Intelligent control system for thermal power generating unit adapting to deep peak regulation

Abstract

This invention relates to the field of automated control technology for thermal power generation, and discloses an intelligent control system for thermal power units adapted to deep peak shaving. The system includes an operating condition sensing module, a coal quality characteristic analysis module, a multi-objective optimization decision-making module, a combustion stability optimization module, a boiler efficiency optimization module, a pollution emission optimization module, an online parameter early warning module, a measure formulation module, an instruction execution feedback module, and a visualization management module. By constructing a soft measurement model of coal calorific value and performing multi-objective optimization decisions, a comprehensive evaluation index is established based on multi-source data fusion, including an online soft measurement model of coal quality, a combustion stability margin index, a comprehensive deviation coefficient of operating condition parameters, and a coal quality-combustion matching degree coefficient. This enables the system to accurately identify coal quality changes in real time and to conduct integrated online quantitative evaluation and collaborative decision-making on combustion stability, operational economy, and environmental matching degree, effectively overcoming the shortcomings of blind coal quality adjustment in traditional control systems.

Description

Technical Field

[0001] This invention relates to the field of automated control technology for thermal power generation, specifically to an intelligent control system for thermal power units adapted to deep peak shaving. Background Technology

[0002] With the increasing proportion of renewable energy power generation, traditional coal-fired power units need to undertake more arduous deep peak shaving tasks. Under deep peak shaving conditions with low and variable loads, the units face multiple severe challenges: First, frequent fluctuations in the quality of coal fed into the furnace lead to poor combustion stability, easily causing safety accidents such as fire extinguishing and deflagration, and making it difficult to stably control the main steam parameters; second, economic indicators such as boiler efficiency and coal consumption for power generation deteriorate significantly, resulting in a substantial reduction in operational economy; third, in order to maintain low-load combustion, the temperature field and air distribution characteristics in the furnace change, leading to increased fluctuations in the generation and concentration of pollutants such as nitrogen oxides, making it more difficult to achieve environmental compliance control. Existing distributed control systems (DCS) and traditional coordinated control strategies mainly rely on fixed or simple variable parameter set curves, making it difficult to perceive dynamic changes in coal quality online. They lack the ability to coordinate and optimize the conflict of multiple objectives such as combustion stability, economy, and environmental protection, resulting in lagging control response and heavy reliance on the experience and judgment of operators. They can no longer meet the high standards of deep peak shaving for unit flexibility, safety, and coordinated optimization of economy and environmental protection. Summary of the Invention

[0003] (a) Technical problems to be solved

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent control system for thermal power units adapted to deep peak shaving. It has the advantages of multi-objective collaborative optimization, online coal quality self-adaptation, intelligent early warning and closed-loop control. It solves the problems of combustion instability, efficiency reduction and emission exceedance caused by coal quality fluctuations under deep peak shaving conditions, as well as the difficulty in multi-objective collaborative optimization and control response lag.

[0005] (II) Technical Solution

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control system for thermal power units adapted to deep peak shaving, comprising an operating condition sensing module, a coal quality characteristic analysis module, a multi-objective optimization decision-making module, a combustion stability optimization module, a boiler efficiency optimization module, a pollution emission optimization module, a parameter online early warning module, a measure formulation module, an instruction execution feedback module, and a visualization management module;

[0007] The operating condition sensing module is used to acquire unit operating status parameters in real time.

[0008] The coal quality characteristic analysis module acquires multi-source data in real time during the combustion process by deploying collection stations, and establishes a soft measurement model of coal calorific value based on this data, thereby accurately identifying the dynamic changes in coal calorific value and combustion characteristics online.

[0009] The multi-objective optimization decision-making module calculates the combustion stability margin index based on the collected data. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient It is used to formulate control strategies that coordinate safety, environmental protection, and economic objectives;

[0010] The combustion stability optimization module is based on the combustion stability margin index. Anticipate and prevent risks of fire extinguishing, deflagration, or significant fluctuations in main steam pressure caused by sudden changes in coal quality;

[0011] The boiler efficiency optimization module uses a comprehensive deviation coefficient based on operating parameters. Assess the degree of deviation between the current operating conditions and the optimal operating conditions, identify efficiency losses caused by improper parameter matching, and guide the optimization and adjustment of boiler-side operating parameters;

[0012] The pollution emission optimization module is based on the coal quality-combustion matching coefficient. Assess the compatibility and environmental compliance of current coal quality and combustion methods, identify the risk of exceeding emission standards and incomplete combustion losses caused by changes in coal quality, and guide the optimization of air distribution, burner adjustment and denitrification system operation;

[0013] The parameter online early warning module compares the real-time data of a single key parameter received online with the set safety threshold, economic range or environmental protection standard, and provides graded online early warning.

[0014] The measure formulation module automatically generates and prioritizes targeted control measures based on the optimization direction output by the multi-objective optimization decision module and the specific alarm information from the parameter online early warning module.

[0015] The instruction execution feedback module receives specific control instructions issued by the measure formulation module, drives the relevant actuators in the unit's DCS to perform actions according to priority, and monitors the instruction execution status and final effect in real time.

[0016] The visualization management module connects the data flow and business flow of all other modules in the system, providing operators with a unified intelligent monitoring and decision support platform.

[0017] Preferably, the operating condition sensing module collects unit load, main steam pressure, furnace negative pressure, superheated steam temperature, primary air pressure, feedwater flow rate and power grid frequency signals through a distributed sensor network, and filters and synchronizes the collected data before transmitting it to the multi-objective optimization decision module via the network.

[0018] Preferably, the coal quality characteristic analysis module includes a pulverized coal sampling point at the furnace inlet, a flue gas composition analysis point, a boiler efficiency calculation point, and a fly ash carbon content detection point. Based on the data collected from the stations, a soft measurement model of coal calorific value is established to calculate the lower heating value, volatile matter content, and ash characteristics of the coal entering the furnace in real time, and the coal quality fluctuation information is transmitted to the multi-objective optimization decision module.

[0019] Preferably, the pulverized coal sampling point at the furnace inlet collects pulverized coal samples entering the furnace using an automatic sampling device, and detects real-time data on elemental and industrial analysis of the pulverized coal; the flue gas composition analysis point collects flue gas samples from the tail flue of the boiler using a high-temperature sampling probe, and detects real-time data on the concentration of the main components in the flue gas.

[0020] Preferably, the boiler efficiency calculation point uses an integrated online calculation model to back-calculate and verify the comprehensive combustion characteristics and effective calorific value of the current coal in real time; the fly ash carbon content detection point uses an online laser-induced breakdown spectroscopy analyzer or microwave measurement device to directly analyze and obtain real-time detection data of unburned carbon content in fly ash, and outputs the real-time boiler efficiency value for verification of the coal quality soft measurement model.

[0021] Preferably, the multi-objective optimization decision module calculates the combustion stability margin index based on the collected data. The calculation formula is as follows: , In the formula, This indicates the standard deviation of the furnace negative pressure sliding window. Indicates the maximum permissible standard deviation, and µ represents the furnace negative pressure weighting coefficient. This represents the current measured wind pressure value. This indicates the primary air pressure setpoint corresponding to the current load. This represents the primary wind pressure weighting coefficient. Indicates that the product is dry and free of ash and volatile matter. This indicates the baseline value of volatile matter for the designed coal type. This represents the weighting coefficient for sudden changes in coal quality. This represents the rate of change of volatile matter.

[0022] Preferably, the multi-objective optimization decision module calculates the comprehensive deviation coefficient of operating condition parameters based on the collected data. The calculation formula is as follows: , In the formula, This represents the measured value of the superheated steam temperature. This indicates the superheated steam temperature setpoint corresponding to the current load. This represents the measured value of the main steam pressure. This indicates the main steam pressure setpoint. This indicates the current boiler efficiency. This represents the maximum achievable boiler efficiency under the current coal quality. This represents the measured value of the water supply flow rate. This indicates the optimal water supply flow rate under the current operating conditions. This represents the weighting coefficient for steam temperature deviation. This represents the weighting coefficient for steam pressure deviation. This represents the efficiency loss weighting coefficient. This represents the weighting coefficient for water supply flow deviation.

[0023] Preferably, the multi-objective optimization decision module calculates the coal quality-combustion matching coefficient based on the collected data. The calculation formula is as follows: , In the formula, Indicating in flue gas Measured concentration Indicates environmental standards Emission limits, Indicating in flue gas Measured concentration Indicates environmental standard S Emission limits, This indicates the carbon content of fly ash. This indicates the standard operating limit for carbon content control in fly ash. Indicates the excess air coefficient. This represents the optimal excess air coefficient under the current coal quality. Indicates that the product is dry and free of ash and volatile matter. Indicates the volatile matter content of the designed coal type. , , , , They represent Emissions, S Emissions, incomplete combustion loss, air-coal ratio matching and coal quality adaptability weighting coefficients.

[0024] Preferably, the online parameter warning module has built-in threshold management and logical judgment functions. When any parameter exceeds the limit, it automatically triggers an audible and visual alarm and generates a warning event, while simultaneously transmitting the abnormal information and the combustion stability margin index. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient The associated optimization index status is synchronously pushed to the measure formulation module and the visualization management module; the measure formulation module integrates an expert rule base and optimization algorithm, transforming abstract optimization objectives and specific abnormal parameters into a list of executable control instructions. When the index is low and accompanied by a negative pressure fluctuation warning, priority should be given to formulating measures to stabilize the gas supply; when When the index is high and a smoke exhaust temperature warning is issued, efficiency improvement measures will be formulated.

[0025] Preferably, the instruction execution feedback module sends control instructions to the DCS and simultaneously monitors the feedback signals of the actuator and the subsequent parameter changes of the working condition perception module. When the instruction execution deviation exceeds the limit or the relevant parameters / optimization index do not improve within the expected time, it is determined to be an execution abnormality or poor effect. The system automatically feeds the information back to the measure formulation module for strategy retuning and simultaneously prompts in the visualization management module, forming a closed-loop control loop of instruction-execution-feedback-optimization.

[0026] Compared with existing technologies, this invention provides an intelligent control system for thermal power units that adapts to deep peak shaving, and has the following beneficial effects:

[0027] 1. This invention establishes an online soft measurement model for coal quality and a combustion stability margin index based on multi-source data fusion by constructing a soft measurement model for coal calorific value and performing multi-objective optimization decisions. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient The comprehensive evaluation index enables the system to accurately identify changes in coal quality in real time, and to conduct integrated online quantitative evaluation and collaborative decision-making on combustion stability, operational economy and environmental protection matching, effectively overcoming the drawbacks of blind adjustment of coal quality in traditional control.

[0028] 2. This invention achieves the beneficial effect of rapidly triggering system optimization from "point" parameter anomalies by designing a closed-loop linkage mechanism of online early warning of design parameters, measure formulation and instruction execution feedback modules. It also transforms abstract decisions into specific executable instructions, ultimately forming a closed-loop intelligent control of perception-decision-execution-verification, which greatly improves the system's adaptability and rapid response capability to complex working conditions of deep peak shaving.

[0029] 3. By integrating a visualization management module, this invention presents multi-source data, optimization indices, early warning information, control measures, and execution feedback in a unified and visualized manner. This achieves the beneficial effect of providing operators with an integrated intelligent platform that provides panoramic monitoring, decision support, and human-machine collaborative intervention. Ultimately, it reduces the difficulty and risk of deep peak shaving operations and improves the overall management and control level of the system. Attached Figure Description

[0030] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please see Figure 1 An intelligent control system for thermal power units adapted to deep peak shaving includes an operating condition sensing module, a coal quality characteristic analysis module, a multi-objective optimization decision-making module, a combustion stability optimization module, a boiler efficiency optimization module, a pollution emission optimization module, a parameter online early warning module, a measure formulation module, an instruction execution feedback module, and a visualization management module.

[0033] The operating condition sensing module is used to acquire unit operating status parameters in real time;

[0034] The coal quality characteristics analysis module acquires multi-source data related to the combustion process in real time by deploying acquisition sites, and establishes a soft measurement model of coal calorific value based on this data, thereby accurately identifying the dynamic changes of coal calorific value and combustion characteristics online.

[0035] The multi-objective optimization decision module calculates the combustion stability margin index based on the collected data. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient It is used to formulate control strategies that coordinate safety, environmental protection, and economic objectives;

[0036] The combustion stability optimization module is based on the combustion stability margin index. Anticipate and prevent risks of fire extinguishing, deflagration, or significant fluctuations in main steam pressure caused by sudden changes in coal quality (such as a sudden change in volatile matter);

[0037] The boiler efficiency optimization module integrates deviation coefficients based on operating parameters. Assess the degree of deviation between the current operating conditions and the optimal operating conditions, identify the efficiency loss caused by improper parameter matching, guide the optimization and adjustment of boiler-side operating parameters, and achieve economical operation under deep peak shaving conditions;

[0038] The pollution emission optimization module is based on the coal quality-combustion matching coefficient. Assess the current matching degree between coal quality and combustion mode and environmental compliance, identify the risk of exceeding emission standards and loss due to incomplete combustion caused by changes in coal quality, guide the optimization of air distribution, burner adjustment and denitrification system operation, and ensure that environmental indicators meet the standards during deep peak shaving;

[0039] The parameter online early warning module compares real-time data of a single key parameter (such as furnace negative pressure, main steam temperature, oxygen content, NOx concentration, etc.) received online with the set safety threshold, economic range or environmental protection standard, and performs online early warning in a graded (such as early warning, alarm) and classified (safety, economy, environmental protection) manner.

[0040] The measure formulation module automatically generates and prioritizes targeted control measures based on the optimization direction (Hs, Kr, Uo trends) output by the multi-objective optimization decision module and the specific alarm information from the parameter online early warning module.

[0041] Command execution feedback module: Receives specific control commands issued by the measure formulation module, drives the relevant actuators (such as coal feeder, damper, ammonia injection regulating valve) in the unit DCS (distributed control system) to act, and monitors the command execution status and final effect in real time;

[0042] Visual management module: Connects the data flow and business flow of all other modules in the system, providing operators with a unified intelligent monitoring and decision support platform.

[0043] The operating condition sensing module collects unit load, main steam pressure, furnace negative pressure, superheated steam temperature, primary air pressure, feedwater flow rate and power grid frequency signals through a distributed sensor network, and filters and synchronizes the collected data before transmitting it to the multi-objective optimization decision module via the network.

[0044] The coal quality characteristics analysis module includes four data collection points: the pulverized coal sampling point at the furnace inlet, the flue gas composition analysis point, the boiler efficiency calculation point, and the fly ash carbon content detection point. Based on the data from these four data collection points, a soft measurement model for coal calorific value is established to calculate the lower heating value, volatile matter content, and ash characteristics of the coal entering the furnace in real time, and the coal quality fluctuation information is transmitted to the multi-objective optimization decision-making module.

[0045] The pulverized coal sampling point at the furnace inlet is used to collect pulverized coal samples entering the furnace using an automatic sampling device. These samples are then sent to an online coal quality analyzer (such as a PGNAA transient gamma neutron activation analyzer) for testing, obtaining real-time data on the elemental analysis (C, H, O, N, S) and industrial analysis (moisture, ash, volatile matter) of the pulverized coal. The flue gas composition analysis point is used to collect flue gas samples from the boiler tail flue using a high-temperature sampling probe. These samples are then sent to a multi-component flue gas analyzer for testing, obtaining data on oxygen (O2), carbon monoxide (CO), carbon dioxide (CO2), and nitrogen oxides (NOx) in the flue gas. x Real-time data on sulfur dioxide (SO2) concentration.

[0046] The boiler efficiency calculation point uses an online calculation model that integrates boiler inputs (fuel quantity, coal calorific value), outputs (main steam flow, temperature, pressure), and various heat losses (flue gas loss, unburned carbon loss, etc.) to back-calculate and verify the comprehensive combustion characteristics and effective calorific value of the current coal in real time. The fly ash carbon content detection point uses an online laser-induced breakdown spectroscopy (LIBS) analyzer or microwave measurement device to directly analyze fly ash samples in the inlet flue of the electrostatic precipitator, obtain real-time detection data of unburned carbon content in fly ash, and output the real-time boiler efficiency value for verification of the coal quality soft measurement model.

[0047] The advantages are: by constructing a multi-source data acquisition network that includes the PGNAA online coal quality analyzer, LIBS fly ash carbon content detection, and flue gas composition analysis, a soft measurement model for coal calorific value is established, enabling the system to accurately identify the dynamic changes in the calorific value, volatile matter, and ash content of coal in real time, realizing the transparency of coal quality information, providing prior knowledge for combustion optimization, and thus solving the defects of traditional control systems that cannot detect changes in coal quality and respond passively.

[0048] The multi-objective optimization decision module calculates the combustion stability margin index based on the collected data. The calculation formula is as follows: , In the formula, The standard deviation of the furnace negative pressure sliding window is derived from the distributed sensor network - furnace negative pressure measuring points. = , ( This represents the i-th sampled value within the window, with a sampling period typically between 1 and 2 seconds. (represents the window average). This represents the maximum permissible standard deviation (design value), derived from boiler design specifications. µ represents the furnace negative pressure weighting coefficient, ranging from 0.4 to 0.5. This represents the current measured wind pressure value. This indicates the primary air pressure setpoint corresponding to the current load. This represents the primary wind pressure weighting coefficient, with a value ranging from 0.3 to 0.4. The percentage of dry, ash-free volatile matter is derived from the Coal Quality Characteristics Analysis Module - Coal Quality Soft Measurement Model. This represents the baseline value for volatile matter in the design coal, derived from the boiler design coal quality parameters. This represents the coal quality mutation weighting coefficient, with a value ranging from 0.2 to 0.3. This indicates the rate of change of volatile matter (% / min).

[0049] The advantage is that it allows for the calculation of the combustion stability margin index. It will be updated in real time. Used to predict and prevent risks of fire extinguishing, deflagration, or significant fluctuations in main steam pressure caused by sudden changes in coal quality (such as abrupt changes in volatile matter); when When the concentration is less than 15%, it is considered a stable zone, indicating that the combustion in the furnace is stable, the negative pressure fluctuation in the furnace is slight, the primary air pressure is well matched, the coal quality has no sudden changes, the burner flame detection is normal, and it is in a safe and controllable state; when 15% ≤ When the combustion margin is less than 30%, it is considered a warning zone, indicating a decrease in combustion margin. One of the following situations may occur: increased furnace negative pressure fluctuations approaching the allowable upper limit, primary air pressure deviating from the set value by more than 10%, or a volatile matter change rate exceeding 3% / min is detected, requiring a warning. When the concentration is ≥30%, it is considered a danger zone, indicating a risk of unstable combustion. This may result in: severe fluctuations in furnace negative pressure (approaching the fire extinguishing protection setpoint), serious mismatch in primary air pressure, or sudden changes in volatile matter (>8% / min) leading to flame flickering / instability, posing a risk of fire extinguishing, deflagration, or significant fluctuations in main steam pressure.

[0050] The multi-objective optimization decision-making module calculates the comprehensive deviation coefficient of operating condition parameters based on the collected data. The calculation formula is as follows: , In the formula, The measured value (°C) of superheated steam temperature is derived from a distributed sensor network. This indicates the superheated steam temperature setpoint (°C) corresponding to the current load, derived from the operating procedures. The measured value of the main steam pressure (MPa) is derived from a distributed sensor network. This indicates the main steam pressure setpoint (MPa). This indicates the current boiler efficiency (%). This indicates the maximum achievable boiler efficiency (%) under the current coal quality. This represents the measured value of water supply flow rate (t / h). This indicates the optimal water supply flow rate (t / h) under the current operating conditions. This represents the weighting coefficient for steam temperature deviation, with a value ranging from 0.25 to 0.30. This represents the weighting coefficient for steam pressure deviation, with a value ranging from 0.25 to 0.30. This represents the efficiency loss weighting coefficient, with a value ranging from 0.30 to 0.35. This represents the weighting coefficient for water supply flow deviation, with a value ranging from 0.10 to 0.15.

[0051] The advantage is that it calculates the comprehensive deviation coefficient of the operating condition parameters. It will be updated in real time. Used to quantitatively assess the deviation between current operating conditions and optimal operating conditions, identify efficiency losses caused by improper parameter matching, guide the optimization and adjustment of boiler-side operating parameters, and achieve economical operation under deep peak-shaving conditions. When ≤5%, it indicates that the current operating conditions are excellent, the superheated steam temperature and main steam pressure accurately track the set values, the boiler efficiency is close to the maximum achievable efficiency under the current coal quality, the feedwater flow rate is reasonably configured, all parameters are matched and coordinated, and the economy is at its optimal state; when 5% < When the deviation is ≤12%, it indicates that there is room for optimization, and one of the following situations may occur: steam temperature / pressure deviates from the set value by 2% to 5%, boiler efficiency loss is 2% to 5%, or feedwater flow rate deviates from the optimal value by 3% to 8%, resulting in a decrease in economy but not exceeding the limit; when ≥12% indicates a significant deviation in operating conditions, which may include: severe deviation in steam temperature / pressure (>5%), boiler efficiency loss >5% (such as abnormal increase in flue gas temperature or excessive carbon content in fly ash), or severe mismatch in feedwater flow, leading to a sharp increase in coal consumption or a threat to equipment safety.

[0052] The multi-objective optimization decision-making module calculates the coal quality-combustion matching coefficient based on the collected data. The calculation formula is as follows: , In the formula, Indicating in flue gas Measured concentration (mg / Nm³, 6%) (Converted), derived from flue gas composition analysis points, Indicates environmental standards Emission limits (mg / Nm³). Indicating in flue gas Measured concentration (mg / Nm³, 6%) (Converted), derived from flue gas composition analysis points, Indicates environmental standard S Emission limits (mg / Nm³). The percentage of carbon content in fly ash is indicated by the fly ash carbon content monitoring point. This indicates the standard control limit (%) for carbon content in fly ash. Indicates the excess air coefficient. This represents the optimal excess air coefficient under the current coal quality. Indicates dry, ash-free volatile matter (%). This indicates the volatile matter content (%) of the designed coal type. , , , , They represent Emissions (values ​​range from 0.25 to 0.30), S Weighting coefficients for emissions (values ​​0.20~0.25), incomplete combustion loss (values ​​0.20~0.25), air-coal ratio matching (values ​​0.15~0.20), and coal quality adaptability (values ​​0.10~0.15);

[0053] The advantage is that it allows for the calculation of the coal quality-combustion matching coefficient. This is used to assess the compatibility of current coal quality and combustion methods with environmental compliance, identify the risk of excessive emissions and incomplete combustion losses due to changes in coal quality, guide air distribution optimization, burner adjustment, and denitrification system operation, and ensure that environmental indicators meet standards during deep peak shaving. A value less than 0.8 indicates an excellent coal quality, meaning the current coal quality and combustion method are well-matched. and The emission margin is sufficient, the fly ash carbon content is low, the air-to-coal ratio is reasonable, the combustion efficiency is high, and the environmental protection indicators are far superior to the limits; when 0.8≤ When the value is less than 1, it is a warning zone, indicating that the current coal quality and combustion method are close to the optimal matching boundary. or Emissions approaching limits, or fly ash carbon content nearing control limits, or the air-fuel ratio deviating from optimal values, require optimization and adjustment to prevent exceeding standards; when A value ≥1 indicates a danger zone, signifying an imbalance between the current coal quality and combustion method, and the presence of [problems / issues]. or The risk of exceeding emission standards, or the risk of excessive carbon content in fly ash leading to a sharp increase in losses due to incomplete combustion, or the risk of combustion instability due to sudden changes in coal quality, requires immediate intervention.

[0054] The online parameter warning module has built-in threshold management and logic judgment functions. When any parameter exceeds the limit, it automatically triggers an audible and visual alarm and generates a warning event, while simultaneously displaying the abnormal information and the combustion stability margin index. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient The status of the associated optimization index is synchronously pushed to the measure formulation module and the visualization management module, enabling rapid location and response triggering of system problems from "point" parameter anomalies to "surface" system problems.

[0055] The advantages are: by establishing a two-layer early warning mechanism of single-parameter point early warning and multi-exponential surface correlation, the system can quickly map and locate from local anomalies to system risks, avoiding the problem of traditional DCS systems where a large number of single-point alarm messages overwhelm operators and fail to quickly identify core contradictions; by linking single-parameter limit exceeding with... , , The three optimization indices are automatically linked, providing operators with a complete decision-making chain and significantly shortening the response time and improving the accuracy of handling abnormal system conditions.

[0056] The measure formulation module integrates an expert rule base and optimization algorithms, transforming abstract optimization objectives and specific anomaly parameters into a list of executable control instructions. When the index is low and accompanied by a negative pressure fluctuation warning, priority should be given to implementing measures to stabilize combustion, such as adjusting the coal feeding rate and primary air volume ratio; when When the index is high and the flue gas temperature is under warning, efficiency improvement measures should be formulated, such as optimizing the opening of the secondary air damper.

[0057] The advantages are: by integrating an expert rule base (based on operating procedures and historical experience) with multi-objective optimization algorithms (such as NSGA-II or weighted TOPSIS), it achieves a leap from human experience-based decision-making to intelligent assisted decision-making, freeing operators from the pressure of massive information analysis and decision-making; by based on , , The system automatically matches and prioritizes control strategies based on real-time status, thereby ensuring the dynamic balance and coordinated optimization of safety, economy, and environmental protection objectives under deep peak-shaving conditions, and ultimately avoiding the phenomenon that optimization of a single objective leads to the deterioration of other objectives.

[0058] The instruction execution feedback module sends control instructions to the DCS and simultaneously monitors the feedback signals of the actuators and the subsequent parameter changes of the working condition perception module. When the instruction execution deviation exceeds the limit or the relevant parameters / optimization index do not improve within the expected time, it is judged as an execution abnormality or poor effect. The system automatically feeds the information back to the measure formulation module for strategy retuning and simultaneously prompts in the visualization management module, forming a closed-loop control loop of instruction-execution-feedback-optimization.

[0059] The advantages are: by establishing a closed-loop control loop of command issuance-execution monitoring-effect evaluation-strategy retuning, adaptive correction and continuous optimization of the control strategy are achieved, solving the problems of inability to evaluate the strategy execution effect and inability to adaptively adjust to abnormal operating conditions in traditional open-loop control or simple closed-loop control; through seamless integration with the DCS system, advanced optimization functions are superimposed while retaining the basic control functions of the DCS, which can make full use of existing control equipment and improve the system control quality, and has the advantages of convenient engineering implementation and low transformation cost.

[0060] The visualization management module integrates panoramic flowcharts, real-time trend curves, optimization index dashboards, early warning lists, measure execution logs, and performance statistics reports. It visualizes complex multi-source information and decision-making processes, enabling operators to intuitively grasp the overall plant status, understand system optimization logic, confirm or intervene in automatic control measures, and manage historical data and optimization cases. It is the core interactive interface for realizing human-machine collaboration and intelligent system management.

[0061] The advantages are: A global data integration network built through a unified architecture of Industrial Ethernet and OPC UA enables millisecond-level synchronization and seamless sharing of data across all functional modules of the system, ultimately breaking down information silos in traditional control systems; this module transforms complex optimization algorithms and multi-source data into graphical information that operators can understand, trust, and intervene in through panoramic visualization and intelligent decision support, achieving efficient collaboration; and through historical data management and case study functions, it accumulates valuable knowledge assets for deep peak-shaving operation, providing data support for continuous optimization and personnel training, thereby improving the intelligence level and knowledge transfer capabilities of the system's thermal power unit operation management.

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

Claims

1. An intelligent control system for thermal power units adapted to deep peak shaving, characterized in that, It includes a working condition perception module, a coal quality characteristic analysis module, a multi-objective optimization decision-making module, a combustion stability optimization module, a boiler efficiency optimization module, a pollution emission optimization module, a parameter online early warning module, a measure formulation module, an instruction execution feedback module, and a visualization management module; The operating condition sensing module is used to acquire unit operating status parameters in real time. The coal quality characteristic analysis module acquires multi-source data in real time during the combustion process by deploying collection stations, and establishes a soft measurement model of coal calorific value based on this data, thereby accurately identifying the dynamic changes in coal calorific value and combustion characteristics online. The multi-objective optimization decision-making module calculates the combustion stability margin index based on the collected data. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient It is used to formulate control strategies that coordinate safety, environmental protection, and economic objectives; The combustion stability optimization module is based on the combustion stability margin index. Anticipate and prevent risks of fire extinguishing, deflagration, or significant fluctuations in main steam pressure caused by sudden changes in coal quality; The boiler efficiency optimization module uses a comprehensive deviation coefficient based on operating parameters. Assess the degree of deviation between the current operating conditions and the optimal operating conditions, identify efficiency losses caused by improper parameter matching, and guide the optimization and adjustment of boiler-side operating parameters; The pollution emission optimization module is based on the coal quality-combustion matching coefficient. Assess the compatibility and environmental compliance of current coal quality and combustion methods, identify the risk of exceeding emission standards and incomplete combustion losses caused by changes in coal quality, and guide the optimization of air distribution, burner adjustment and denitrification system operation; The parameter online early warning module compares the real-time data of a single key parameter received online with the set safety threshold, economic range or environmental protection standard, and provides graded online early warning. The measure formulation module automatically generates and prioritizes targeted control measures based on the optimization direction output by the multi-objective optimization decision module and the specific alarm information from the parameter online early warning module. The instruction execution feedback module receives specific control instructions issued by the measure formulation module, drives the relevant actuators in the unit's DCS to perform actions according to priority, and monitors the instruction execution status and final effect in real time. The visualization management module connects the data flow and business flow of all other modules in the system, providing operators with a unified intelligent monitoring and decision support platform.

2. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The operating condition sensing module collects unit load, main steam pressure, furnace negative pressure, superheated steam temperature, primary air pressure, feedwater flow rate and power grid frequency signals through a distributed sensor network, and filters and synchronizes the collected data before transmitting it to the multi-objective optimization decision module via the network.

3. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The coal quality characteristic analysis module includes a pulverized coal sampling point at the furnace inlet, a flue gas composition analysis point, a boiler efficiency calculation point, and a fly ash carbon content detection point. Based on the data collected from the stations, a soft measurement model of coal calorific value is established to calculate the lower heating value, volatile matter content, and ash characteristics of the coal entering the furnace in real time, and the coal quality fluctuation information is transmitted to the multi-objective optimization decision module.

4. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 3, characterized in that, The pulverized coal sampling point at the furnace inlet collects pulverized coal samples from the furnace using an automatic sampling device, and detects real-time data on elemental and industrial analysis of the pulverized coal. The flue gas composition analysis point collects flue gas samples from the boiler tail flue using a high-temperature sampling probe, and detects real-time data on the concentration of the main components in the flue gas.

5. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 3, characterized in that, The boiler efficiency calculation point uses an integrated online calculation model to back-calculate and verify the comprehensive combustion characteristics and effective calorific value of the current coal in real time; the fly ash carbon content detection point uses an online laser-induced breakdown spectroscopy analyzer or microwave measurement device to directly analyze the real-time detection data of the unburned carbon content in fly ash, and outputs the real-time boiler efficiency value for verification of the coal quality soft measurement model.

6. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The multi-objective optimization decision-making module calculates the combustion stability margin index based on the collected data. The calculation formula is as follows: ， In the formula, This indicates the standard deviation of the furnace negative pressure sliding window. Indicates the maximum permissible standard deviation, and µ represents the furnace negative pressure weighting coefficient. This represents the current measured wind pressure value. This indicates the primary air pressure setpoint corresponding to the current load. This represents the primary wind pressure weighting coefficient. Indicates that the product is dry and free of ash and volatile matter. This indicates the baseline value of volatile matter for the designed coal type. This represents the weighting coefficient for sudden changes in coal quality. This represents the rate of change of volatile matter.

7. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The multi-objective optimization decision-making module calculates the comprehensive deviation coefficient of operating condition parameters based on the collected data. The calculation formula is as follows: ， In the formula, This represents the measured value of the superheated steam temperature. This indicates the superheated steam temperature setpoint corresponding to the current load. This represents the measured value of the main steam pressure. This indicates the main steam pressure setpoint. This indicates the current boiler efficiency. This represents the maximum achievable boiler efficiency under the current coal quality. This represents the measured value of the water supply flow rate. This indicates the optimal water supply flow rate under the current operating conditions. This represents the weighting coefficient for steam temperature deviation. This represents the weighting coefficient for steam pressure deviation. This represents the efficiency loss weighting coefficient. This represents the weighting coefficient for water supply flow deviation.

8. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The multi-objective optimization decision-making module calculates the coal quality-combustion matching coefficient based on the collected data. The calculation formula is as follows: ， In the formula, Indicating in flue gas Measured concentration Indicates environmental standards Emission limits, Indicating in flue gas Measured concentration Indicates environmental standard S Emission limits, This indicates the carbon content of fly ash. This indicates the standard operating limit for carbon content control in fly ash. Indicates the excess air coefficient. This represents the optimal excess air coefficient under the current coal quality. Indicates that the product is dry and free of ash and volatile matter. Indicates the volatile matter content of the designed coal type. , , , , They represent Emissions, S Emissions, incomplete combustion loss, air-coal ratio matching and coal quality adaptability weighting coefficients.

9. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The online parameter early warning module has built-in threshold management and logical judgment functions. When any parameter exceeds the limit, it automatically triggers an audible and visual alarm and generates an early warning event, while simultaneously transmitting the abnormal information and the combustion stability margin index. Comprehensive deviation coefficient of operating parameters Coal quality-combustion matching coefficient The associated optimization index status is synchronously pushed to the measure formulation module and the visualization management module; the measure formulation module integrates an expert rule base and optimization algorithm, transforming abstract optimization objectives and specific abnormal parameters into a list of executable control instructions. When the index is low and accompanied by a negative pressure fluctuation warning, priority should be given to formulating measures to stabilize the gas supply; when When the index is high and a smoke exhaust temperature warning is issued, efficiency improvement measures will be formulated.

10. The intelligent control system for thermal power units adapted to deep peak shaving according to claim 1, characterized in that, The instruction execution feedback module sends control instructions to the DCS and simultaneously monitors the feedback signals of the actuator and the subsequent parameter changes of the working condition perception module. When the instruction execution deviation exceeds the limit or the relevant parameters / optimization index do not improve within the expected time, it is determined to be an execution abnormality or poor effect. The system automatically feeds the information back to the measure formulation module for strategy retuning and simultaneously prompts in the visualization management module, forming a closed-loop control loop of instruction-execution-feedback-optimization.

Images