Intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation

The intelligent control system with dynamic boundary adaptation calculates and adjusts the load, pressure and temperature control boundaries of thermal power units in real time, solving the problem of fixed boundaries in deep peak shaving control and improving peak shaving safety and load response flexibility.

CN123113590APending Publication Date: 2026-11-17FUJIAN HUADIAN SHAOWU CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611466100.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the deep peak shaving control of thermal power units, the control boundary is fixed, the control parameters lack adaptive adjustment, and the boundary conditions lack dynamic quantification. This leads to a large difference between the unit's operating state and the design conditions, poor control quality, and problems such as equipment risks and limited regulation capabilities.

Method used

The system adopts an intelligent control system based on dynamic boundary adaptation. Through the unit data acquisition, preprocessing, dynamic boundary calculation and feedback correction modules, it calculates and adjusts the safety control boundaries in real time, including the upper and lower limits of load, pressure and temperature, to achieve adaptive adjustment under all operating conditions.

Benefits of technology

It improves peak-shaving safety and load response flexibility, avoids the risk of equipment exceeding limits, and ensures that the unit maintains excellent control quality across the entire operating range.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113590A_ABST
    Figure CN123113590A_ABST
Patent Text Reader

Abstract

This invention relates to the field of automatic control technology for thermal power generation, and discloses an intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation. The system includes a unit data acquisition module, a preprocessing module, a dynamic boundary calculation module, and a feedback correction module. The data acquisition module collects unit load, main steam pressure, temperature, valve opening, and coal consumption in real time. The preprocessing module removes out-of-limit values ​​from the raw data and performs low-pass filtering. The dynamic boundary calculation module calculates the dynamic safety control boundaries of load, pressure, and temperature in real time based on the preprocessed data. The load boundary is dynamically corrected based on the pressure change rate and combustion stability; the pressure boundary is dynamically corrected based on the pressure change rate and valve opening; and the temperature boundary is dynamically corrected based on the temperature change rate and feedwater flow deviation. The feedback correction module adjusts the correction coefficients online based on the actual control effect. This invention achieves dynamic adaptation of the control boundaries under deep peak shaving conditions in thermal power units.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for thermal power generation, specifically to a deep peak-shaving intelligent control system for thermal power units based on dynamic boundary self-adaptation. Background Technology

[0002] With the rapid growth of installed capacity of new energy power generation such as wind power and photovoltaic power, the power system's requirements for the peak-shaving capacity of thermal power units are increasing. Thermal power units need to frequently participate in deep peak shaving, flexibly adjusting output within a wide load range to mitigate the volatility of new energy power generation such as wind power and photovoltaic power. Through equipment modification and control optimization, the lower limit of power generation load regulation under normal operating conditions of thermal power units can be reduced from the traditional minimum stable combustion load to 30% of rated power or even lower. However, under deep peak shaving conditions, the overall operating characteristics of the unit deviate significantly from the high-load design operating parameters, placing higher demands on the quality of the coordinated control system.

[0003] Currently, the main problems with deep peak shaving control of thermal power units are as follows:

[0004] First, the control boundaries are fixed. Existing control systems often set fixed limits (such as maximum pressure and minimum load) based on design conditions. When a unit operates under deep peak shaving at low load, the actual operating state differs significantly from the design conditions. Fixed boundaries are either too conservative, limiting the unit's regulation capabilities, or too aggressive, easily leading to parameter exceeding limits or even equipment tripping. Studies have indicated that during deep peak shaving, units operating at their stable boundaries are prone to system risks when subjected to significant disturbances.

[0005] Second, the control parameters lack adaptive adjustment. Most existing thermal power unit coordinated control systems adopt traditional boiler-turbine coordinated control strategies, using the same control parameters in both low-load and high-load periods. This results in poor adaptability to changes in coal quality, poor load control performance, and large fluctuations in main steam pressure. During the load variation process of deep peak shaving, the dynamic characteristics of the controlled object will change significantly, and traditional fixed-parameter control strategies are difficult to maintain excellent control quality across the entire operating range.

[0006] Third, the boundary conditions lack dynamic quantification. Existing methods mostly handle the boundary conditions of unit operation statically, failing to dynamically adjust the control boundaries based on the real-time operating status of the unit (such as pressure change rate, temperature change rate, valve opening, feedwater flow deviation, etc.). When boundary conditions change, operators often need to manually modify the control parameters, which is lagging and difficult to guarantee optimality. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a deep peak-shaving intelligent control system for thermal power units based on dynamic boundary adaptation. It features dynamic analytical calculation to correct control boundaries in real time, enabling adaptive adjustment under all operating conditions. This system improves peak-shaving safety while enhancing load response flexibility. It also boasts advantages such as simple structure and ease of engineering implementation, thus solving the aforementioned technical problems.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a deep peak-shaving intelligent control system for thermal power units based on dynamic boundary adaptation, comprising: a unit data acquisition module, a preprocessing module, a dynamic boundary calculation module, and a feedback correction module;

[0009] The unit data acquisition module is used to collect the unit's operating data in real time;

[0010] The preprocessing module is used to preprocess the collected data;

[0011] The dynamic boundary calculation module is used to calculate the safety control boundary in real time based on the preprocessed data, including the upper and lower limits of load, pressure, and temperature.

[0012] The feedback correction module is used to adjust the correction coefficients in the dynamic boundary calculation online based on the actual control effect.

[0013] As a preferred embodiment of the present invention, the operating data includes unit load, main steam pressure, main steam temperature, turbine control valve opening, and coal consumption.

[0014] As a preferred embodiment of the present invention, the data preprocessing includes removing obviously excessive values ​​from the collected raw data and performing first-order low-pass filtering.

[0015] As a preferred embodiment of the present invention, the upper and lower load limits are calculated by the load boundary calculation unit in the dynamic boundary calculation module, and the relevant expressions are as follows: ; ; in, Indicates time The maximum load limit; Indicates time The lower limit of the load; Indicates the rated power of the unit; This indicates the unit's minimum stable combustion load; This represents the load limit correction factor; This represents the load lower limit correction factor.

[0016] As a preferred technical solution of the present invention, the load limit correction coefficient The adjustment is dynamically based on the rate of pressure change, and the adjustment formula is as follows: ; in, Indicates the adjustment factor; This indicates the rate of change of the main steam pressure;

[0017] Lower load limit correction factor When the furnace pressure fluctuation exceeds the set value Increase.

[0018] As a preferred embodiment of the present invention, the upper and lower pressure limits are calculated by the pressure boundary calculation unit in the dynamic boundary calculation module, and the calculation steps are as follows:

[0019] Step A1: Calculate the upper and lower limits of the pressure reference based on the collected data, as shown in the following expression: ; ; in, Indicates current load The upper limit of the pressure benchmark; Indicates current load The lower limit of the pressure reference; This represents the slope coefficient of the upper pressure limit; This represents the upper limit intercept coefficient for pressure. This indicates the current real-time load of the generating unit; This represents the slope coefficient of the lower pressure limit; Indicates the lower limit intercept coefficient of pressure;

[0020] Step A2: Calculate the dynamic pressure correction amount based on the current pressure change rate and valve opening. ;

[0021] Step A3: Calculate the final real-time pressure upper and lower limits. The relevant expressions are as follows: ; ; in, Indicates time Real-time pressure limit; Indicates time The real-time lower limit of pressure.

[0022] As a preferred embodiment of the present invention, the pressure dynamic correction amount in step A2 The expression is as follows: ; in, Represents the dynamic correction amount for the pressure boundary; This represents the correction factor for the rate of pressure change; This indicates the rate of change of the main steam pressure; Indicates the valve opening coefficient; Represents absolute value; This indicates the opening degree of the turbine control valve.

[0023] As a preferred embodiment of the present invention, the upper and lower temperature limits are calculated by the temperature boundary calculation unit in the dynamic boundary calculation module, and the calculation steps are as follows:

[0024] Step B1: Calculate the upper and lower limits of the temperature reference based on the collected data. The relevant expressions are as follows: ; ; in, Indicates current load The upper limit of the temperature reference; Indicates current load The lower limit of the temperature reference; This represents the slope coefficient of the upper temperature limit; Indicates the lower limit slope coefficient; This represents the upper temperature intercept coefficient; Indicates the lower limit intercept coefficient;

[0025] Step B2: Calculate the dynamic correction amount based on the current temperature change rate and water flow deviation. ;

[0026] Step B3: Calculate the final real-time upper and lower temperature limits, as shown in the following expressions: ; ; in, Indicates time The real-time upper limit of temperature; Indicates time The real-time lower limit of temperature.

[0027] As a preferred embodiment of the present invention, the dynamic correction amount in step B2 The expression is as follows: ; in, This represents the correction factor for the rate of temperature change. This indicates the rate of change of the main steam temperature; Indicates deviation in water supply flow rate; This represents the water supply flow correction factor; It represents the absolute value.

[0028] As a preferred technical solution of the present invention, the feedback correction module is used to adjust the correction coefficients in the dynamic boundary calculation online according to the actual control effect. Specifically, if the number of exceedances reaches 3 or more within a fixed time window, the correction coefficients are reduced by 10% of their original values. If there are no exceedances within a fixed time window, the original correction coefficients are maintained.

[0029] Compared with existing technologies, this invention provides a deep peak-shaving intelligent control system for thermal power units based on dynamic boundary adaptation, which has the following beneficial effects:

[0030] 1. This invention, through a dynamic boundary calculation module, directly calculates the safety control boundaries of pressure, temperature, and load using analytical formulas based on real-time unit operating data, overcoming the shortcomings of traditional methods where control boundaries are fixed. Specifically, the pressure boundary is dynamically corrected based on the pressure change rate and valve opening, the temperature boundary is dynamically corrected based on the temperature change rate and feedwater flow deviation, and the load boundary is dynamically adjusted based on the pressure change rate and combustion stability. This ensures that the control boundaries always match the actual operating state of the unit, effectively solving the problem of poor adaptability of static boundaries under deep peak shaving conditions.

[0031] 2. This invention achieves an adaptive protection mechanism where the larger the disturbance, the narrower the boundary, through differentiated bidirectional contraction design of the upper limit correction amount and lower limit correction amount for pressure boundary, and the upper limit correction amount and lower limit correction amount for temperature boundary. When the unit experiences severe operational disturbances, the control boundary automatically narrows, guiding the unit to operate within a safer parameter range. This effectively avoids equipment damage and unplanned shutdowns caused by parameter exceeding limits, thus improving operational safety under deep peak shaving conditions. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0033] 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.

[0034] Please see Figure 1 The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation includes: a unit data acquisition module, a preprocessing module, a dynamic boundary calculation module, and a feedback correction module.

[0035] The unit data acquisition module is used to collect the unit's operating data in real time, including unit load, main steam pressure, main steam temperature, turbine control valve opening, and coal consumption.

[0036] The preprocessing module is used to preprocess the collected data. Data preprocessing includes removing obviously excessive values ​​from the collected raw data and performing first-order low-pass filtering.

[0037] The dynamic boundary calculation module is used to calculate the safety control boundaries in real time based on the preprocessed data, including the upper and lower limits of load, pressure, and temperature.

[0038] The upper and lower limits of the load are calculated by the load boundary calculation unit in the dynamic boundary calculation module, and the relevant expressions are as follows: ; ; in, Indicates time The maximum load limit; Indicates time The lower limit of the load; Indicates the rated power of the unit; This indicates the unit's minimum stable combustion load; This represents the load limit correction factor; This represents the load lower limit correction factor;

[0039] When thermal power units operate under deep peak shaving conditions, their permissible load range is constrained by multiple factors, including boiler combustion stability and equipment strength; rated power It is the upper limit of the output allowed by the equipment design, and the minimum stable combustion load. This is the lower limit of the baseline value for maintaining stable combustion in a boiler without oil-assisted combustion. However, in dynamic processes, when the main steam pressure changes significantly, there is a risk of overpressure if the boiler continues to operate at rated power; when combustion is unstable, there is a risk of flameout if the boiler continues to operate at the minimum stable combustion load. Therefore, this invention introduces a load upper limit correction coefficient. and load lower limit correction factor The basic load boundary is dynamically corrected based on the pressure change rate and combustion stability, so that the load boundary random group can adaptively adjust the real-time operating status.

[0040] Load limit correction factor The adjustment is dynamically based on the rate of pressure change, and the adjustment formula is as follows: ; in, Indicates the adjustment factor; This indicates the rate of change of the main steam pressure;

[0041] Lower load limit correction factor When the furnace pressure fluctuation exceeds the set value Increase;

[0042] During deep peak shaving operation, the main steam pressure change rate of thermal power units It is an important indicator reflecting the operational stability of the unit. The larger its absolute value, the more drastic the change in heat storage on the boiler side, and the closer the unit is to a dangerous state of overpressure or instability. If the rated power is still used as the upper limit of the load at this time, it may lead to further pressure runaway and trigger the safety valve or turbine overspeed protection. Therefore, this invention designs a load upper limit correction coefficient. Its initial value is set to 1.0, and it is made to decrease linearly as the absolute value of the pressure change rate increases, thereby dynamically reducing the upper limit of the load and preventing overpressure accidents under severe disturbances; adjustment coefficient Used to control the correction intensity so that the correction response matches the actual inertial characteristics of the unit;

[0043] The upper and lower pressure limits are calculated by the pressure boundary calculation unit in the dynamic boundary calculation module. The calculation steps are as follows:

[0044] Step A1: Calculate the upper and lower limits of the pressure reference based on the collected data, as shown in the following expression: ; ; in, Indicates current load The upper limit of the pressure benchmark; Indicates current load The lower limit of the pressure reference; This represents the slope coefficient of the upper pressure limit; This represents the upper limit intercept coefficient for pressure. This indicates the current real-time load of the generating unit; This represents the slope coefficient of the lower pressure limit; Indicates the lower limit intercept coefficient of pressure;

[0045] Under steady-state operation, thermal power units exhibit a clear positive correlation between main steam pressure and unit load; that is, the higher the load, the greater the boiler evaporation demand, and the higher the required main steam pressure. Based on this physical characteristic, this invention uses a linear function to model the pressure reference value, and through coefficients... , and , The upper and lower pressure limits are described respectively as they change with load. The coefficients are obtained by least squares regression fitting of the unit's historical operating data, which can accurately reflect the static pressure-load characteristics of the unit under normal steady-state conditions. Using the fitted linear function as a benchmark, the corresponding pressure benchmark value can be directly calculated at any continuous load point, providing a basic reference value for subsequent dynamic correction.

[0046] Step A2: Calculate the dynamic pressure correction amount based on the current pressure change rate and valve opening. Its expression is as follows: ; in, Represents the dynamic correction amount for the pressure boundary; This represents the correction factor for the rate of pressure change; This indicates the rate of change of the main steam pressure; Indicates the valve opening coefficient; Represents absolute value; Indicates the opening degree of the turbine control valve;

[0047] During the operation of thermal power units, the pressure reference value is only meaningful under steady-state conditions. In actual dynamic processes, the main steam pressure will deviate from the reference value due to factors such as load changes and combustion disturbances. In order to accurately characterize the degree of this dynamic deviation and adjust the control boundary accordingly, this invention selects two key variables to form the correction quantity. First, the rate of change of main steam pressure. Its absolute value directly reflects the trend and severity of the current pressure deviating from the steady state; secondly, the turbine control valve opening. Its magnitude represents the degree of influence of the steam turbine on the boiler-side pressure release capability. The larger the valve opening, the more sensitive the dynamic pressure response, and the higher the boundary safety risk. By multiplying both by a correction factor... , The correction amount obtained by superposition It can dynamically track changes in the actual operating status of the unit and realize real-time online correction of the pressure boundary;

[0048] In the above formula, The physical meaning of (dynamic pressure correction) is the magnitude by which the pressure boundary needs to contract relative to the steady-state reference value under current dynamic conditions. A larger value indicates a more severe current operational disturbance, and a greater need to narrow the pressure boundary to ensure safety. The pressure change rate correction factor is used to convert the pressure change rate dimension to the pressure correction dimension, and its magnitude reflects the sensitivity of the system to the pressure change rate. The absolute value of the rate of change of pressure represents the degree of deviation of the pressure from the steady state. Whether the pressure is rising or falling, a rapid change indicates that the system is in an unsteady state. (Regulator opening coefficient) converts the regulator opening dimension into the pressure correction dimension, and its magnitude reflects the degree to which the regulator action consumes the pressure boundary safety margin. (Regulator valve opening) characterizes the impact of the turbine side on the boiler side's heat storage and release capacity; the physical meaning of this formula is: the more severe the pressure disturbance and the larger the regulator valve opening, the greater the dynamic correction amount. The larger the value, the more the pressure boundary shrinks towards the reference value, thus realizing the adaptive protection function of actively reserving a safety margin and preventing parameters from exceeding limits during dynamic processes.

[0049] Step A3: Calculate the final real-time pressure upper and lower limits. The relevant expressions are as follows: ; ; in, Indicates time Real-time pressure limit; Indicates time The real-time lower limit of pressure;

[0050] The pressure reference value reflects the normal operating range of the unit under steady-state conditions, while the dynamic correction amount... This represents the compression requirement of the current dynamic disturbance on the safety boundary; the present invention superimposes the two to form the final real-time pressure boundary; specifically, it adds to the upper limit of the pressure reference. This further raises the upper limit, but it should be noted that this "raising" is relative to the baseline upper limit, in conjunction with step A2. The characteristic that increases with increasing disturbance essentially means that the larger the disturbance, the closer the real-time upper limit becomes to the reference upper limit, or even that it is compressed (i.e., the increment of the effective value of the upper limit relative to the reference value is canceled out by the disturbance), while the lower limit is the difference between the reference lower limit and the lower limit. This design causes the upper and lower boundaries to narrow towards the center, forming an adaptive protection mechanism where the greater the disturbance, the narrower the boundary. Through this design, the real-time pressure boundary can be dynamically adjusted according to the operating status, ensuring that the unit always operates within a safe range.

[0051] The upper and lower temperature limits are calculated by the temperature boundary calculation unit in the dynamic boundary calculation module. The calculation steps are as follows:

[0052] Step B1: Calculate the upper and lower limits of the temperature reference based on the collected data. The relevant expressions are as follows: ; ; in, Indicates current load The upper limit of the temperature reference; Indicates current load The lower limit of the temperature reference; This represents the slope coefficient of the upper temperature limit; Indicates the lower limit slope coefficient; This represents the upper temperature intercept coefficient; Indicates the lower limit intercept coefficient;

[0053] Under steady-state operation, thermal power units exhibit a clear correlation between main steam temperature and unit load. As load increases, boiler combustion intensity rises, and the main steam temperature correspondingly increases to a higher level; conversely, as load decreases, the temperature drops. Based on this physical characteristic, this invention uses a linear function to model the temperature reference value, through coefficients... , and , The upper and lower temperature limits are described respectively as a function of load. The coefficients are obtained by least squares regression fitting of the historical operating data of the unit, which can accurately reflect the static temperature-load characteristics of the unit under normal steady-state conditions. Using the fitted linear function as a benchmark, the corresponding temperature benchmark value can be directly calculated at any continuous load point, providing a basic reference value for subsequent dynamic correction.

[0054] Step B2: Calculate the dynamic correction amount based on the current temperature change rate and water flow deviation. Its expression is as follows: ; in, This represents the correction factor for the rate of temperature change. This indicates the rate of change of the main steam temperature; Indicates deviation in water supply flow rate; This represents the water supply flow correction factor; Represents absolute value;

[0055] The temperature reference value is only meaningful under steady-state conditions. In actual dynamic processes, the main steam temperature will deviate from the reference value due to factors such as load changes, feedwater disturbances, and combustion adjustments. To accurately characterize this dynamic deviation and adjust the temperature control boundary accordingly, this invention selects two key variables to constitute the correction amount. Firstly, the rate of change of main steam temperature. The absolute value of the temperature change directly reflects the trend and severity of the temperature deviation from the steady state. The faster the temperature changes, the more unstable the system is, and the narrower the boundary needs to be. Secondly, the deviation in water flow rate... The absolute value of the deviation reflects the degree to which the feedwater regulation deviates from normal operating conditions. Feedwater flow is the core means of controlling the main steam temperature; the larger the deviation, the more abnormal the current temperature control is. Multiplying both by a correction factor... , The correction amount obtained by superposition It can dynamically track changes in operating temperature; it should be noted that the temperature correction amount... The treatment method in the final boundary calculation is different from that of pressure. The temperature boundary is lowered by removing the correction amount from the upper limit of the reference and the lower limit is raised by adding the correction amount to the lower limit of the reference. That is, the upper and lower boundaries narrow towards the middle at the same time to prevent the superheater tube wall from burning due to excessive temperature or the turbine water hammer caused by excessive temperature.

[0056] Step B3: Calculate the final real-time upper and lower temperature limits, as shown in the following expressions: ; ; in, Indicates time The real-time upper limit of temperature; Indicates time The real-time lower limit of temperature;

[0057] The temperature reference value reflects the normal operating range of the unit under steady-state conditions, while the dynamic correction amount... This characterizes the current dynamic disturbance's compression requirement on the safety boundary; this invention superimposes the two to form the final real-time temperature boundary, and the processing method is the opposite of the pressure boundary: subtracting from the upper limit of the temperature reference. This lowers the upper limit; it adds to the lower limit of the temperature reference. This design raises the lower limit, meaning that both the upper and lower boundaries narrow towards the middle simultaneously. Since excessively high temperatures can cause overheating and burnout of the superheater and reheater tube walls, while excessively low temperatures can lead to serious accidents such as water hammer in the turbine, the bidirectional contraction design of the temperature boundary ensures that the main steam temperature, whether deviating from the reference value upwards or downwards, can be limited to a safer range by narrowing the operating corridor, effectively preventing the two extreme risks of overheating and undercooling.

[0058] The feedback correction module is used to adjust the correction coefficients in the dynamic boundary calculation online according to the actual control effect. Specifically, if the number of exceedances reaches 3 or more within a fixed time window, the correction coefficients will be reduced by 10% of their original values. If there are no exceedances within a fixed time window, the original correction coefficients will be maintained.

[0059] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A deep peak-shaving intelligent control system for thermal power units based on dynamic boundary adaptation, characterized in that: include: The unit consists of a data acquisition module, a preprocessing module, a dynamic boundary calculation module, and a feedback correction module; The unit data acquisition module is used to collect the unit's operating data in real time; The preprocessing module is used to preprocess the collected data; The dynamic boundary calculation module is used to calculate the safety control boundary in real time based on the preprocessed data, including the upper and lower limits of load, pressure, and temperature. The feedback correction module is used to adjust the correction coefficients in the dynamic boundary calculation online based on the actual control effect.

2. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 1, characterized in that: The operating data includes unit load, main steam pressure, main steam temperature, turbine control valve opening, and coal consumption.

3. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 2, characterized in that: The data preprocessing includes removing obviously excessive values ​​from the collected raw data and performing a first-order low-pass filter.

4. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 1, characterized in that: The upper and lower limits of the load are calculated by the load boundary calculation unit in the dynamic boundary calculation module, and the relevant expressions are as follows: ; ; in, Indicates time The maximum load limit; Indicates time The lower limit of the load; Indicates the rated power of the unit; This indicates the unit's minimum stable combustion load; This represents the load limit correction factor; This represents the load lower limit correction factor.

5. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 4, characterized in that: The load limit correction factor The adjustment is dynamically based on the rate of pressure change, and the adjustment formula is as follows: ; in, Indicates the adjustment factor; This indicates the rate of change of the main steam pressure; Lower load limit correction factor When the furnace pressure fluctuation exceeds the set value Increase.

6. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 1, characterized in that: The upper and lower pressure limits are calculated by the pressure boundary calculation unit in the dynamic boundary calculation module. The calculation steps are as follows: Step A1: Calculate the upper and lower limits of the pressure reference based on the collected data, as shown in the following expression: ; ; in, Indicates current load The upper limit of the pressure benchmark; Indicates current load The lower limit of the pressure reference; This represents the slope coefficient of the upper pressure limit; This represents the upper limit intercept coefficient for pressure. This indicates the current real-time load of the generating unit; This represents the slope coefficient of the lower pressure limit; Indicates the lower limit intercept coefficient of pressure; Step A2: Calculate the dynamic pressure correction amount based on the current pressure change rate and valve opening. ; Step A3: Calculate the final real-time pressure upper and lower limits. The relevant expressions are as follows: ; ; in, Indicates time Real-time pressure limit; Indicates time The real-time lower limit of pressure.

7. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 6, characterized in that: The pressure dynamic correction amount in step A2 The expression is as follows: ; in, Represents the dynamic correction amount for the pressure boundary; This represents the pressure change rate correction factor; This indicates the rate of change of the main steam pressure; Indicates the valve opening coefficient; Represents absolute value; This indicates the opening degree of the turbine control valve.

8. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 1, characterized in that: The upper and lower temperature limits are calculated by the temperature boundary calculation unit in the dynamic boundary calculation module. The calculation steps are as follows: Step B1: Calculate the upper and lower limits of the temperature reference based on the collected data. The relevant expressions are as follows: ; ; in, Indicates current load The upper limit of the temperature reference; Indicates current load The lower limit of the temperature reference; This represents the slope coefficient of the upper temperature limit; Indicates the slope coefficient of the lower limit of the temperature; This represents the upper temperature intercept coefficient; Indicates the lower limit intercept coefficient; Step B2: Calculate the dynamic correction amount based on the current temperature change rate and water flow deviation. ; Step B3: Calculate the final real-time upper and lower temperature limits, as shown in the following expressions: ; ; in, Indicates time The real-time upper limit of temperature; Indicates time The real-time lower limit of temperature.

9. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 8, characterized in that: The dynamic correction amount in step B2 The expression is as follows: ; in, This represents the correction factor for the rate of temperature change. This indicates the rate of change of the main steam temperature; Indicates deviation in water supply flow rate; This represents the water supply flow correction factor; It represents the absolute value.

10. The intelligent control system for deep peak shaving of thermal power units based on dynamic boundary adaptation according to claim 1, characterized in that: The feedback correction module is used to adjust the correction coefficients in the dynamic boundary calculation online according to the actual control effect. Specifically, if the number of exceedances reaches 3 or more within a fixed time window, the correction coefficients are reduced by 10% of their original values. If there are no exceedances within a fixed time window, the original correction coefficients are maintained.