Thermal power plant boiler combustion optimization adjustment system and method

Abstract

This invention discloses a combustion optimization and adjustment system and method for thermal power plant boilers. The combustion optimization and adjustment system includes a data acquisition module, an intelligent analysis and command module, an actuator, and a feedback adjustment module. The data acquisition module includes an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter. The intelligent analysis and command module includes a combustion digital twin model constructed using a deep learning algorithm. The intelligent analysis and command module generates combined commands based on the data acquired by the data acquisition module. The actuator controls the coal feed rate and speed of the variable frequency coal feeder, the damper opening degree of the intelligent damper actuator, and / or the blowing angle according to the combined commands. The feedback adjustment module dynamically corrects the combustion digital twin model. According to the embodiments of this invention, the thermal power plant boiler combustion optimization and adjustment system, combined with reinforcement learning to generate the optimal control strategy, ensures the stability and efficiency of the combustion process and forms an effective closed-loop optimization mechanism.

Description

Technical Field

[0001] This invention relates to the field of thermal power plants, and in particular to a combustion optimization and adjustment system and method for thermal power plant boilers. Background Technology

[0002] In the operation of thermal power plants, boiler combustion efficiency and stability are key factors affecting the overall economic efficiency and safety of the plant. As the core equipment of a thermal power plant, the boiler's combustion status directly determines fuel utilization, power generation costs, and pollutant emission levels. Currently, with the continuous growth of energy demand and increasingly stringent environmental standards, optimizing the boiler combustion process in thermal power plants to achieve efficient and clean combustion has become a significant challenge for these plants.

[0003] Currently, combustion adjustments in some traditional thermal power plant boilers rely heavily on the experience of operators. Operators manually adjust key parameters such as coal feed and air supply based on boiler load changes, fuel characteristics, and instrument parameters like main steam pressure, temperature, and furnace negative pressure. However, this experience-based adjustment method has several limitations. Firstly, due to varying levels of experience among operators, their judgments and adjustment strategies for the same operating conditions can differ significantly, making it difficult to guarantee the accuracy and consistency of combustion adjustments. Even experienced operators struggle to quickly and accurately find the optimal combination of combustion parameters when faced with complex and changing operating conditions, thus affecting boiler combustion efficiency.

[0004] Therefore, it is necessary to invent a combustion optimization and adjustment system for thermal power plant boilers to solve the above problems. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, one objective of this invention is to propose a combustion optimization and adjustment system for thermal power plant boilers. By enabling a multi-modal sensor array in the data acquisition module to achieve three-dimensional synchronous measurement of the temperature field, concentration field, and flow field within the furnace, accurate and complete data support is provided to the intelligent analysis and command module. This refined data acquisition method significantly improves the accuracy of the combustion digital twin model, thereby enhancing the accuracy of combustion state assessment and contributing to a more efficient combustion process. By employing deep learning algorithms to construct the combustion digital twin model in the intelligent analysis and command module, combined with reinforcement learning to generate the optimal control strategy, the stability and efficiency of the combustion process are ensured. The feedback adjustment module dynamically corrects the control strategy by comparing the model's predicted values ​​with actual operating data in real time, forming an effective closed-loop optimization mechanism. This mechanism can promptly detect and correct model deviations, ensuring the continuity and stability of combustion optimization and adjustment.

[0006] The present invention also proposes a method for optimizing and adjusting the combustion of a thermal power plant boiler using the above-mentioned thermal power plant boiler combustion optimization and adjustment system.

[0007] According to a first aspect of the present invention, a combustion optimization and adjustment system for a thermal power plant boiler includes: a data acquisition module comprising an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter, wherein the infrared thermal imager includes an infrared sensor array and is used to detect the temperature field inside the furnace, the laser gas analyzer includes a laser sensor array and is used to detect the gas concentration field inside the furnace, and the ultrasonic flow meter includes an ultrasonic sensor array and is used to detect the flow field inside the furnace; an intelligent analysis and instruction module comprising a combustion digital twin model constructed by a deep learning algorithm, wherein the intelligent analysis and instruction module is used to generate combined instructions based on the data acquired by the data acquisition module; an actuator comprising a variable frequency coal feeder and an intelligent damper actuator, used to control the coal feed rate and coal feed speed of the variable frequency coal feeder, and the damper opening and / or blowing angle of the intelligent damper actuator according to the combined instructions; and a feedback adjustment module for comparing real-time data with the calculation results of the intelligent analysis module and dynamically correcting the combustion digital twin model.

[0008] According to an embodiment of the present invention, the combustion optimization and adjustment system for thermal power plant boilers enables the multi-modal sensor array of the data acquisition module to achieve three-dimensional synchronous measurement of the temperature field, concentration field, and flow field within the furnace. This provides accurate and complete data support for the intelligent analysis and command module. This refined data acquisition method significantly improves the accuracy of the combustion digital twin model, thereby enhancing the accuracy of combustion state assessment and contributing to a more efficient combustion process. By employing deep learning algorithms to construct the combustion digital twin model in the intelligent analysis and command module, and combining this with reinforcement learning to generate the optimal control strategy, the stability and efficiency of the combustion process are ensured. The feedback adjustment module dynamically corrects the control strategy by comparing the model's predicted values ​​with actual operating data in real time, forming an effective closed-loop optimization mechanism. This mechanism can promptly detect and correct model deviations, ensuring the continuity and stability of combustion optimization and adjustment.

[0009] According to some embodiments of the present invention, the data acquisition module further includes a data fusion algorithm, which is used to fuse the data acquired by the infrared thermal imager, the laser gas analyzer and the ultrasonic flow meter to remove noise and outliers.

[0010] According to some embodiments of the present invention, the intelligent analysis and instruction module includes an intelligent analysis module and a control instruction generation module. The intelligent analysis module includes a multi-parameter fusion unit and the combustion digital twin model. The multi-parameter fusion unit is used to perform feature extraction and correlation analysis on the data collected by the data acquisition module and transmit it to the combustion digital twin model. The combustion digital twin model is used to generate a control strategy. The control instruction generation module is used to generate the combined instruction according to the control strategy.

[0011] According to some embodiments of the present invention, the control command generation module includes a multi-objective optimization algorithm, which is used to generate combined commands under constraints based on the calculation results of the intelligent analysis module; the constraints include: boiler load, combustion efficiency, NO... x Emissions and coking risks; the combined commands include: the coal feeding speed and coal feeding amount of the variable frequency coal feeder, and the damper opening and blowing angle of the intelligent damper actuator.

[0012] According to some embodiments of the present invention, the actuator includes an adaptive adjustment device, which includes a PID controller and a fuzzy control algorithm. The adaptive adjustment device is used to compare the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator with the combined command, and adjust the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator to match the combined command. The response time of the adaptive adjustment device is ≤2 seconds.

[0013] According to a second aspect of the present invention, a method for optimizing and adjusting the combustion of a thermal power plant boiler is used, which utilizes the combustion optimization and adjustment system for a thermal power plant boiler described in a first aspect of the present invention. The method includes: a data acquisition module performing three-dimensional measurements of the temperature field, concentration field, and flow field within the furnace using an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter to obtain measurement data; an intelligent analysis and instruction module constructing a combustion digital twin model based on a deep learning algorithm, and performing feature extraction and correlation analysis on the parameters acquired by the data acquisition module using principal component analysis and neural network methods to generate combined instructions; an actuator changing the coal feed rate and coal feed speed of the variable frequency coal feeder, and the opening degree and / or angle of the intelligent damper actuator according to the combined instructions; and a feedback adjustment module dynamically correcting the combustion digital twin model by comparing the predicted values ​​of the combustion digital twin model with the actual operating data acquired by the data acquisition module in real time.

[0014] According to the combustion optimization and adjustment method for thermal power plant boilers of the present invention, the data acquisition module performs three-dimensional synchronous measurement of the temperature field, concentration field, and flow field in the furnace, providing accurate and complete data support for the intelligent analysis and command module. This refined data acquisition method significantly improves the accuracy of the combustion digital twin model, thereby enhancing the accuracy of combustion state assessment and contributing to a more efficient combustion process. By employing deep learning algorithms to construct the combustion digital twin model and combining reinforcement learning to generate the optimal control strategy, the stability and efficiency of the combustion process are ensured. The feedback adjustment module dynamically corrects the combustion digital twin model by comparing the model's predicted values ​​with actual operating data in real time, forming an effective closed-loop optimization mechanism. This mechanism can promptly detect and correct model deviations, ensuring the continuity and stability of combustion optimization and adjustment.

[0015] According to some embodiments of the present invention, the feedback adjustment module, in the process of real-time comparison between model prediction values ​​and actual operating data, quickly adjusts the control strategy based on deviations in key parameters such as temperature and pressure using a model predictive control algorithm.

[0016] According to some embodiments of the present invention, the combustion optimization and adjustment system for a thermal power plant boiler is a combustion optimization and adjustment system for a thermal power plant boiler according to the first aspect of the present invention. The intelligent analysis module adopts a combination of principal component analysis and neural network method, and uses the combustion digital twin model to generate control strategies; the control command generation module is based on a multi-objective optimization algorithm, considering combustion efficiency, NO... x Under constraints such as emissions and coking risks, the optimal combination of coal feed rate, damper opening, and burner angle is generated according to the control strategy.

[0017] According to some embodiments of the present invention, during the process of the intelligent analysis module generating the optimal control strategy and the control command generation module generating the optimal combination command, while optimizing combustion efficiency, pollution emission factors are comprehensively considered to control NO in the boiler of thermal power plants. x Emissions ≤30mg / Nm³, SO2 emissions ≤20mg / Nm³.

[0018] According to some embodiments of the present invention, the intelligent analysis and instruction module adopts a dynamic adaptive algorithm, predicts the combustion trend through an LSTM network, adjusts the control parameters, and adapts to the changes in the maximum continuous output load of the boiler from 30% to 120%.

[0019] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a method for optimizing and adjusting the combustion of a thermal power plant boiler according to some embodiments of the present invention. Detailed Implementation

[0021] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0022] The following is for reference. Figure 1 A combustion optimization and adjustment system for a thermal power plant boiler is described according to an embodiment of the present invention.

[0023] According to a first aspect of the present invention, a boiler combustion optimization and adjustment system for thermal power plants includes: a data acquisition module, an intelligent analysis and instruction module, an actuator, and a feedback adjustment module.

[0024] The data acquisition module includes an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter. The infrared thermal imager includes an infrared sensor array and is used to detect the temperature field inside the furnace. The laser gas analyzer includes a laser sensor array and is used to detect the gas concentration field inside the furnace. The ultrasonic flow meter includes an ultrasonic sensor array and is used to detect the flow field inside the furnace. The intelligent analysis and command module includes a combustion digital twin model constructed using deep learning algorithms. This module generates combined commands based on the data acquired by the data acquisition module. The actuators include a variable frequency coal feeder and an intelligent damper actuator, used to control the coal feed rate and speed of the variable frequency coal feeder, and the damper opening and / or blowing angle of the intelligent damper actuator according to the combined commands. The feedback adjustment module compares real-time data with the calculation results of the intelligent analysis module to dynamically correct the combustion digital twin model.

[0025] The data acquisition module integrates an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter, acquiring various types of data in a multimodal manner. The infrared thermal imager measures the temperature inside the furnace, with an infrared sensor array distributed in three-dimensional space to detect the temperature at different locations within the furnace, forming a three-dimensional temperature field. The laser gas analyzer emits a specific laser beam through the flue gas, and by analyzing the absorption of the laser, it detects oxygen (O2), carbon monoxide (CO), and nitrogen oxides (NOx). xThe furnace employs a multi-position laser sensor array to detect the concentrations of key gases such as sulfur dioxide (SO2) at different locations, forming a three-dimensional concentration field. Ultrasonic flow meters measure wind speed and direction by utilizing the speed difference of ultrasound propagation in fluids, providing insight into gas flow patterns. An ultrasonic sensor array is distributed throughout the three-dimensional space of the furnace to detect gas flow at different locations, creating a three-dimensional flow field. This combination of data provides a more comprehensive and accurate understanding of the conditions within the furnace.

[0026] The data acquisition module provides accurate and complete data support for the intelligent analysis and command module. This refined data acquisition method significantly improves the accuracy of the combustion digital twin model, thereby enhancing the accuracy of combustion state assessment and helping to achieve a more efficient combustion process.

[0027] By incorporating a combustion digital twin model built using deep learning algorithms into the intelligent analysis and instruction module, and using this model to input various data collected by the data acquisition module, the system calculates the strategy with the highest boiler combustion efficiency based on matching the power demand of the thermal power plant and meeting power production requirements. Based on this strategy, a combination of instructions for controlling the operation of the actuators is generated to improve fuel combustion efficiency.

[0028] The actuators include a variable frequency coal feeder and an intelligent damper actuator. For example, the actuators can control the coal feed rate and speed of the variable frequency coal feeder, and the damper opening and blowing angle of the intelligent damper actuator according to combined commands. For example, the actuators can support adjustment with an accuracy of ±0.1%.

[0029] By including a feedback adjustment module in the system, the calculation results of the combustion digital twin model are compared with real-time data, and the combustion digital twin model is dynamically corrected and optimized in reverse, making it increasingly accurate. The model continuously evolves based on real-time feedback. An effective closed-loop optimization mechanism can promptly detect and correct model deviations, ensuring the continuity and stability of combustion optimization adjustments.

[0030] According to an embodiment of the present invention, the combustion optimization and adjustment system for thermal power plant boilers enables the multi-modal sensor array of the data acquisition module to achieve three-dimensional synchronous measurement of the temperature field, concentration field, and flow field within the furnace. This provides accurate and complete data support for the intelligent analysis and command module. This refined data acquisition method significantly improves the accuracy of the combustion digital twin model, thereby enhancing the accuracy of combustion state assessment and contributing to a more efficient combustion process. By employing deep learning algorithms to construct the combustion digital twin model in the intelligent analysis and command module, and combining this with reinforcement learning to generate the optimal control strategy, the stability and efficiency of the combustion process are ensured. The feedback adjustment module dynamically corrects the control strategy by comparing the model's predicted values ​​with actual operating data in real time, forming an effective closed-loop optimization mechanism. This mechanism can promptly detect and correct model deviations, ensuring the continuity and stability of combustion optimization and adjustment.

[0031] According to some embodiments of the present invention, the data acquisition module further includes a data fusion algorithm, which is used to fuse data acquired by the infrared thermal imager, the laser gas analyzer and the ultrasonic flow meter to remove noise and outliers.

[0032] By aligning and correlating the originally discrete data from the infrared thermal imager, laser gas analyzer, and ultrasonic flow meter in time and space through the data acquisition module, the information on the temperature field, gas concentration field, and flow field is matched under a unified spatiotemporal coordinate system, facilitating subsequent analysis in the intelligent analysis and command module. Furthermore, by removing noise and outliers from each data point, the problem of occasional distortion in single-sensor data under harsh furnace conditions is resolved, making the data more accurate and reliable, and providing a more precise data foundation for the combustion digital twin model.

[0033] According to some embodiments of the present invention, the intelligent analysis and command module includes an intelligent analysis module and a control command generation module. The intelligent analysis module includes a multi-parameter fusion unit and a combustion digital twin model. The multi-parameter fusion unit is used to extract features and perform correlation analysis on the data collected by the data acquisition module, and transmits the data to the combustion digital twin model, which is used to generate a control strategy. The control command generation module is used to generate combined commands based on the control strategy.

[0034] The data acquisition module simultaneously acquires raw physical information about the temperature field, gas concentration field, and flow field within the furnace using an infrared thermal imager, laser gas analyzer, and ultrasonic flow meter, and transmits the collected data to the intelligent analysis and command module. The multi-parameter fusion unit in the intelligent analysis module performs deeper feature extraction and correlation analysis on the data, and then transmits the analysis results to the combustion digital twin model. Based on deep learning methods, the combustion digital twin model calculates the dynamic impact of different strategies on the combustion state, ultimately generating a strategy that maximizes boiler combustion efficiency while matching the power plant's electricity demand and meeting power production requirements, providing a decision-making basis for subsequent control command generation.

[0035] According to some embodiments of the present invention, the control command generation module includes a multi-objective optimization algorithm, which generates combined commands under constraints based on the calculation results of the intelligent analysis module. The constraints include: boiler load, combustion efficiency, and NO0. x Emissions and coking risks. Combined instructions include: the coal feeding speed and quantity of the variable frequency coal feeder, and the damper opening and blowing angle of the intelligent damper actuator.

[0036] By setting up a control command generation module, the abstract control strategy generated by the digital twin model is specifically transformed into combined command signals that can be directly executed by equipment such as variable frequency coal feeders and intelligent damper actuators.

[0037] Furthermore, in addition to constraints on boiler load and combustion efficiency, constraints on nitrogen oxide emissions and coking risk have been added. This approach optimizes economic indicators while also taking into account safety and environmental protection goals, avoiding the risk of boiler coking, and suppressing nitrogen oxide emissions. This maximizes combustion economy while ensuring unit safety and meeting ultra-low emission requirements.

[0038] For example, constraints also include: SO x Emissions.

[0039] According to some embodiments of the present invention, the actuator includes an adaptive adjustment device, which includes a PID controller and a fuzzy control algorithm. The adaptive adjustment device is used to compare the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator with the combined command, and adjust the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator to match the combined command. The response time of the adaptive adjustment device is ≤2 seconds.

[0040] By incorporating an adaptive adjustment device into the actuator, the system rapidly and accurately executes commands from the control command generation module using a PID controller and fuzzy control algorithm. Furthermore, when discrepancies exist between the actual situation and the commands, the system adjusts the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator to match the combined commands. This achieves dynamic matching of coal feed rate and air supply, allowing for more precise adjustments to the variable frequency coal feeder and intelligent damper actuator based on the control commands. This results in more accurate system execution and more effective realization of the results calculated by the intelligent analysis and command module, thereby maximizing combustion efficiency while meeting the boiler's requirements.

[0041] Reference Figure 1 According to a second aspect of the present invention, the method for optimizing and adjusting the combustion of a thermal power plant boiler utilizes a combustion optimization and adjustment system for a thermal power plant boiler according to a first aspect of the present invention. The method includes: The data acquisition module uses an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter to perform three-dimensional measurements of the temperature field, concentration field, and flow field inside the furnace, and obtains measurement data.

[0042] The intelligent analysis and instruction module constructs a combustion digital twin model based on deep learning algorithms, and combines principal component analysis and neural network methods to extract features and perform correlation analysis on the parameters collected by the data acquisition module to generate combined instructions.

[0043] The actuator changes the coal feed rate, coal feed speed, and opening degree and / or angle of the intelligent damper actuator of the variable frequency coal feeder according to the combined instructions.

[0044] The feedback adjustment module dynamically corrects the combustion digital twin model by comparing the predicted values ​​of the combustion digital twin model with the actual operating data collected by the data acquisition module in real time.

[0045] According to the combustion optimization and adjustment method for thermal power plant boilers of the present invention, the data acquisition module performs three-dimensional synchronous measurement of the temperature field, concentration field, and flow field in the furnace, providing accurate and complete data support for the intelligent analysis and command module. This refined data acquisition method significantly improves the accuracy of the combustion digital twin model, thereby enhancing the accuracy of combustion state assessment and contributing to a more efficient combustion process. By employing deep learning algorithms to construct the combustion digital twin model and combining reinforcement learning to generate the optimal control strategy, the stability and efficiency of the combustion process are ensured. The feedback adjustment module dynamically corrects the combustion digital twin model by comparing the model's predicted values ​​with actual operating data in real time, forming an effective closed-loop optimization mechanism. This mechanism can promptly detect and correct model deviations, ensuring the continuity and stability of combustion optimization and adjustment.

[0046] Reference Figure 1According to some embodiments of the present invention, the feedback adjustment module, in the process of comparing the model prediction values ​​with the actual operating data in real time, quickly adjusts the control strategy based on the deviation of key parameters such as temperature and pressure through the model predictive control algorithm.

[0047] By enabling the feedback adjustment module to quickly adjust the control strategy through model predictive control algorithms, the stability and accuracy of combustion optimization adjustments can be ensured, forming a more effective closed-loop optimization.

[0048] Reference Figure 1 According to some embodiments of the present invention, the combustion optimization and adjustment system for thermal power plant boilers is a combustion optimization and adjustment system for thermal power plant boilers according to the first aspect of the present invention. The intelligent analysis module adopts a combination of principal component analysis and neural network method, and uses a combustion digital twin model to generate control strategies.

[0049] After the data acquisition module transmits the collected data to the intelligent analysis and command module, the multi-parameter fusion unit in the intelligent analysis module performs deeper feature extraction and correlation analysis on the data, and then transmits the analysis results to the combustion digital twin model. Based on deep learning methods, the combustion digital twin model calculates the dynamic impact of different strategies on the combustion state, ultimately generating a strategy that maximizes boiler combustion efficiency while matching the power plant's electricity demand and meeting power production requirements, providing a decision-making basis for subsequent control command generation.

[0050] Reference Figure 1 According to some embodiments of the present invention, the control command generation module is based on a multi-objective optimization algorithm, considering combustion efficiency, NO... x Under constraints such as emissions and coking risks, the optimal combination of coal feed rate, damper opening, and burner angle is generated based on the control strategy.

[0051] For example, the risk of coking can be assessed by detecting flue gas temperature. If the flue gas temperature is too high, especially in critical areas such as the burner region, the ash particles will reach a molten or semi-molten state, making them more likely to adhere to the heated surfaces and form coke. By accurately measuring the flue gas temperature at different locations in the furnace using equipment such as infrared thermal imagers, and combining this with the boiler's structure and combustion characteristics, it is possible to determine which areas are likely to experience coking due to excessively high temperatures.

[0052] The control command generation module transforms the abstract control strategy generated by the digital twin model into a combination of command signals that can be directly executed by equipment such as variable frequency coal feeders and intelligent damper actuators.

[0053] Reference Figure 1According to some embodiments of the present invention, during the process of the intelligent analysis module generating the optimal control strategy and the control command generation module generating the optimal combination command, while optimizing combustion efficiency, pollution emission factors are comprehensively considered to control NO in the boiler of thermal power plants. x Emissions ≤30mg / Nm³, SO2 emissions ≤20mg / Nm³.

[0054] By enabling the intelligent analysis module to generate the optimal control strategy and the control command generation module to generate the optimal combination of commands, pollution emission factors are comprehensively considered. By optimizing combustion parameters, the system can achieve ultra-low emission standards for pollutants while improving combustion efficiency. This feature not only enhances the economic efficiency of thermal power plants but also significantly improves their environmental performance, helping thermal power plants meet increasingly stringent environmental requirements.

[0055] Reference Figure 1 According to some embodiments of the present invention, the intelligent analysis and instruction module adopts a dynamic adaptive algorithm, predicts the combustion trend through an LSTM network, adjusts the control parameters, and adapts to the changes in the maximum continuous output load of the boiler from 30% to 120%.

[0056] By employing a dynamic adaptive algorithm in the intelligent analysis and instruction module, predicting combustion trends through an LSTM network, and adjusting control parameters in real time using a reinforcement learning algorithm, fuel combustion can adapt to load changes ranging from 30% to 120% of the boiler's maximum continuous output, and to power generation under different electricity demands. Furthermore, the intelligent analysis and instruction module can optimize control strategies in a timely manner based on load changes, further enhancing the system's adaptability and robustness.

[0057] The following reference Figure 1 A method for optimizing and adjusting the combustion of a thermal power plant boiler according to an embodiment of the present invention is described.

[0058] First, the data acquisition module performs three-dimensional synchronous measurement of the temperature field, concentration field, and flow field inside the furnace using a multi-modal sensor array (infrared thermal imager, laser gas analyzer, and ultrasonic flow meter), and uses a data fusion algorithm to remove noise and outliers, providing accurate data for the intelligent analysis and command module.

[0059] Then, the intelligent analysis module constructs a combustion digital twin model based on deep learning algorithms, and combines principal component analysis and neural network methods to extract features and perform correlation analysis on parameters such as temperature, pressure, and oxygen to generate the optimal control strategy.

[0060] Next, the control command generation module, based on a multi-objective optimization algorithm, considers combustion efficiency, NO... x Under constraints such as emissions and coking risks, the optimal combination of coal feed rate, damper opening, and burner angle is generated.

[0061] Subsequently, the actuator uses a variable frequency coal feeder, an intelligent damper actuator, and a PID-based fuzzy control algorithm to achieve dynamic matching of coal feed rate and air supply rate, with a response time of ≤2 seconds. The coal feeder's feeding speed and rate, as well as the intelligent damper actuator's damper opening and angle, support ±0.1% accuracy adjustment.

[0062] Finally, the feedback adjustment module compares the model prediction values ​​with the actual operating data in real time. For deviations in key parameters such as temperature and pressure, it dynamically corrects the control strategy through the model predictive control algorithm to form a closed-loop optimization and adapt to changes in the boiler's maximum output load from 30% to 120%.

[0063] In addition, the system uses an LSTM network and reinforcement learning to adjust control parameters in real time to ensure that pollution emissions meet NO standards. x The ultra-low standards of ≤30mg / Nm³ and SO2≤20mg / Nm³ integrate pollution control into the core process of combustion optimization.

[0064] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0065] In the description of this invention, "first feature" and "second feature" may include one or more of the features.

[0066] In the description of this invention, "a plurality of" means two or more.

[0067] In the description of this invention, the first feature being "above" or "below" the second feature may include the first and second features being in direct contact, or it may include the first and second features not being in direct contact but being in contact through another feature between them.

[0068] In the description of this invention, the terms "above," "over," and "on top" for the first feature and the second feature include the first feature being directly above or diagonally above the second feature, or simply indicating that the first feature is at a higher horizontal level than the second feature.

[0069] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0070] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A combustion optimization and adjustment system for a thermal power plant boiler, characterized in that, include: The data acquisition module includes an infrared thermal imager, a laser gas analyzer, and an ultrasonic flow meter. The infrared thermal imager includes an infrared sensor array and is used to detect the temperature field inside the furnace. The laser gas analyzer includes a laser sensor array and is used to detect the gas concentration field inside the furnace. The ultrasonic flow meter includes an ultrasonic sensor array and is used to detect the flow field inside the furnace. The intelligent analysis and instruction module includes a combustion digital twin model constructed using deep learning algorithms. The intelligent analysis and instruction module is used to generate combined instructions based on the data collected by the data acquisition module. The actuator includes a variable frequency coal feeder and an intelligent damper actuator, used to control the coal feeding amount and coal feeding speed of the variable frequency coal feeder, and the damper opening and / or blowing angle of the intelligent damper actuator according to the combined instructions; The feedback adjustment module is used to compare real-time data with the calculation results of the intelligent analysis module and dynamically correct the combustion digital twin model.

2. The combustion optimization adjustment system for a utility boiler as set forth in claim 1, wherein The data acquisition module also includes a data fusion algorithm, which is used to fuse the data collected by the infrared thermal imager, the laser gas analyzer and the ultrasonic flow meter to remove noise and outliers.

3. The combustion optimization adjustment system for a utility boiler as set forth in claim 1, wherein The intelligent analysis and instruction module includes an intelligent analysis module and a control instruction generation module. The intelligent analysis module includes a multi-parameter fusion unit and the combustion digital twin model. The multi-parameter fusion unit is used to perform feature extraction and correlation analysis on the data collected by the data acquisition module and transmit it to the combustion digital twin model. The combustion digital twin model is used to generate control strategies. The control instruction generation module is used to generate the combined instructions according to the control strategy.

4. The combustion optimization adjustment system for a utility boiler as set forth in claim 3, wherein The control command generation module includes a multi-objective optimization algorithm, which is used to generate combined commands under constraints based on the calculation results of the intelligent analysis module. The constraints include: boiler load, combustion efficiency, NO x x emissions, coking risk; The combined commands include: the coal feeding speed and coal feeding amount of the variable frequency coal feeder, and the damper opening and blowing angle of the intelligent damper actuator.

5. The combustion optimization adjustment system for a utility boiler of claim 1, wherein, The actuator includes an adaptive adjustment device, which includes a PID controller and a fuzzy control algorithm. The adaptive adjustment device is used to compare the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator with the combined command, and adjust the coal feed rate of the variable frequency coal feeder and the damper opening of the intelligent damper actuator to match the combined command. The response time of the adaptive adjustment device is ≤2 seconds.

6. A method for optimizing adjustment of combustion in a boiler of a thermal power plant, characterized in that, The boiler combustion optimization and adjustment system for thermal power plants according to any one of claims 1-5 is used for adjustment, and the boiler combustion optimization and adjustment method for thermal power plants includes: The data acquisition module performs three-dimensional measurements of the temperature field, concentration field, and flow field inside the furnace using the infrared thermal imager, the laser gas analyzer, and the ultrasonic flow meter to obtain measurement data. The intelligent analysis and instruction module constructs the combustion digital twin model based on the deep learning algorithm, and performs feature extraction and correlation analysis on the parameters collected by the data acquisition module by combining principal component analysis and neural network methods to generate combined instructions; The actuator changes the coal feed rate and coal feed speed of the variable frequency coal feeder, and the opening degree and / or angle of the intelligent damper actuator according to the combined instructions; The feedback adjustment module dynamically corrects the combustion digital twin model by comparing the predicted values ​​of the combustion digital twin model with the actual operating data collected by the data acquisition module in real time.

7. The method of claim 6, wherein, The feedback adjustment module, in the process of comparing the model prediction values ​​with the actual operating data in real time, quickly adjusts the control strategy based on the deviation of key parameters such as temperature and pressure through the model predictive control algorithm.

8. The method of claim 6, wherein, The combustion optimization and adjustment system for thermal power plant boilers is the combustion optimization and adjustment system for thermal power plant boilers according to claim 4. The intelligent analysis module adopts a combination of principal component analysis and neural network method, and uses the combustion digital twin model to generate control strategies. The control instruction generation module generates, based on a multi-objective optimization algorithm, optimal combination instructions of the coal feeding amount, the damper opening degree, and the burner angle according to the control strategy under the constraints of the combustion efficiency, the NO x emission, the coking risk, and the like.

9. The method for optimizing and adjusting combustion in a thermal power plant boiler according to claim 8, characterized in that, During the process of the intelligent analysis module generating the optimal control strategy and the control command generation module generating the optimal combination command, while optimizing combustion efficiency, pollution emission factors are comprehensively considered to control NO in the boiler of thermal power plants. x Emissions ≤30mg / Nm³, SO2 emissions ≤20mg / Nm³.

10. The method of claim 6, wherein, The intelligent analysis and instruction module adopts a dynamic adaptive algorithm, which uses an LSTM network to predict combustion trends and adjust control parameters to adapt to changes in the boiler's maximum continuous output load from 30% to 120%.

Images