physical-data hybrid prediction model that integrates pipe wall temperature fluctuation characteristics, and includes a Kalman filter-based algorithm for assimilating prediction and measured data. The control execution module executes control commands for adjusting combustion conditions and adding corrosion inhibitors based on the prediction results from the data processing module. The online monitoring module monitors the physical state of the heated pipe wall in real time, verifies the corrosion prediction model results, and enables online early warning and anomaly handling for high-temperature corrosion.

[0013] The second technical solution of the present invention is further characterized in that, The data acquisition module includes a flue gas temperature sensor, a flue gas composition analyzer, a heated surface tube wall temperature sensor, a boiler load sensor, and an air volume sensor; the data acquisition module also includes an optical probe array for measuring the chemiluminescence radiation intensity of OH* and CH* in the local flame of the furnace, and an online HCl analyzer in the near-wall area; Flue gas temperature sensors and flue gas composition analyzers are installed at the furnace outlet and near the wall area of ​​each heating surface, respectively; heating surface tube wall temperature sensors are arranged at different monitoring sections of the heating surface tube walls of the water-cooled wall, superheater, and reheater; air volume sensors are installed in the air supply ducts of each air chamber; boiler load sensors are installed on the coal feeder body and generator outlet; the signal output terminals of the flue gas temperature sensor, flue gas composition analyzer, heating surface tube wall temperature sensor, boiler load sensor, air volume sensor, optical probe array, and near-wall area HCl online analyzer are all electrically connected to the signal input terminal of the data processing module. The control and execution module includes a combustion adjustment device and a corrosion inhibitor addition device; the combustion adjustment device includes an air distribution control valve, an air volume actuator, and a secondary air swirl intensity regulator; the air distribution control valve and the air volume actuator are installed in series on the air supply ducts of each air chamber; the secondary air swirl intensity regulator is an adjustable blade swirler, installed at the inlet of the secondary air nozzles on each floor. The corrosion inhibitor addition device includes an inhibitor storage tank, a metering pump, a triboelectric charger, and an injector, and uses water-cooled wall tubes as grounding electrodes to form an electric field guiding mechanism. The corrosion inhibitor storage tank and metering pump are arranged on the boiler furnace side platform. The inlet of the metering pump and the outlet of the storage tank are connected by a pipeline. The triboelectric static charger is a tubular triboelectric static charger, which is installed in series on the conveying pipeline between the outlet of the metering pump and the inlet of the precision injector. The nozzle of the precision injector is installed at the furnace wall opening corresponding to the high-risk corrosion area in the furnace, with the nozzle facing the near wall area inside the furnace. The grounding terminal of the water-cooled wall tube bank is electrically connected to the grounding terminal of the triboelectric static charger, forming the grounding electrode of the electric field guiding mechanism. The control output terminal of the data processing module is electrically connected to the control input terminals of the air distribution control valve, air volume actuator, secondary air swirl intensity regulator, metering pump, triboelectric static charger and injector respectively.

[0014] The online monitoring module includes an ultrasonic thickness gauge, a pipe wall corrosion imaging device, and an early warning terminal. The non-contact probe array of the ultrasonic thickness gauge is arranged at the corresponding monitoring positions on the outer walls of the water-cooled walls, superheaters, and reheaters. The pipe wall corrosion imaging device is installed at the furnace inspection port or viewing port, with its lens aimed at the inner wall of the heating surface. The early warning terminal is located on the control panel in the boiler control room. The signal output terminals of both the ultrasonic thickness gauge and the pipe wall corrosion imaging device are electrically connected to the feedback signal input terminal of the data processing module. The early warning signal output terminal of the data processing module is electrically connected to the signal input terminal of the early warning terminal. When the detected data exceeds the preset threshold, the early warning terminal issues an early warning. The linkage signal output terminal of the early warning terminal is electrically connected to the control input terminal of the data processing module, and the abnormal data is fed back to the data processing module. After being assimilated by Kalman filtering, the linkage control execution module executes enhanced prevention and control measures such as increasing the amount of inhibitor added and optimizing local air distribution.

[0015] The beneficial effects of this invention are: (1) The method and system for online real-time monitoring and control of high temperature corrosion of pulverized coal boiler in power plant provided by the present invention eliminates the main causes of high temperature corrosion from the source through closed-loop control of "real-time monitoring-precise prediction-control", effectively reduces the corrosion rate of the heating surface tube wall, avoids problems such as tube wall thinning and tube bursting, significantly extends the service life of boiler heating surface and related equipment, and reduces the frequency of equipment replacement and maintenance; in particular, by monitoring the chemiluminescence intensity of flame OH* and CH*, it can more directly reveal the change of equivalent ratio in the near wall area, make up for the lag of traditional macroscopic flue gas analysis, and combined with the characteristics of tube wall temperature fluctuation, enable the prediction model to capture the effect of thermal fatigue accelerated corrosion, significantly improving the sensitivity and predictability of corrosion prediction.

[0016] (2) The method and system for online real-time monitoring and control of high temperature corrosion of pulverized coal boiler in power plant provided by the present invention, based on the combustion condition optimization of corrosion prediction results, can not only reduce the generation of corrosive media, but also make the combustion in the furnace more complete, improve the thermal efficiency and combustion efficiency of the boiler, reduce coal consumption, achieve energy saving and consumption reduction, and improve the economic benefits of the power plant; after introducing the online adjustment of the intensity of secondary air swirl, it can effectively destroy the near-wall reducing atmosphere boundary layer, accurately control the wall atmosphere without significantly changing the total air volume, and reduce the disturbance of combustion adjustment to the main combustion zone.

[0017] (3) The method and system for online real-time monitoring and control of high temperature corrosion of pulverized coal boiler in power plant provided by the present invention can significantly improve the utilization efficiency of corrosion inhibitors by accurately adding corrosion inhibitors, reduce the amount of inhibitors used, and reduce the cost of reagents; at the same time, the online monitoring module can realize early detection and early treatment of corrosion problems, avoid large-scale maintenance and unplanned shutdowns caused by corrosion deterioration, and significantly reduce the operation and maintenance costs and downtime losses of boilers; through triboelectric electrostatic charging and electric field guidance, the inhibitor particles are directionally deposited on the high-risk corrosion pipe wall under the action of Coulomb force, which greatly improves the near-wall enrichment efficiency of inhibitors.

[0018] (4) The method and system for online real-time monitoring and control of high-temperature corrosion of pulverized coal boilers provided by the present invention transform passive corrosion prevention into active and precise corrosion prevention. Through real-time monitoring of multiple parameters and dynamic prediction of the model, high-risk corrosion areas are accurately identified to achieve targeted prevention and control. The online early warning function can detect corrosion anomalies in a timely manner, effectively avoid safety production accidents caused by high-temperature corrosion, and ensure the safe, stable and continuous operation of the boiler. Kalman filtering is used to realize the online assimilation of the prediction model and ultrasonic thickness measurement data, continuously correct the prediction deviation, and provide a more reliable corrosion status assessment for early warning and control. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the online real-time monitoring and control method for high-temperature corrosion in pulverized coal boilers of the present invention. Figure 2 This is a schematic diagram of the structure of the online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants according to the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0021] The present invention provides a method for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers for power plants, such as... Figure 1 As shown, this includes real-time acquisition of core boiler operating parameters, flue gas temperature in the furnace combustion zone and near-wall zone, and tube wall temperature data of the heated surfaces. Based on the acquired data, combined with boiler design parameters, coal quality characteristics, and laboratory high-temperature corrosion test data, a high-temperature corrosion prediction model is constructed. This model dynamically calculates and predicts the corrosion rate in each region of the heated surface and identifies high-risk areas. Based on the prediction results, combustion condition optimization and corrosion inhibitor addition are implemented simultaneously. Meanwhile, the tube wall condition of the heated surfaces is monitored in real time. The measured data and model prediction results are assimilated and fused using a Kalman filter algorithm to form a closed-loop control system. Specifically, the following steps are included: Step 1: Using sensors and detection equipment, real-time data are collected at key monitoring points of the boiler to obtain core operating parameters, including flue gas temperature, flue gas composition, heating surface tube wall temperature, boiler load, and excess air coefficient. Real-time data are also collected on the OH* and CH* chemiluminescence radiation intensity of the flame in the combustion zone of the furnace, as well as the online data on the HCl concentration in the flue gas near the wall, characterizing the local combustion equivalence ratio and corrosive atmosphere activity. In Step 1, the flue gas temperature measurement range is 0–1200℃; the flue gas composition includes CO, H2S, O2, SO2, and HCl; the heating surface tube wall temperature measurement range is 0–800℃; the boiler load is 0–120% of the rated load; key monitoring points include the furnace and the near-wall area of ​​the heating surface; the OH* and CH* chemiluminescence radiation intensity is collected using a photomultiplier tube array with filters, with a sampling frequency of not less than 10Hz.

[0022] Step 2: Based on the collected real-time operating data, combined with boiler design parameters, coal quality characteristic data, and environmental high-temperature corrosion test data, a high-temperature corrosion prediction model is established. The high-temperature corrosion prediction model dynamically calculates and predicts the high-temperature corrosion rate of the tube wall in different areas of the heating surface according to the real-time monitoring parameters, and identifies high-risk corrosion areas. The high-temperature corrosion prediction model is a hybrid model that integrates physical mechanism-driven and data-driven approaches. The input features of the data-driven sub-model include the standard deviation of tube wall temperature fluctuation and the tube wall temperature change rate extracted from the time series data of the heating surface tube wall temperature, reflecting the accelerating effect of thermal fatigue on corrosion. The boiler design parameters in Step 2 include the heating surface material, structural dimensions, and internal flow field characteristics. The coal quality characteristic data includes industrial analysis and elemental analysis data of the coal.

[0023] Step 3: Based on the corrosion rate and high-risk area information output by the high-temperature corrosion prediction model, control commands are sent to the boiler control system in real time to dynamically adjust the boiler's air distribution mode, air volume distribution in each air chamber, and excess air coefficient, thereby optimizing the combustion conditions in the furnace and reducing the generation of corrosive media. The control commands also include online adjustment of the secondary air swirl intensity in each layer, eliminating the local reducing atmosphere boundary layer by changing the turbulent mixing intensity in the near-wall area. In Step 3, the corrosive media include SO3 and HCl. Optimizing the combustion conditions in the furnace includes eliminating the local reducing atmosphere. The secondary air swirl intensity is adjusted online through an adjustable blade cyclone separator, with the swirl number adjustment range being 0.4 to 1.2.

[0024] Step 4: Based on the prediction results of the corrosion prediction model, and combined with the characteristics of the flue gas flow field and temperature distribution in the furnace, control the amount, location, and rate of corrosion inhibitor addition to ensure that the inhibitor accurately reaches the high-risk corrosion area. During the addition process, the corrosion inhibitor powder is subjected to triboelectric charging, causing the particles to carry a unipolar charge. A DC electric field is formed in the furnace using the water-cooled wall tubes as grounding electrodes to guide the charged inhibitor particles to migrate and deposit directionally towards the high-risk corrosion area near the wall. The amount, location, and rate of corrosion inhibitor addition in Step 4 are dynamically adjusted according to the real-time prediction results of the high-risk corrosion area to ensure that the inhibitor reacts fully with the corrosive medium. A tubular triboelectric charger is used for triboelectric charging to achieve a charge-to-mass ratio of 0.5–1.5 μC / g for the inhibitor particles.

[0025] Step 5: Utilize non-contact online monitoring equipment to monitor the thickness changes and localized pitting corrosion of the heated pipe wall in real time. Verify the pipe wall condition data with the corrosion prediction model results. When the pipe wall thickness change exceeds a threshold or abnormal corrosion occurs, issue an early warning signal and trigger enhanced prevention and control measures. Verification involves using a Kalman filter algorithm to assimilate the predicted corrosion rate output by the corrosion prediction model with the ultrasonic thickness measurement data to obtain a fused corrosion state estimate. The non-contact online monitoring equipment in Step 5 is an ultrasonic thickness gauge. Early warning signals include audible and visual warnings and system pop-up warnings. Enhanced prevention and control measures include increasing the amount of inhibitor added and optimizing local air distribution.

[0026] This invention also provides an online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants, such as... Figure 2 As shown, it includes a data acquisition module connected to a data processing module, which outputs prediction results. The data processing module contains a physical-data hybrid-driven prediction model that integrates pipe wall temperature fluctuation characteristics. It also includes a control execution module that executes control commands for adjusting combustion conditions and adding corrosion inhibitors based on the prediction results from the data processing module. Furthermore, it includes an online monitoring module that monitors the physical state of the heated pipe wall in real time, verifies the corrosion prediction model results, and enables online early warning and anomaly handling for high-temperature corrosion.

[0027] The data acquisition module includes a flue gas temperature sensor, a flue gas composition analyzer, a heated surface tube wall temperature sensor, a boiler load sensor, and an air volume sensor. It also includes an optical probe array for measuring the chemiluminescent radiation intensity of OH* and CH* in the local flame of the furnace, and an online HCl analyzer for the near-wall area. The flue gas temperature sensor and flue gas composition analyzer are installed at the furnace outlet and near the wall of each heated surface, respectively. The heated surface tube wall temperature sensors are arranged at different monitoring sections of the heated surface tube walls of the water-cooled wall, superheater, and reheater. The air volume sensor is installed in the air supply duct of each air chamber. The boiler load sensor is installed on the coal feeder body and the generator outlet. The signal output terminals of the flue gas temperature sensor, flue gas composition analyzer, heated surface tube wall temperature sensor, boiler load sensor, air volume sensor, optical probe array, and online HCl analyzer for the near-wall area are all electrically connected to the signal input terminal of the data processing module. The control execution module includes a combustion adjustment device and a corrosion inhibitor addition device. The combustion adjustment device includes an air distribution control valve, an air volume actuator, and a secondary air swirl intensity regulator. The system includes: air distribution control valves and air volume actuators connected in series on the air supply ducts of each air chamber; a secondary air swirl intensity regulator, an adjustable blade swirl converter, installed at the inlet of each secondary air nozzle; a corrosion inhibitor addition device comprising an inhibitor storage tank, a metering pump, a triboelectric charger, and an injector, utilizing water-cooled wall tubes as grounding electrodes to form an electric field guiding mechanism; the corrosion inhibitor storage tank and metering pump are located on the boiler furnace-side platform, with the metering pump inlet connected to the storage tank outlet via a pipeline; and a tubular triboelectric charger. The charger is connected in series on the conveying pipeline between the outlet of the metering pump and the inlet of the precision injector; the nozzle of the precision injector is installed at the furnace wall opening corresponding to the high-risk corrosion area of ​​the furnace, with the nozzle facing the near-wall area inside the furnace; the grounding terminal of the water-cooled wall tube bank is electrically connected to the grounding terminal of the triboelectric charger, forming the grounding electrode of the electric field guiding mechanism; the control output terminal of the data processing module is electrically connected to the control input terminals of the air distribution control valve, air volume actuator, secondary air swirl intensity regulator, metering pump, triboelectric charger and injector respectively.

[0028] The online monitoring module includes an ultrasonic thickness gauge, a pipe wall corrosion imaging device, and an early warning terminal. The non-contact probe array of the ultrasonic thickness gauge is positioned at corresponding monitoring locations on the outer walls of the water-cooled walls, superheaters, and reheaters. The pipe wall corrosion imaging device is installed at the furnace inspection port or viewing port, with its lens aimed at the inner wall of the heating surface. The early warning terminal is located on the control panel in the boiler control room. The signal output terminals of both the ultrasonic thickness gauge and the pipe wall corrosion imaging device are electrically connected to the feedback signal input terminal of the data processing module. The early warning signal output terminal of the data processing module is electrically connected to the signal input terminal of the early warning terminal. When the detected data exceeds a preset threshold, the early warning terminal issues an early warning. The linkage signal output terminal of the early warning terminal is electrically connected to the control input terminal of the data processing module, and the abnormal data is fed back to the data processing module. After Kalman filtering and assimilation, the linkage control execution module executes enhanced prevention and control measures, such as increasing the amount of inhibitor added and optimizing local air distribution.

[0029] The present invention provides a method and system for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers for power plants. This system uses a distributed sensor network to synchronously collect multi-dimensional data, including boiler operating parameters, furnace flame OH* / CH* chemiluminescence, concentrations of corrosive gases such as HCl in the near-wall region, and the condition of the heated tube walls. A hybrid prediction model integrating physical mechanisms and data-driven approaches is constructed to quantify the impact of conventional corrosion factors and incorporate tube wall temperature fluctuation characteristics to correct for thermal fatigue-coupled corrosion effects. The system dynamically calculates corrosion rates in each region and accurately identifies high-risk areas. Based on this, the system simultaneously optimizes combustion conditions (dynamically adjusting air distribution and secondary air swirl intensity to eliminate localized reducing atmospheres) and precisely adds electric field-guided corrosion inhibitors (guiding particle deposition through triboelectric charging and the grounding electric field of the water-cooled wall). Simultaneously, the system uses a Kalman filter algorithm to assimilate measured tube wall thickness data with model predictions, continuously correcting model errors and strengthening prevention and control measures, forming a complete online real-time monitoring and closed-loop control system for high-temperature corrosion.

[0030] Example 1 This embodiment provides a method for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers of power plants, including the following steps performed in sequence: Step 1: Real-time monitoring and acquisition of multiple parameters; High-precision sensors and detection equipment are used to collect core boiler operating parameters in real time at various key monitoring points. These parameters include flue gas temperature, flue gas composition, heated surface tube wall temperature, boiler load, and excess air coefficient, ensuring the real-time, accurate, and comprehensive nature of the monitoring data. Simultaneously, a photomultiplier tube array with filters is installed near the burner wall in the furnace to collect the chemiluminescence radiation intensity of OH* (wavelength 308 nm) and CH* (wavelength 431 nm) in real time at a sampling frequency of 20 Hz, used for online characterization of local equivalence ratios and the distribution of combustion intermediate products. Furthermore, a laser spectroscopy HCl analyzer is used separately in the near-wall region to acquire online HCl concentration data to assess the instantaneous activity of the corrosive atmosphere.

[0031] The flue gas temperature measurement range is 0–1200℃, with an accuracy of ±1℃; the flue gas composition includes CO, H2S, O2, SO2, and HCl, with a measurement accuracy of ±1% FS; the heating surface tube wall temperature measurement range is 0–800℃, with an accuracy of ±0.5℃; the boiler load is 0–120% of the rated load, with an accuracy of ±0.5%; the excess air coefficient is calculated, with an accuracy of ±0.05; key monitoring points include the furnace outlet, the near-wall surface of the burner area, the middle and lower parts of the water-cooled wall, the inlet of the screen-type superheater, the inlet of the high-temperature superheater, and the inlet of the reheater. At least one set of flue gas sampling and analysis equipment and temperature sensors is installed at each monitoring point. The sampling pump draws flue gas at a flow rate of 2–5 L / min, which is then filtered by a ceramic filter and dehydrated by a condenser before being sent to the flue gas composition analyzer. Non-dispersive infrared absorption is used to detect CO and SO2, electrochemical methods are used to detect H2S and O2, and laser spectroscopy is used to detect HCl. All sensors and detection equipment use high-temperature resistant and corrosion-resistant industrial-grade components and are equipped with water-cooled or air-cooled protective sleeves.

[0032] Step 2: Establish a high-temperature corrosion prediction model; Based on the real-time operating data collected in step 1, combined with boiler design parameters, coal quality characteristics data, and laboratory high-temperature corrosion test data, a high-precision high-temperature corrosion prediction model is established through big data analysis and algorithm optimization.

[0033] The specific construction process is as follows: (1) Data preparation: Collect at least 6 months of historical boiler operation data, heating surface material, structural dimensions, and flow field characteristics in the boiler design drawings; coal quality characteristic data include industrial analysis and elemental analysis of each batch of coal; laboratory high-temperature corrosion test data are obtained by on-site hanging plate method, standard test pieces are placed in different areas of the boiler, the material is the same as the heating surface, and after running for 100 to 500 hours, they are taken out to measure corrosion weight loss, and the correlation curve between corrosion rate and flue gas composition, temperature, and wall temperature is obtained.

[0034] (2) Model Architecture: A hybrid model is adopted, including a physical mechanism-based corrosion rate benchmark model and a machine learning-based error compensation model. The mechanistic model is based on the Arrhenius equation and the diffusion-reaction kinetics of SO3 and HCl to calculate the theoretical corrosion rate; the machine learning model takes real-time monitored parameters, including flue gas temperature, O2, CO, H2S, SO2, HCl, pipe wall temperature, load, and excess air coefficient, as inputs and outputs a correction coefficient for the corrosion rate. In addition to flue gas temperature, O2, CO, H2S, SO2, HCl, pipe wall temperature, load, and excess air coefficient, the input features of the machine learning model also include the standard deviation of pipe wall temperature fluctuation and the pipe wall temperature change rate extracted from the time series data of the heated surface pipe wall temperature, in order to quantify the accelerating effect of thermal fatigue on corrosion; at the same time, the chemiluminescence ratio of OH* and CH* is introduced as a substitute characterization quantity for the local equivalent ratio.

[0035] (3) Model training and updating: The model was trained using laboratory data and field plate data. Every 72 hours, the model parameters were fine-tuned online using the measured pipe wall thinning data from the most recent week. The final output of the model is: the instantaneous corrosion rate (unit: mm / 1000h) of each region of the heated surface, the cumulative corrosion thinning (unit: mm), and the coordinates and risk level of the high-risk corrosion areas, i.e., areas with a corrosion rate greater than 0.1 mm / 1000h. The risk levels include: Level I: 0.1~0.3 mm / 1000h; Level II: 0.3~0.5 mm / 1000h; Level III: >0.5 mm / 1000h.

[0036] Step 3: Optimize and adjust combustion conditions; Based on the corrosion rate and high-risk area information output by the high-temperature corrosion prediction model in step 2, control commands are sent to the boiler control system in real time. The control strategy adopts model predictive control with a control cycle of 30 seconds.

[0037] Adjust the air distribution method: For the identified local reducing atmosphere area, the opening of the secondary air damper of the corresponding burner is increased by 5% to 15%, while the concentration of pulverized coal in the adjacent primary air is reduced, so that the excess air coefficient in this area is increased to 1.05 to 1.15.

[0038] Airflow distribution in each air chamber: Adjust the airflow ratio of burnout air and secondary air in each layer according to the location of high-risk corrosion areas in the furnace height direction. If high-temperature corrosion is concentrated in the burner area, move some of the burnout air downward to increase the oxidizing atmosphere in the main combustion zone; if corrosion is concentrated in the screen-type superheater area, increase the burnout air rate to 25% to 30%.

[0039] Excess air coefficient optimization: While ensuring that the boiler thermal efficiency does not decrease, the excess air coefficient at the furnace outlet is dynamically adjusted from the conventional 1.15-1.20 to a range of 1.10-1.25. When the H2S concentration in the flue gas is detected to be >50 ppm or the CO concentration to be >1000 ppm, the excess air coefficient is automatically increased to above 1.20 and maintained for 5 minutes before returning to normal.

[0040] In addition to the aforementioned air distribution regulation, the control commands also include online adjustment of the swirl intensity of secondary air at each level. Adjustable blade cyclones installed in the secondary air ducts continuously adjust the swirl number within the range of 0.4 to 1.2. When the chemiluminescence ratio of OH* and CH* indicates the presence of a strongly reducing atmosphere boundary layer in a certain area, the secondary air swirl intensity in that area is increased to a swirl number of 0.9 to 1.1 to enhance near-wall turbulent mixing and rapidly disrupt the reducing atmosphere of the boundary layer, while maintaining a relatively constant airflow. All control commands have safety limits: the change in the opening of a single damper does not exceed 10% / min, and the change in total airflow does not exceed 5% / min, to avoid impacting combustion stability.

[0041] Step 4: Precisely add corrosion inhibitor; Based on the prediction results of the corrosion prediction model in step 2, and combined with the characteristics of the flue gas flow field and temperature distribution in the furnace, the amount, location, and rate of corrosion inhibitor addition are controlled to ensure that the inhibitor can accurately reach high-risk corrosion areas and fully react with the corrosive medium, significantly improving the utilization efficiency of the inhibitor and avoiding resource waste caused by blind addition. The inhibitor is selected from magnesium-based Mg(OH)2 or MgO, or calcium-based Ca(OH)2 or CaCO3 powder, with a particle size D90 ≤ 45 μm and a purity ≥ 95%.

[0042] The amount added is dynamically calculated based on the predicted concentrations of SO3 and HCl generated. Let the SO3 concentration be C. SO3 (ppm), HCl concentration is C HCl (ppm), theoretical inhibitor requirement Q (kg / h) = k1·C SO3 ·V flue + k2·C HCl ·V flue V flue Flue gas volumetric flow rate (Nm³) 3 / h), k1 and k2 are reaction coefficients (for magnesium-based, k1=0.032, k2=0.045; for calcium-based, k1=0.028, k2=0.038). The actual addition amount is multiplied by a safety factor of 1.1 to 1.3. When the corrosion high-risk area level is III, the upper limit of 1.3 is taken. The addition location is to pre-embed multiple injection points at different heights in the burner area, reduction zone, burnout zone, and horizontal flue inlet of the furnace, with each injection point corresponding to a specific coverage area. The control system selects the nearest 1 to 3 injection points for injection based on the location of the high-risk area. The injection points are pneumatically conveyed with a conveying air pressure of 0.4 to 0.6 MPa.

[0043] Before the inhibitor powder is ejected from the injection point, the particles are triboelectrically charged using a tubular triboelectric charger, giving the powder a unipolar charge with a charge-to-mass ratio controlled at 0.5–1.5 μC / g. Simultaneously, the water-cooled wall tubes are used as grounding electrodes, creating a DC electric field pointing towards the tube wall in the near-wall region of the furnace. After leaving the nozzle, the charged inhibitor powder, under the combined action of the flow field drag and the Coulomb force of the electric field, migrates along the flue gas flow, deflecting towards the water-cooled wall surface and depositing in high-risk corrosion areas, thus significantly increasing the inhibitor concentration and reaction efficiency near the wall. The addition rate uses pulsed injection, with each injection lasting 0.5–2 seconds, and the interval dynamically adjusted based on the corrosion rate change predicted by the model. For Level III high-risk areas, the injection interval is shortened to 10 seconds, with two injections per cycle. After injection, the changes in SO2 and HCl concentrations are monitored by a downstream flue gas composition analyzer. If the concentrations do not drop to the target values ​​of SO2 < 500 ppm and HCl < 30 ppm, the addition amount is automatically increased by 10% until the target is met.

[0044] Step 5: Online monitoring and early warning of the condition of the heated surface tube wall; Non-contact online inspection equipment is used to monitor the thickness changes and local pitting corrosion of the heated pipe wall in real time. Specific equipment and parameters are as follows: The ultrasonic thickness gauge uses a high-temperature electromagnetic ultrasonic sensor, requiring no coupling agent. The probe is temperature resistant to 500℃, with a measurement range of 1–50 mm and an accuracy of ±0.05 mm. A measuring point is placed every 2 meters on each heated pipe surface, focusing on monitoring bends and areas near welds. Measurements are taken automatically every 5 minutes, and data is wirelessly transmitted to the data processing module. The pipe wall corrosion imaging device uses a pulsed eddy current array probe, capable of scanning the wall thickness distribution within a 0.5m × 0.5m area with a resolution of 1mm × 1mm. High-risk areas are automatically scanned weekly.

[0045] The data processing module compares the measured pipe wall thickness with the initial thickness to calculate the actual corrosion rate. A Kalman filter algorithm is used to assimilate the predicted corrosion rate output by the high-temperature corrosion prediction model with the measured corrosion rate calculated by ultrasonic thickness measurement. The model prediction is used as a priori estimate of the state equation, and ultrasonic thickness measurement data is used as the observed value for updating. The resulting fused corrosion state estimate is then output and used to correct the internal state of the model. The warning terminal issues a warning signal if any of the following occurs: ① The fused corrosion state estimate exceeds 1.5 times the predicted value three consecutive times; ② The cumulative thinning at any measuring point exceeds 10% of the original wall thickness; ③ The local pitting depth exceeds 1.5 mm. The warning signal includes an audible and visual alarm, a pop-up window on the control room screen, and a text message sent to the on-duty engineer.

[0046] After the warning is triggered, the system automatically links to strengthen prevention and control measures: ① immediately increase the amount of inhibitor added in step 4 by 50% and continue spraying; ② increase the excess air coefficient of the corresponding area in step 3 by 0.05; ③ if the warning level is high, generate an operation and maintenance suggestion work order, suggesting that the pipe section in the area be replaced or sprayed with anti-corrosion coating during the next shutdown.

[0047] Example 2 This embodiment provides a system for implementing the method of Embodiment 1, including a data acquisition module, a data processing module, a control execution module, and an online monitoring module. These modules work collaboratively to achieve integrated operation of data acquisition, processing, prediction, regulation, and monitoring. The specific composition and workflow of each module are as follows: (a) Data Acquisition Module It is used to collect various core parameters during boiler operation in real time and accurately, and transmit the data to the data processing module without delay; The components and technical parameters are as follows:

[0048] The optical probe array consists of 12 photomultiplier tubes with 308 nm and 431 nm narrowband filters, installed near the walls of the burners at the four corners of the furnace, collecting the chemiluminescence intensity of OH* and CH* at a frequency of 20 Hz. All equipment is designed for high temperature and corrosion resistance; thermocouples and sensor probes are coated with high-temperature resistant ceramic coatings; and the flue gas sampling pipeline is heated to over 200°C using heating cables to prevent SO2 condensation. Data is transmitted in real-time via a PROFIBUS DP industrial bus with a refresh rate of 1 Hz.

[0049] (ii) Data Processing Module The data processing module is an industrial-grade control unit (IPC) equipped with an Intel Xeon processor, 32GB RAM, and a 1TB SSD, running a Linux real-time operating system and self-developed corrosion prediction software. Median filtering is applied to the raw data to remove outliers significantly exceeding the measurement range, and cubic spline interpolation is used to supplement missing data. A built-in hybrid model based on the Python TensorFlow framework is invoked, which integrates pipe wall temperature fluctuation characteristics and OH* and CH* data, outputting a full furnace corrosion rate distribution map every 30 seconds. Simultaneously, a Kalman filter assimilation algorithm is run to fuse ultrasonic thickness measurement data with model predictions online. All raw data and prediction results are stored in a time-series database (InfluxDB) for at least 5 years, supporting retrieval by time and region.

[0050] (III) Control Execution Module Based on the prediction results from the data processing module, control commands for adjusting combustion conditions and adding corrosion inhibitors are executed. The combustion adjustment device includes an electrically operated air distribution control valve and air volume actuator connected in parallel with the boiler's original air system, as well as adjustable blade cyclones installed in the secondary air ducts on each floor, with a cyclone number adjustment range of 0.4 to 1.2. The control system directly writes the damper opening setpoint to the DCS via a 4-20mA signal. For safety reasons, all control commands must undergo amplitude limiting checks by the DCS.

[0051] The corrosion inhibitor dosing device includes: a volume of 10m³ 3 The storage tank features a fluidized bottom and a vibratory arch breaker to prevent powder caking; a metering pump and screw feeder provide a feed rate of 0–500 kg / h with an accuracy of ±2%; a tubular friction electrostatic charger generates frictional charge by using high-speed airflow to carry powder and impact a special lining on the inner wall of the pipe, stabilizing the particle charge-to-mass ratio at 0.5–1.5 μC / g; and a precision injector employs pneumatic conveying, with each injection point equipped with an independent pneumatic switching valve and adjustable nozzle. The injection system and the water-cooled wall tube bank form an electric field loop, with the tube bank reliably grounded as a collecting electrode.

[0052] (iv) Online monitoring module This system monitors the physical state of the heated pipe wall in real time, verifies corrosion prediction model results, and enables online early warning and anomaly handling for high-temperature corrosion. The equipment includes an ultrasonic thickness gauge, a pipe wall corrosion imaging device, and an early warning terminal. The ultrasonic thickness gauge uses a high-temperature electromagnetic ultrasonic thickness gauge, automatically measuring and recording every 5 minutes. When the difference between two consecutive measurements at a certain measuring point exceeds 0.1 mm, the measurement frequency is automatically increased to once per minute. The pipe wall corrosion imaging device uses a pulsed eddy current array system with a scanning speed of 0.1 m / s. 2The system generates a pseudo-color map of wall thickness at a speed of [speed per minute], automatically marking points with corrosion depths exceeding 1 mm. The early warning terminal is an industrial control computer deployed in the control room, equipped with visualization software to display a real-time corrosion rate cloud map, historical trends, and alarm list. When early warning conditions are met, an audible and visual alarm and a full-screen pop-up window are triggered, and a maintenance work order containing the location, corrosion level, and recommended measures is automatically generated.

[0053] (V) Specific workflow of each module: ① Data acquisition stage: The sampling pump extracts the high-temperature flue gas near the furnace wall, and after pretreatment such as ash removal and dehydration, it is sent to the flue gas composition analyzer to accurately analyze the concentration of flue gas components such as CO, H2S, O2, and SO2; at the same time, flue gas temperature sensors, pipe wall temperature sensors, boiler load sensors and other sensors collect various parameters simultaneously, and optical probe arrays collect OH* and CH* chemiluminescence data. All monitoring data are transmitted to the data processing module in real time through the industrial communication bus.

[0054] ② Data processing and corrosion prediction stage: The data processing module cleans and filters the raw monitoring data to remove interference data; then the effective data is substituted into the high-temperature corrosion prediction model, combined with boiler design parameters and real-time coal quality data, and based on the high-temperature corrosion rate curve obtained from experimental research, the algorithm calculates and outputs the high-temperature corrosion rate of different areas of the heated surface in real time, and accurately locates high-risk corrosion areas.

[0055] ③ Control Phase: The data processing module transmits corrosion rate and high-risk area information to the control execution module. The combustion adjustment device dynamically adjusts the air distribution method, air volume of each air chamber, excess air coefficient, and secondary air swirl intensity according to the instructions, optimizes the combustion conditions in the furnace, eliminates local reducing atmosphere, and reduces the generation of SO3 and HCl. The corrosion inhibitor addition device precisely controls the amount of inhibitor added, the injection position, and the rate according to the instructions, so that the inhibitor can fully react with the corrosive medium in the high-risk corrosion area, and efficiently deposits the inhibitor in the target area through electrostatic charging and electric field guidance, thereby improving the anti-corrosion effect.

[0056] ④ Online monitoring and early warning stage: The ultrasonic thickness gauge of the online monitoring module detects the thickness change of the heated surface pipe wall in real time. The data processing module uses Kalman filtering to assimilate and fuse the detection data with the results of the corrosion prediction model. If the fused corrosion assessment exceeds the preset threshold or abnormal corrosion occurs, the early warning terminal immediately issues an early warning signal. At the same time, the data processing module links with the control execution module to strengthen the control measures. Maintenance personnel can take timely on-site treatment measures based on the early warning information to prevent the corrosion problem from worsening.

[0057] Implementation Notes: ① The sensors and analyzers of the data acquisition module need to be calibrated regularly to ensure the accuracy of the monitoring data; all detection equipment must be protected against high temperatures and corrosion to adapt to the harsh operating environment of the boiler. Optical probes must be equipped with purge protection gas to prevent lens contamination.

[0058] ② The high-temperature corrosion prediction model needs to be optimized and updated regularly based on the actual operation of the boiler to improve the prediction accuracy of the model.

[0059] ③ The selection of corrosion inhibitors should be tailored to the characteristics of the coal and the type of corrosive media in the furnace to ensure efficient reaction with media such as SO3 and HCl. The lining material of the triboelectric charger should be selected based on the work function characteristics of the inhibitor type to ensure stable charging.

[0060] ④ Communication between the various modules of the system must adopt an industrial-grade stable communication method to ensure real-time and delay-free data transmission and avoid affecting the control effect due to communication failure.

[0061] (vi) Overall system deployment Taking a 300MW subcritical tangential pulverized coal boiler as an example, the system deployment is as follows: A flue gas sampling and analysis device was installed at 0.5m, 1.5m, and 2.5m above the burners at the four corners of the furnace, totaling 12 measuring points. An optical probe array was installed near each measuring point. Twenty tubes were selected on the front and rear walls of the water-cooled wall, with two tube wall temperature measuring points and two ultrasonic thickness measuring probes installed on each tube. Six and four inhibitor injection points were installed in the burner area and the burnout air area, respectively. The data processing host communicated with the existing DCS via OPCUA to read signals such as coal feed rate, total air volume, and the opening degree of secondary air dampers on each floor.

[0062] Operational Procedure: At a certain moment, the data processing module predicts that the corrosion rate in the front wall burner area reaches 0.35 mm / 1000 h (Level II risk). Simultaneously, the flue gas analyzer in this area shows CO = 2500 ppm, H2S = 80 ppm, and the OH* / CH* ratio significantly deviates from the normal range. System Execution: Send a command to the DCS to increase the opening of the secondary air damper at the bottom of the front wall from 40% to 50%, correspondingly decrease the opening of the burnout damper by 5%, and increase the secondary air swirl number in this area from 0.6 to 1.0, while keeping the total air volume unchanged.

[0063] The nearest injection point to the area was activated, spraying electrostatically charged magnesium-based inhibitors at a rate of 150 kg / h in a pulsed pattern, with a 1-second on, 5-second off cycle, efficiently covering the near-wall area under the guidance of an electric field. After 10 minutes, CO in the area dropped to 800 ppm, H2S to 25 ppm, and the predicted corrosion rate dropped to 0.12 mm / 1000 h. The system automatically restored the injection interval to 30 seconds, and the damper remained open.

[0064] Meanwhile, the ultrasonic thickness gauge in the online monitoring module measured a cumulative thinning of 0.8 mm in this area. The corrosion state estimate after Kalman filtering exceeded the threshold of 0.6 mm, issuing a yellow warning and displaying a pop-up message in the control room: "The corrosion of the water-cooled wall in the front burner area is accelerating. It is recommended to shut down the furnace for inspection next time."

[0065] Example 3 This embodiment provides the online real-time monitoring and control method for high-temperature corrosion of pulverized coal boilers in power plants provided by the present invention, such as... Figure 1 As shown, this includes real-time acquisition of core boiler operating parameters, flue gas temperature in the furnace combustion zone and near-wall zone, and tube wall temperature data of the heated surfaces. Based on the acquired data, combined with boiler design parameters, coal quality characteristics, and laboratory high-temperature corrosion test data, a high-temperature corrosion prediction model is constructed. This model dynamically calculates and predicts the corrosion rate in each region of the heated surface and identifies high-risk areas. Based on the prediction results, combustion condition optimization and corrosion inhibitor addition are implemented simultaneously. Meanwhile, the tube wall condition of the heated surfaces is monitored in real time. The measured data and model prediction results are assimilated and fused using a Kalman filter algorithm to form a closed-loop control system. Specifically, the following steps are included: Step 1: Using sensors and detection equipment, collect the core operating parameters of the boiler in real time at various key monitoring points of the boiler. The core parameters include flue gas temperature, flue gas composition, heating surface tube wall temperature, boiler load, and excess air coefficient. Also, collect the OH* and CH* chemiluminescence radiation intensity of the flame in the combustion zone of the furnace and the online data of HCl concentration in the flue gas near the wall to characterize the local combustion equivalence ratio and corrosive atmosphere activity. Step 2: Based on the collected real-time operating data, combined with boiler design parameters, coal quality characteristics data, and environmental high-temperature corrosion test data, a high-temperature corrosion prediction model is established. The high-temperature corrosion prediction model dynamically calculates and predicts the high-temperature corrosion rate of the pipe wall in different areas of the heating surface according to the real-time monitoring parameters, and identifies high-risk corrosion areas. The high-temperature corrosion prediction model is a hybrid model that integrates physical mechanism-driven and data-driven approaches. The input features of the data-driven sub-model include the standard deviation of pipe wall temperature fluctuation and the pipe wall temperature change rate extracted from the time series data of the heating surface pipe wall temperature, reflecting the accelerating effect of thermal fatigue on corrosion. Step 3: Based on the corrosion rate and high-risk area information output by the high-temperature corrosion prediction model, send control commands to the boiler control system in real time to dynamically adjust the boiler's air distribution mode, air volume distribution in each air chamber, and excess air coefficient, optimize the combustion conditions in the furnace, and reduce the generation of corrosive media; the control commands also include online adjustment of the intensity of secondary air swirl in each layer, and eliminate the local reducing atmosphere boundary layer by changing the turbulent mixing intensity in the near-wall area. Step 4: Based on the prediction results of the corrosion prediction model, combined with the characteristics of flue gas flow field and temperature distribution in the furnace, control the amount, location and rate of corrosion inhibitor addition to ensure that the inhibitor accurately reaches the high-risk corrosion area; during the addition process, the corrosion inhibitor powder is subjected to triboelectric charging to make the particles carry unipolar charge, and the water-cooled wall tube bank is used as a grounding electrode to form a DC electric field in the furnace to guide the charged inhibitor particles to migrate and deposit in a directional manner towards the high-risk corrosion area near the wall. Step 5: Use non-contact online detection equipment to monitor the thickness change and local pitting corrosion of the heated pipe wall in real time, and cross-validate the pipe wall status data with the corrosion prediction model results; when the pipe wall thickness change exceeds the threshold or abnormal corrosion occurs, issue an early warning signal and link up to strengthen prevention and control measures; cross-validation includes using the Kalman filter algorithm to assimilate the predicted corrosion rate output by the corrosion prediction model with the ultrasonic thickness measurement data to obtain the fused corrosion status estimate.

[0066] This embodiment synchronously collects corrosion characteristic data such as basic boiler operating parameters, flame OH* / CH* chemiluminescence, and HCl concentration in the near-wall region through a distributed sensor network. It constructs a hybrid prediction model that integrates physical mechanisms and data-driven approaches. This model quantifies the impact of conventional corrosion factors and incorporates pipe wall temperature fluctuation characteristics to correct for the thermal fatigue coupled corrosion effect, accurately identifying high-risk corrosion areas. Simultaneously, it optimizes combustion conditions and precisely adds electric field-guided inhibitors to reduce the generation of corrosive media at the source and achieve targeted treatment. Finally, it uses Kalman filtering to assimilate the measured pipe wall thickness data with the model prediction results, forming a complete active prevention and control closed loop.

[0067] Example 4 Based on Example 3, in step 1, the flue gas temperature measurement range is 0–1200℃; the flue gas composition includes CO, H2S, O2, SO2, and HCl; the heating surface tube wall temperature measurement range is 0–800℃; the boiler load is 0–120% of the rated load; key monitoring points include the furnace and the near-wall area of ​​the heating surface; the chemiluminescence radiation intensity of OH* and CH* is collected by a photomultiplier tube array with filters, and the sampling frequency is not less than 10Hz. In step 2, the boiler design parameters include the heating surface material, structural dimensions, and in-furnace flow field characteristics; the coal quality characteristic data includes industrial analysis and elemental analysis data of the coal. In step 3, the corrosive media include SO3 and HCl; optimizing the in-furnace combustion conditions includes eliminating local reducing atmospheres; the secondary air swirl intensity is adjusted online by an adjustable blade cyclone separator, and the swirl number adjustment range is 0.4–1.2. In step 4, the amount, location, and rate of corrosion inhibitor addition are dynamically adjusted based on real-time predictions of high-risk corrosion areas to ensure sufficient reaction between the inhibitor and the corrosive medium. A tubular triboelectric charger is used for triboelectric charging to achieve a charge-to-mass ratio of 0.5–1.5 μC / g for the inhibitor particles. In step 5, the non-contact online detection device is an ultrasonic thickness gauge; early warning signals include audible and visual warnings and system pop-up warnings; enhanced control measures include increasing the inhibitor dosage and optimizing local ventilation.

[0068] This embodiment utilizes a high-precision sensor network deployed in the furnace and near the heating surface wall area to simultaneously collect core boiler operating parameters such as flue gas temperature (0–1200℃), flue gas composition (containing CO, H2S, O2, SO2, and HCl), heating surface tube wall temperature (0–800℃), and 0–120% rated load. Simultaneously, a photomultiplier tube array with filters is used to collect flame OH- at a sampling frequency of no less than 10Hz. and CH Chemiluminescence radiation intensity, combined with HCl concentration data in the near-wall region, accurately characterizes the local combustion state and corrosive atmosphere activity. Based on this, combined with boiler design parameters such as heating surface material, structural dimensions, and in-furnace flow field characteristics, as well as industrial and elemental analysis data of coal, a physical-data hybrid-driven prediction model is constructed. The characteristics of pipe wall temperature fluctuation are introduced to correct the thermal fatigue coupled corrosion effect, dynamically calculate the corrosion rate of each region, and identify high-risk areas.

[0069] Example 5 Based on Example 4, this example provides an online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants. The data acquisition module is used to collect core parameters during boiler operation in real time and transmit the data to the data processing module. The data acquisition module also includes an optical probe array for measuring the chemiluminescence radiation intensity of OH* and CH* in the local flame of the furnace, and an online HCl analyzer in the near-wall area. The data processing module cleans, filters, analyzes, and models the collected data to achieve dynamic prediction of high-temperature corrosion rate and identification of high-risk corrosion areas, and outputs the prediction results. The data processing module is equipped with a physical-data hybrid prediction model that integrates pipe wall temperature fluctuation characteristics, and includes a Kalman filter-based algorithm for assimilating prediction and measured data. The control execution module executes control commands for adjusting combustion conditions and adding corrosion inhibitors based on the prediction results of the data processing module. The control execution module includes an online adjustment device for the intensity of secondary air swirl and an electrostatic addition device for electrostatic charging and electric field guidance of corrosion inhibitor powder. The online monitoring module monitors the physical state of the heated pipe wall in real time, verifies the corrosion prediction model results, and enables online early warning and anomaly handling for high-temperature corrosion.

[0070] This embodiment provides an online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants, corresponding to the method. The system consists of a data acquisition module, a data processing module, a control execution module, and an online monitoring module, forming a complete proactive prevention and control system. The data acquisition module is responsible for collecting core boiler operating parameters in real time. It also integrates an optical probe array for measuring the chemiluminescence radiation intensity of OH* and CH* in the local flame of the furnace, as well as an online HCl analyzer in the near-wall area, providing specific characteristic data for corrosion prediction. The data processing module is responsible for data preprocessing and analysis. After cleaning and filtering the collected data, it uses an internally deployed physical-data hybrid prediction model that integrates the characteristics of pipe wall temperature fluctuations to achieve dynamic calculation of high-temperature corrosion rate and accurate identification of high-risk areas. It also has a built-in Kalman filter-based algorithm for assimilating prediction and measured data to correct model errors. The control execution module receives the prediction results from the data processing module and executes control commands for adjusting combustion conditions and adding corrosion inhibitors. It is specially equipped with an online adjustment device for secondary air swirl intensity and an electrostatic addition device for electrostatic charging and electric field guidance of corrosion inhibitor powder to achieve precise prevention and control. The online monitoring module monitors the physical state of the heated pipe wall in real time, verifies the accuracy of the corrosion prediction model, and completes online early warning and anomaly handling for high-temperature corrosion.

[0071] Example 6 Based on Example 5, the data acquisition module includes a flue gas temperature sensor, a flue gas composition analyzer, a heated surface tube wall temperature sensor, a boiler load sensor, and an air volume sensor. The data acquisition module also includes an optical probe array for measuring the chemiluminescence radiation intensity of OH* and CH* in the local flame of the furnace, and an online HCl analyzer in the near-wall area. The flue gas temperature sensor and flue gas composition analyzer are respectively installed at the furnace outlet and in the near-wall area of ​​each heated surface. The heated surface tube wall temperature sensors are arranged at different monitoring sections of the heated surface tube walls of the water-cooled wall, superheater, and reheater. The air volume sensor is installed in the air supply duct of each air chamber. The boiler load sensor is installed... The signal output terminals of the coal feeder body and generator outlet, flue gas temperature sensor, flue gas composition analyzer, heated surface tube wall temperature sensor, boiler load sensor, air volume sensor, optical probe array, and near-wall HCl online analyzer are all electrically connected to the signal input terminal of the data processing module; the control execution module includes a combustion adjustment device and a corrosion inhibitor addition device; the combustion adjustment device includes an air distribution control valve, an air volume actuator, and a secondary air swirl intensity regulator; the air distribution control valve and air volume actuator are installed in series on the air supply ducts of each air chamber; the secondary air swirl intensity regulator is an adjustable blade swirler, installed at the inlet of the secondary air nozzles of each layer; The corrosion inhibitor addition device includes an inhibitor storage tank, a metering pump, a triboelectric charger, and an injector, and utilizes a water-cooled wall tube bank as a grounding electrode to form an electric field guiding mechanism. The corrosion inhibitor storage tank and metering pump are arranged on the boiler furnace side platform, and the inlet of the metering pump is connected to the outlet of the storage tank through a pipeline. The triboelectric charger is a tubular triboelectric charger, which is installed in series on the conveying pipeline between the outlet of the metering pump and the inlet of the precision injector. The nozzle of the precision injector is installed at the furnace wall opening corresponding to the high-risk corrosion area in the furnace, with the nozzle facing the near-wall area inside the furnace. The grounding terminal of the water-cooled wall tube bank is electrically connected to the grounding terminal of the triboelectric charger, forming the grounding electrode of the electric field guiding mechanism. The control output terminal of the data processing module is electrically connected to the control input terminals of the air distribution control valve, the air volume actuator, the secondary air swirl intensity regulator, the metering pump, the triboelectric charger, and the injector, respectively.

[0072] The online monitoring module includes an ultrasonic thickness gauge, a pipe wall corrosion imaging device, and an early warning terminal. The non-contact probe array of the ultrasonic thickness gauge is positioned at corresponding monitoring locations on the outer walls of the water-cooled walls, superheaters, and reheaters. The pipe wall corrosion imaging device is installed at the furnace inspection port or viewing port, with its lens aimed at the inner wall of the heating surface. The early warning terminal is located on the control panel in the boiler control room. The signal output terminals of both the ultrasonic thickness gauge and the pipe wall corrosion imaging device are electrically connected to the feedback signal input terminal of the data processing module. The early warning signal output terminal of the data processing module is electrically connected to the signal input terminal of the early warning terminal. When the detected data exceeds a preset threshold, the early warning terminal issues an early warning. The linkage signal output terminal of the early warning terminal is electrically connected to the control input terminal of the data processing module, and the abnormal data is fed back to the data processing module. After Kalman filtering and assimilation, the linkage control execution module executes enhanced prevention and control measures, such as increasing the amount of inhibitor added and optimizing local air distribution.

Claims

1. A method for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers of power plants, characterized in that, This includes real-time acquisition of core boiler operating parameters, flue gas temperature in the furnace combustion zone and near-wall zone, and tube wall temperature data of the heated surfaces. Based on the acquired data, combined with boiler design parameters, coal quality characteristics, and laboratory high-temperature corrosion test data, a high-temperature corrosion prediction model is constructed. The model dynamically calculates and predicts the corrosion rate of each area of ​​the heated surface and identifies high-risk areas. Based on the prediction results, combustion condition optimization and corrosion inhibitor addition are implemented simultaneously. At the same time, the tube wall condition of the heated surfaces is monitored in real time. The measured data and model prediction results are assimilated and fused using a Kalman filter algorithm to form a closed-loop control system.

2. The method for online real-time monitoring and control of high-temperature corrosion in power plant pulverized coal boilers according to claim 1, characterized in that, Specifically, the following steps are included: Step 1: Using sensors and detection equipment, collect the core operating parameters of the boiler in real time at various key monitoring points of the boiler. The core parameters include flue gas temperature, flue gas composition, heating surface tube wall temperature, boiler load, and excess air coefficient. The OH* and CH* chemiluminescence radiation intensity of the flame in the combustion zone of the furnace and the online data of HCl concentration in the flue gas near the wall were collected in real time to characterize the local combustion equivalence ratio and the activity of corrosive atmosphere. Step 2: Based on the collected real-time operating data, combined with boiler design parameters, coal quality characteristic data, and environmental high-temperature corrosion test data, a high-temperature corrosion prediction model is established. The high-temperature corrosion prediction model dynamically calculates and predicts the high-temperature corrosion rate of the pipe wall in different areas of the heating surface according to the real-time monitoring parameters, and identifies high-risk corrosion areas. The high-temperature corrosion prediction model is a hybrid model that integrates physical mechanism-driven and data-driven approaches. The input features of the data-driven sub-model include the standard deviation of pipe wall temperature fluctuation and the pipe wall temperature change rate extracted from the time series data of the heating surface pipe wall temperature, reflecting the accelerating effect of thermal fatigue on corrosion. Step 3: Based on the corrosion rate and high-risk area information output by the high-temperature corrosion prediction model, send control commands to the boiler control system in real time to dynamically adjust the boiler's air distribution mode, air volume distribution in each air chamber, and excess air coefficient, optimize the combustion conditions in the furnace, and reduce the generation of corrosive media; the control commands also include online adjustment of the intensity of secondary air swirl in each layer, and eliminate the local reducing atmosphere boundary layer by changing the turbulent mixing intensity in the near-wall area. Step 4: Based on the prediction results of the corrosion prediction model, combined with the characteristics of flue gas flow field and temperature distribution in the furnace, control the amount, location and rate of corrosion inhibitor addition so that the inhibitor can accurately reach the high-risk corrosion area. During the addition process, the corrosion inhibitor powder is subjected to triboelectric charging, which makes the particles carry unipolar charges. The water-cooled wall tube bank is used as a grounding electrode to form a DC electric field in the furnace, which guides the charged inhibitor particles to migrate and deposit in a directional manner towards the high-risk area of ​​near-wall corrosion. Step 5: Use non-contact online detection equipment to monitor the thickness change and local pitting corrosion of the heated pipe wall in real time, and verify the pipe wall status data with the corrosion prediction model results; when the pipe wall thickness change exceeds the threshold or abnormal corrosion occurs, issue an early warning signal and link up to strengthen prevention and control measures; the mutual verification includes using the Kalman filter algorithm to assimilate the predicted corrosion rate output by the corrosion prediction model with the ultrasonic thickness measurement data to obtain the fused corrosion status estimate.

3. The method for online real-time monitoring and control of high-temperature corrosion in power plant pulverized coal boilers according to claim 2, characterized in that, The flue gas temperature measurement range in step 1 is 0–1200℃; the flue gas composition includes CO, H2S, O2, SO2, and HCl; the heating surface tube wall temperature measurement range is 0–800℃; the boiler load is 0–120% of the rated load; the key monitoring points include the furnace and the near-wall area of ​​the heating surface; the OH* and CH* chemiluminescence radiation intensity is collected by a photomultiplier tube array with a filter, and the sampling frequency is not less than 10Hz.

4. The method for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers of power plants according to claim 2, characterized in that, The boiler design parameters mentioned in step 2 include the material of the heating surface, structural dimensions, and flow field characteristics inside the furnace; the coal quality characteristic data include industrial analysis and elemental analysis data of the coal.

5. The method for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers of power plants according to claim 2, characterized in that, In step 3, the corrosive medium includes SO3 and HCl; the optimization of the combustion conditions in the furnace includes eliminating local reducing atmosphere; the intensity of the secondary air swirl is adjusted online by an adjustable blade cyclone separator, and the swirl number adjustment range is 0.4 to 1.

2.

6. The method for online real-time monitoring and control of high-temperature corrosion in power plant pulverized coal boilers according to claim 2, characterized in that, In step 4, the amount, location, and rate of corrosion inhibitor addition are dynamically adjusted based on the real-time prediction results of high-risk corrosion areas to ensure that the inhibitor reacts fully with the corrosive medium. The triboelectric electrostatic charging uses a tubular triboelectric electrostatic charger to achieve a charge-to-mass ratio of 0.5–1.5 μC / g for the inhibitor particles.

7. The method for online real-time monitoring and control of high-temperature corrosion in pulverized coal boilers of power plants according to claim 2, characterized in that, The non-contact online detection device mentioned in step 5 is an ultrasonic thickness gauge; the early warning signals include audible and visual early warnings and system pop-up warnings; the enhanced prevention and control measures include increasing the amount of inhibitor added and optimizing local air distribution.

8. A real-time online monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants, characterized in that, The monitoring and control method according to any one of claims 1-7 includes a data acquisition module connected to a data processing module, which outputs prediction results. The data processing module contains a physical-data hybrid-driven prediction model that integrates pipe wall temperature fluctuation characteristics. It also includes a control execution module that executes control commands for adjusting combustion conditions and adding corrosion inhibitors based on the prediction results from the data processing module. Furthermore, it includes an online monitoring module that monitors the physical state of the heated pipe wall in real time, verifies the corrosion prediction model results, and enables online early warning and anomaly handling for high-temperature corrosion.

9. The online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants according to claim 8, characterized in that, The data acquisition module includes a flue gas temperature sensor, a flue gas composition analyzer, a heated surface tube wall temperature sensor, a boiler load sensor, and an air volume sensor. The data acquisition module includes an optical probe array for measuring the chemiluminescence radiation intensity of OH* and CH* in the local flame of the furnace, and an online HCl analyzer in the near-wall region; The flue gas temperature sensor and flue gas composition analyzer are respectively installed at the furnace outlet and near the wall area of ​​each heating surface; the heating surface tube wall temperature sensor is arranged at different monitoring sections of the water-cooled wall, superheater, and reheater heating surface tube wall; the air volume sensor is installed in the air supply duct of each air chamber; the boiler load sensor is installed at the coal feeder body and generator outlet; the signal output terminals of the flue gas temperature sensor, flue gas composition analyzer, heating surface tube wall temperature sensor, boiler load sensor, air volume sensor, optical probe array, and near-wall area HCl online analyzer are all electrically connected to the signal input terminal of the data processing module; The control execution module includes a combustion adjustment device and a corrosion inhibitor addition device; the combustion adjustment device includes an air distribution control valve, an air volume actuator, and a secondary air swirl intensity regulator; the air distribution control valve and the air volume actuator are connected in series on the air supply duct of each air chamber; the secondary air swirl intensity regulator is an adjustable blade swirler, installed at the inlet of the secondary air nozzle of each layer. The corrosion inhibitor addition device includes an inhibitor storage tank, a metering pump, a triboelectric charger, and an injector, and uses water-cooled wall tubes as grounding electrodes to form an electric field guiding mechanism. The corrosion inhibitor storage tank and metering pump are arranged on the boiler furnace side platform, and the inlet of the metering pump is connected to the outlet of the storage tank through a pipeline; the triboelectric charger is a tubular triboelectric charger, which is installed in series on the conveying pipeline between the outlet of the metering pump and the inlet of the precision injector; the nozzle of the precision injector is installed at the furnace wall opening corresponding to the high-risk corrosion area in the furnace, and the nozzle faces the near wall area inside the furnace; the grounding terminal of the water-cooled wall tube is electrically connected to the grounding terminal of the triboelectric charger, forming the grounding electrode of the electric field guiding mechanism; the control output terminal of the data processing module is electrically connected to the control input terminals of the air distribution control valve, the air volume actuator, the secondary air swirl intensity regulator, the metering pump, the triboelectric charger, and the injector.

10. The online real-time monitoring and control system for high-temperature corrosion of pulverized coal boilers in power plants according to claim 8, characterized in that, The online monitoring module includes an ultrasonic thickness gauge, a pipe wall corrosion imaging device, and an early warning terminal. The non-contact probe array of the ultrasonic thickness gauge is arranged at corresponding monitoring positions on the outer walls of the water-cooled walls, superheaters, and reheaters. The pipe wall corrosion imaging device is installed at the furnace inspection port or observation port, with its lens aimed at the inner wall of the heating surface. The early warning terminal is located on the control panel in the boiler control room. The signal output terminals of both the ultrasonic thickness gauge and the pipe wall corrosion imaging device are electrically connected to the feedback signal input terminal of the data processing module. The early warning signal output terminal of the data processing module is electrically connected to the signal input terminal of the early warning terminal. When the detected data exceeds the preset threshold, the early warning terminal issues an early warning. The linkage signal output terminal of the early warning terminal is electrically connected to the control input terminal of the data processing module, and the abnormal data is fed back to the data processing module. After being assimilated by Kalman filtering, the linkage control execution module executes enhanced prevention and control measures such as increasing the amount of inhibitor added and optimizing local air distribution.

Images