Plastic optimal regulation method and system for primary frequency modulation of supercritical unit

Abstract

This invention provides a malleable optimization adjustment method and system for primary frequency regulation of supercritical units, relating to the field of dynamic modeling and control technology for thermal power generating units. The method includes: replacing the pure integral element characterizing energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, outputting boiler-side state variables; calculating the total releaseable gas flow energy storage based on the boiler-side state variables and pre-generated power conversion relationships of each airflow channel of the turbine, combined with real-time operating margins; when a primary frequency regulation event is detected, dividing the frequency regulation process into multiple consecutive stages in time sequence, and within each stage, using the total releaseable gas flow energy storage as a constraint, performing multi-objective rolling optimization on the turbine's main steam valve, reheat valve, and extraction steam valve, outputting airflow distribution control commands. This invention actively shapes the power response curve through model correction and time-sharing optimization, simultaneously improving initial velocity, maximum power, and support time, ensuring safety.

Description

Technical Field

[0001] This invention relates to the field of dynamic modeling and control technology for thermal power generating units, and in particular to a flexible optimization adjustment method and system for primary frequency regulation of supercritical units. Background Technology

[0002] Supercritical once-through boiler units have become the mainstay of thermal power generation due to their high power generation efficiency and excellent peak-shaving performance. Primary frequency regulation, as a crucial function of the unit to automatically and rapidly adjust its output based on grid frequency deviations, directly affects the frequency security of the power system.

[0003] In existing technologies, transient stability calculations of power systems often employ the BPA model framework to simplify the modeling of boiler-turbine dynamic characteristics. For supercritical once-through boilers, traditional models typically equate the steam generation and transport processes to fixed inertia or pure integral elements, lacking a clear mapping relationship with physical variables such as main steam pressure, reheat pressure, and valve opening. This makes it difficult to accurately reflect the pressure-flow dynamics during second-level primary frequency regulation. Furthermore, conventional primary frequency regulation control primarily relies on frequency difference and valve opening, failing to consider the different power contribution delays and overshoot characteristics of the high-pressure cylinder, intermediate-pressure cylinder, and reheat channel. It also fails to fully utilize the short-term transferable steam flow of the extraction steam regeneration system, resulting in a simplistic control strategy.

[0004] Therefore, existing solutions suffer from problems such as insufficient model accuracy and difficulty in balancing regulation speed and support time in primary frequency regulation of supercritical units. Summary of the Invention

[0005] This invention provides a flexible optimization method for primary frequency regulation of supercritical units, which solves the problems of insufficient model accuracy and difficulty in balancing regulation speed and support time in existing schemes for primary frequency regulation of supercritical units.

[0006] This invention provides a flexible optimization method for primary frequency regulation of a supercritical unit. The method includes: replacing the pure integral element characterizing energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, and outputting the boiler-side state variable; calculating the current total energy storage of the releaseable airflow based on the boiler-side state variable and the pre-generated power conversion relationship of each airflow channel of the turbine, combined with the real-time operating margin; when a primary frequency regulation event is detected, dividing the frequency regulation process into multiple consecutive stages with different control objectives in time sequence, and within each stage, using the total energy storage of the releaseable airflow as a constraint, performing multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the turbine, and outputting airflow distribution control commands.

[0007] According to the present invention, a method for plastic optimization adjustment of primary frequency regulation of a supercritical unit is provided. The power conversion relationship of each airflow channel of the pre-generated steam turbine is obtained through the following steps: based on the boiler-side state variables, the power contribution parameters of each airflow channel of the steam turbine are identified; according to the power contribution parameters, a mapping relationship between the flow rate change and power output of each channel is established as the power conversion relationship.

[0008] According to the present invention, a plastic optimization regulation method for primary frequency regulation of a supercritical unit is provided. This method replaces the pure integral element characterizing energy storage characteristics in the dynamic model of a once-through boiler with a first-order inertial element having pressure-flow feedback, outputting boiler-side state variables. The method includes: dividing the once-through boiler into two spatially distributed lumped elements: a steam generation lumped section and a superheated steam transport lumped section; replacing the pure integral element of the original model in each lumped section with a first-order inertial element having pressure-flow feedback, whereby the first-order inertial element characterizes the dynamic coupling relationship between pressure changes and outlet flow changes within the lumped section; and based on this first-order inertial element, outputting the pressure state, pressure change rate, and predicted steam flow rate of each lumped section as the boiler-side state variables.

[0009] According to the present invention, a plastic optimization adjustment method for primary frequency regulation of a supercritical unit replaces the pure integral element of the original model in each lumped section with a first-order inertial element with pressure-flow feedback. This method includes: representing the outlet flow rate of the lumped section as a linear function of the pressure of the lumped section, and closing the pure integral element into the first-order inertial element. The first-order inertial element has a static gain and a time constant. The static gain reflects the strength of the pressure-flow proportional feedback, and the time constant reflects the inertia of steam release and pressure propagation.

[0010] According to the present invention, a method for plastic optimization regulation of primary frequency control of a supercritical unit is provided, which identifies the power contribution parameters of each airflow channel of the turbine based on the boiler-side state variables, including: dividing the controllable objects on the turbine side into high-pressure cylinder channel, reheat and connecting pipe channel, intermediate-pressure cylinder channel, low-pressure cylinder channel and extraction steam channel; and identifying the flow-power coefficient, dynamic overshoot coefficient and inertia constant of each channel as the power contribution parameters.

[0011] According to the present invention, a method for flexible optimization of primary frequency regulation of a supercritical unit is provided. Based on the boiler-side state variables and the pre-generated power conversion relationship of each airflow channel of the turbine, combined with real-time operating margins, the total energy storage of the currently releaseable airflow is calculated. This includes: calculating the boiler-side releaseable steam equivalent based on the boiler-side state variables; calculating the releaseable airflow increments on the main steam pipe and steam chamber sides, the reheat side, and the extraction redistribution side based on the power conversion relationship; subtracting a preset safety reserve from the sum of the boiler-side releaseable steam equivalent, the main steam pipe and steam chamber side releaseable airflow increments, the reheat side releaseable airflow increments, and the extraction redistribution side transferable airflow increments to obtain the total energy storage of the releaseable airflow; wherein the preset safety reserve is determined based on the real-time main steam pressure lower limit margin, the reheat pressure lower limit margin, and the thermal safety margin.

[0012] According to the present invention, a flexible optimization adjustment method for primary frequency regulation of a supercritical unit is provided. The multiple continuous stages with different control objectives include: a fast-segment release stage, a steady-segment diversion stage, a follow-up stage, and a recovery stage. The optimization objectives of the multi-objective rolling optimization include at least one of the following: maximizing the integral power of primary frequency regulation, maximizing the initial regulation speed, maximizing the maximum power increment, and maximizing the power support time.

[0013] According to the present invention, a method for flexible optimization of primary frequency regulation of a supercritical unit is provided. The method further includes: in the fast release phase, with the optimization objective of maximizing the initial regulation speed, controlling the main steam valve to open rapidly at the allowable rate, and prioritizing the release of available airflow in the high-pressure cylinder front chamber and the main steam pipe; in the steady-state diversion phase, with the optimization objective of increasing the maximum power increment and integral power, coordinating the reheat valve opening and temporarily reducing part of the regenerative extraction steam to maintain the power platform; in the follow-up phase, with the optimization objective of extending the power support time and ensuring pressure safety, dynamically adjusting the valve opening according to the boiler combustion response, and gradually replenishing the extraction steam; and in the recovery phase, with the optimization objective of undisturbed recovery, smoothly restoring to steady-state conditions according to the pressure recovery trend.

[0014] According to the present invention, a method for flexible optimization of primary frequency regulation of a supercritical unit includes a recovery phase in which the undisturbed recovery is optimized to smoothly restore the unit to steady-state conditions according to the pressure recovery trend. The method includes: in the recovery phase, based on the energy deficit and the degree of pressure recovery, the opening of the main steam valve, the reheat valve, and the extraction steam valve, as well as the extraction steam quantity corresponding to the extraction steam valve, are smoothly restored to gradually restore the valve opening and extraction steam quantity to the values ​​corresponding to the steady-state conditions.

[0015] According to the present invention, a plastic optimization method for primary frequency regulation of a supercritical unit is provided. The multi-objective rolling optimization is defined by an objective function and a set of constraints. The objective function includes a weighted combination of at least two of the following: primary frequency regulation integral energy term, initial regulation speed term, maximum power increment term, and power support time term; and a penalty term for penalizing behaviors that violate the constraints. The constraints include at least one of the following: main steam pressure constraint, reheat pressure constraint, valve opening and action rate constraint, flow rate constraint, and thermal stress constraint.

[0016] According to the present invention, a flexible optimization adjustment method for primary frequency regulation of a supercritical unit is provided. The method further includes: when the absolute value of the frequency difference is greater than a preset frequency difference threshold and the system frequency drops rapidly, increasing the weights corresponding to the initial adjustment speed and the maximum power increment; when the main steam pressure margin is less than a preset pressure threshold, increasing the weights corresponding to the power support time and pressure safety, and decreasing the weights corresponding to the maximum power increment; when the extraction steam system is constrained by heating, freezing the optimization variables of the restricted extraction stage and increasing the compensation weights of other adjustable channels.

[0017] According to the present invention, a plastic optimization adjustment method for primary frequency regulation of a supercritical unit is provided. The method further includes: establishing initial values ​​of boiler model parameters and turbine power contribution parameters under different load ranges using historical disturbance data and primary frequency regulation test data of the unit; updating the boiler model parameters and turbine power contribution parameters during operation using natural frequency disturbances or small-amplitude adjustment actions as excitations; calculating the model reliability index based on the updated parameters; and switching to a primary frequency regulation control mode that does not include the multi-objective rolling optimization when the model reliability index is lower than a preset reliability threshold.

[0018] According to the present invention, a method for plastic optimization adjustment of primary frequency regulation of a supercritical unit is provided. The method further includes: superimposing the airflow distribution control command with the original primary frequency regulation command to generate a superimposed command; sending the superimposed command to a digital electro-hydraulic control system, which processes the command according to preset amplitude limiting parameters, rate limiting values ​​and protection interlocking logic, and outputting the processed command to the corresponding actuator.

[0019] According to the present invention, a method for plastic optimization adjustment of primary frequency regulation of a supercritical unit is provided. The method further includes: before each execution of the multi-objective rolling optimization, obtaining boiler pressure constraint signal, boiler temperature constraint signal and combustion response constraint signal from a digital distributed control system; and using the boiler pressure constraint signal, the boiler temperature constraint signal and the combustion response constraint signal as additional constraints for the multi-objective rolling optimization.

[0020] According to the present invention, a method for flexible optimization of primary frequency regulation of a supercritical unit includes an output airflow distribution control command comprising: a main steam valve opening increment command, a reheat valve opening increment command, stage extraction steam valve opening increment commands, and an optional bypass valve opening increment command. The method further comprises: executing the airflow distribution control command under preset execution conditions; wherein the preset execution conditions include: the main steam pressure and reheat pressure are respectively within their respective preset allowable pressure ranges; the operating rate of each valve is less than or equal to the preset maximum allowable rate; the flow rate of each channel is within the preset allowable flow rate range; and the thermal stress and turbine vibration amplitude are respectively less than or equal to their corresponding safety thresholds.

[0021] This invention also provides a flexible optimization control system for primary frequency regulation of a supercritical unit. The system includes: a model correction module, used to replace the pure integral element representing energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, and output boiler-side state variables; an energy storage assessment module, connected to the model correction module, used to calculate the total amount of releaseable airflow energy storage based on the boiler-side state variables and the pre-generated power conversion relationship of each airflow channel of the turbine, combined with real-time operating margin; and a multi-objective optimization module, connected to the energy storage assessment module, used to divide the frequency regulation process into multiple consecutive stages with different control objectives in time sequence when a primary frequency regulation event is detected, and within each stage, using the total amount of releaseable airflow energy storage as a constraint, perform multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the turbine, and output airflow distribution control commands.

[0022] The present invention also provides an electronic device, comprising: at least one processor; at least one memory coupled to the processor, the memory storing at least one program instruction or code, the at least one program instruction or code being loaded and executed by the processor to enable the device to implement the plastic optimization regulation method for primary frequency regulation of a supercritical unit as described above.

[0023] The present invention also provides a computer-readable storage medium storing at least one program instruction or code, which, when loaded and executed by a processor, enables the processor to implement the plastic optimization regulation method for primary frequency regulation of a supercritical unit as described above.

[0024] The present invention provides a flexible optimization adjustment method and system for primary frequency regulation of supercritical units. By replacing the pure integral element in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, a corrected model is established that can accurately reflect the second-level steam flow inertia. Based on this, the total amount of releaseable gas flow energy is calculated according to the boiler-side state variables output by the corrected model, combined with the pre-generated power conversion relationship and real-time operating margin. This transforms the frequency regulation capability, which originally relied on experience-based judgment, into a resource boundary that can be quantified online, ensuring that the optimization decision does not exceed the actual steam supply capacity of the boiler. Furthermore, when a primary frequency regulation event is detected, the frequency regulation process is divided into multiple continuous stages with different control objectives. In each stage, the main steam valve, reheat valve, and extraction steam valve are subjected to multi-objective rolling optimization with the total amount of releaseable gas flow energy as a constraint. This allows different stages to adopt matching control strategies, ultimately outputting a gas flow distribution control command that balances rapid response and continuous support capability. This effectively improves the adjustment speed and power support time of primary frequency regulation of supercritical units and ensures the safe operation of the unit. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0026] Figure 1 This is a schematic flowchart of a flexible optimization adjustment method for primary frequency regulation of a supercritical unit provided in an embodiment of the present invention.

[0027] Figure 2 This is a block diagram of a BPA boiler model integrator modified to first-order inertia, provided in an embodiment of the present invention.

[0028] Figure 3 This is a schematic diagram of a turbine airflow plastic optimization control structure provided in an embodiment of the present invention.

[0029] Figure 4 This is a schematic diagram of a primary frequency modulation response curve shaping index provided in an embodiment of the present invention.

[0030] Figure 5 This is a schematic diagram of the architecture of a flexible optimization regulation system for primary frequency regulation of a supercritical unit provided in an embodiment of the present invention.

[0031] Figure 6 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0033] The following is combined Figure 1 This invention describes a flexible optimization method for primary frequency regulation of supercritical units. For consistency, the entity executing this method will be uniformly referred to as the "system," and will not be described further thereafter.

[0034] Figure 1 This is a flowchart illustrating the plastic optimization adjustment method for primary frequency regulation of supercritical units provided in this embodiment of the invention, as shown below. Figure 1 As shown, the method includes the following: S101. Replace the pure integral element representing the energy storage characteristics in the dynamic model of the DC boiler with a first-order inertial element with pressure-flow feedback, and output the boiler-side state variables.

[0035] In some embodiments, a once-through boiler can be divided into two spatially distributed lumped sections: a steam generation lumped section and a superheated steam transport lumped section.

[0036] For example, the steam generation lumped section characterizes the process by which the working fluid near the economizer, water-cooled walls, and critical region absorbs heat to form usable steam equivalent; the superheated steam transport lumped section characterizes the process by which the superheater, main steam pipe, and pre-valve volume transport steam to the turbine regulating valve. Both lumped sections employ first-order inertial elements, and their time constants and gains can be segmented and corrected according to load, pressure, fuel quantity, feedwater flow rate, main steam temperature, and valve opening.

[0037] like Figure 2 The diagram shown is a modified block diagram of the boiler model integrator in the Bonneville Power Administration (BPA) system, which is commonly used in power system stability calculations according to an embodiment of the present invention, after being changed to first-order inertia. Figure 2 The upper part shows the integrator expression of the original boiler model of BPA: after subtracting the boiler inlet steam equivalent or heat input disturbance from the outlet flow disturbance, the pressure state is formed through a pure integration process; Figure 2 The lower part illustrates the first-order inertial correction of the present invention: the outlet flow disturbance is written as a pressure feedback term, thereby closing the integrator into a first-order inertial element with a defined static gain and time constant. The corrected BPA once-through boiler two-section lumped model includes lumped section 1 (water-cooled wall / evaporator section) and lumped section 2 (superheater / main steam section), with the two lumped sections forming a closed-loop coupling through inter-section flow feedback.

[0038] Specifically, the pure integral element of the original model in each lumped segment is replaced with a first-order inertial element with pressure-flow feedback.

[0039] The first-order inertial element is used to characterize the dynamic coupling relationship between pressure changes and outlet flow changes within the lumped section.

[0040] Alternatively, the pure integral element can be closed into a first-order inertial element by expressing the outlet flow rate of the lumped section as a linear function of the pressure of the lumped section.

[0041] In this embodiment of the invention, the first-order inertial element has a static gain and a time constant. The static gain reflects the strength of the pressure-flow ratio feedback, and the time constant reflects the inertia of steam release and pressure propagation.

[0042] For example, the expression for the pure integral energy storage element in the i-th lumped segment of the original BPA boiler model is as follows: .

[0043] in, Let be the pressure deviation of the i-th boiler lumped section. This refers to the equivalent pressure capacity or energy storage coefficient. The equivalent steam or heat input flow rate deviation entering this section is calculated as follows: This refers to the deviation in steam flow rate from this section.

[0044] Considering the approximately linear proportional relationship between the outlet flow rate and the pressure within the lumped section of a once-through boiler, let: .

[0045] in, Let be the pressure-flow feedback coefficient. Substituting the above equation into the original integrator expression, we obtain the modified first-order inertial element transfer function: .

[0046] in, For the static gain of a first-order inertial element, The time constant of a first-order inertial element. Static gain. The time constant reflects the strength of the pressure-flow ratio feedback. It reflects the dynamic inertia determined by the steam release, pressure propagation, and pipeline volume of the boiler's central section.

[0047] Specifically, the dynamic equations for the two-stage once-through boiler can be expressed as: ; .

[0048] in, This refers to the pressure deviation in the steam generation lumped section. The equivalent steam deviation for entering the steam generation lumped section. , These are the static gain and time constant of the first-order inertial element in the steam generation lumped section, respectively. For the pressure deviation of the superheated steam transport lumened section, This is the conversion factor for the pressure in the steam generation lumped section to the inlet flow rate in the superheated steam transport lumped section. The deviation of the required flow rate before the turbine valve. , These are the static gain and time constant of the first-order inertial element in the superheated steam transport lumped section, respectively.

[0049] The model output includes main steam pressure. Main steam flow Pressure change rate and boiler-side release steam equivalent .

[0050] In Digital Electro-Hydraulic Control Systems (DEH), Distributed Control Systems (DCS), or BPA user-defined models, control cycles can be configured. Discretize the above continuous model: .

[0051] in, To control the cycle number, These are the discretization coefficients. , These are the discretization coefficients for the corresponding input items. When pressure, valve position, or load crosses a preset range, online switching or interpolation updates are performed. , and .

[0052] Furthermore, the linearized flow rate of the main steam before entering the turbine regulating valve can be expressed as: .

[0053] in, The equivalent opening of the main steam regulating valve. For pressure-flow gain, For valve flow gain, This is a temperature / specific volume correction term. This relationship couples the boiler pressure state with the turbine valve control variables in the same optimization problem.

[0054] Thus, by dividing the steam generation and transportation process of a DC boiler into two lumped models, this invention replaces the original single integral energy storage stage, accurately reflecting the second-level pressure-flow dynamics, and avoiding the overestimation of support capacity caused by the lack of feedback on pressure-flow dynamics, flow resistance, and pipeline volume in a pure integrator.

[0055] Furthermore, based on the first-order inertial element, the pressure state, pressure change rate, and steam flow prediction values ​​of each lumped section are output as boiler-side state variables.

[0056] For example, boiler-side state parameters include the steam generation lumped section pressure. Pressure of the superheated steam transport assembly section (i.e., main steam pressure) Main steam pressure change rate Predicted main steam flow rate and boiler-side release steam equivalent .

[0057] Specifically, the steam equivalent that can be released on the boiler side It is calculated based on the current main steam pressure, main steam flow rate and boiler energy storage coefficient, reflecting the steam energy that can be released in a short time within the safety constraints of the boiler side.

[0058] Thus, by establishing a modified boiler model based on a first-order inertial pressure-flow link, this invention achieves interpretable modeling of the second-level steam flow characteristics of a DC boiler, providing an accurate model basis for subsequent primary frequency regulation capability assessment and optimized control.

[0059] S102. Based on the boiler-side state variables and the pre-generated power conversion relationships of each airflow channel of the turbine, combined with the real-time operating margin, calculate the total amount of energy that can be released by the airflow.

[0060] Optionally, the power conversion relationships of each airflow channel of the steam turbine can be generated in advance.

[0061] In some embodiments, the power contribution parameters of each airflow channel of the steam turbine can be identified based on boiler-side state variables.

[0062] Optionally, the controllable objects on the turbine side can be divided into high-pressure cylinder passage, reheat and connecting pipe passage, intermediate-pressure cylinder passage, low-pressure cylinder passage and extraction steam passage.

[0063] For example, the controllable objects on the turbine side also include the main steam regulating valve GV, the high-pressure cylinder front steam chamber, the cold reheat / hot reheat passage from the high-pressure cylinder exhaust to the reheater, the intermediate-pressure cylinder valve group, the low-pressure cylinder inlet passage, the reheat extraction valves at each stage, and optional bypass or make-up steam valves.

[0064] Since different airflow channels have different time constants and power contributions, primary frequency control cannot simply use a uniform valve opening command; instead, it should address the issue of multi-channel airflow distribution.

[0065] Furthermore, the flow-power coefficient, dynamic overshoot coefficient, and inertia constant of each channel are identified as power contribution parameters.

[0066] For example, the power contribution parameters to be identified include: High-pressure cylinder channel: Flow-power coefficient Overadjustment coefficient Steam chamber inertial constant ; Reheat and connecting pipe channels: Inertial constant of reheat pipe Inertial constant of connecting pipes ; Intermediate pressure cylinder passage: Flow rate-power coefficient Overadjustment coefficient ; Low-pressure cylinder passage: Flow rate-power coefficient Overadjustment coefficient Inertial constant of the low-pressure cylinder inlet passage ; Steam extraction channel: Steam extraction flow rate - power coefficient Inertial constant of the extraction steam passage.

[0067] Specifically, parameter identification employs a combination of offline identification and online updating: Offline phase: Using historical disturbance data of the unit, primary frequency regulation test data, small disturbance test data of Automatic Generation Control (AGC), and thermodynamic calculation results, initial parameter values ​​for each load range are established.

[0068] Among them, the parameters of the two boiler stages were identified using a step test and a recursive least squares method; the parameters of the high-pressure cylinder channel were identified using a small valve disturbance and a constrained least squares method; the parameters of the reheat / medium-pressure channel were identified using a frequency domain fitting and a time domain verification method; and the parameters of the low-pressure / extraction steam channel were identified using a thermal balance model and a historical disturbance method.

[0069] Online phase: Without affecting unit safety, key parameters are updated using recursive least squares with a forgetting factor, with natural frequency disturbances and small-scale adjustment actions as excitations.

[0070] The forgetting factor is adaptively adjusted according to the stability of the unit's operating conditions. When the unit is in a stable operating state, a smaller forgetting factor is used to maintain parameter stability, while when the unit is in a state of large-scale adjustment, a larger forgetting factor is used to accelerate parameter tracking.

[0071] Verification phase: Verify the power peak, initial slope, and support time error using an independent primary frequency modulation event or BPA simulation condition; if the error exceeds the threshold, trigger parameter re-identification or order reduction control.

[0072] Thus, by identifying turbine power contribution parameters through different channels, this invention can accurately distinguish the differences in the contribution of different airflow channels to the power curve, providing a quantitative basis for multi-channel airflow distribution.

[0073] Furthermore, based on the power contribution parameters, a mapping relationship between the flow rate change and power output of each channel is established as a power conversion relationship.

[0074] For example, in a time window Within the turbine, the maximum supported power increment can be obtained by superimposing the power contributions of each channel: .

[0075] in, This is the increase in the inlet flow rate of the high-pressure cylinder. This is the increase in inlet flow rate of the intermediate pressure cylinder. This is the increase in inlet flow rate of the low-pressure cylinder. This refers to the equivalent flow increment transferred to the main flow direction after reducing or changing the extraction steam distribution. This includes power reduction due to rapid pressure drop, valve saturation, and thermal stress limiting. If the available airflow energy storage is insufficient to support a given target, the controller automatically lowers the peak target and extends the plateau to avoid excessive release in the early stages.

[0076] Thus, by establishing a power mapping relationship for multi-stage airflow channels, this invention enables the quantifiable calculation of the power contribution on the turbine side, laying the foundation for subsequent evaluation of available airflow energy storage.

[0077] In some embodiments, the boiler-side releaseable steam equivalent can be calculated based on boiler-side state quantities.

[0078] For example, the boiler side can release steam equivalent The energy storage coefficient is calculated based on the main steam pressure, main steam flow rate, and boiler energy storage coefficient output from the corrected boiler model, reflecting the releaseable steam energy stored in the boiler heating surface and steam pipes.

[0079] Specifically, the calculation of the boiler-side release steam equivalent takes into account the dynamic relationship between pressure and flow rate changes in the steam generation lumped section and the superheated steam transport lumped section, ensuring that the calculation results conform to the actual steam flow characteristics.

[0080] In other embodiments, based on the power conversion relationship, the releaseable airflow increment on the main steam pipe and steam chamber side, the releaseable airflow increment on the reheat side, and the transferable airflow increment on the extraction and redistribution side are calculated respectively.

[0081] For example, an incremental airflow can be released on the main steam pipe and steam chamber side. Calculated based on main steam pressure margin, main steam pipe volume, high-pressure cylinder front chamber volume, and high-pressure cylinder flow-power coefficient; the reheat side can release an incremental airflow. The incremental transferable airflow on the extraction steam redistribution side is calculated based on the reheat pressure margin, reheater volume, hot reheat pipe volume, and intermediate pressure cylinder flow-power coefficient. It is calculated based on the extraction steam flow rate, extraction steam pressure, and extraction steam flow rate-power coefficient at each stage.

[0082] Specifically, the transferable airflow increment on the extraction redistribution side refers to the steam energy that can be transferred to the main flow of the turbine for power generation by temporarily reducing a portion of the regenerated extraction steam, without affecting the unit's safety and heating demand.

[0083] Furthermore, the total energy storage of the releaseable airflow is obtained by subtracting the preset safety reserve from the sum of the boiler-side releaseable steam equivalent, the releaseable airflow increment on the main steam pipe and steam chamber side, the releaseable airflow increment on the reheat side, and the transferable airflow increment on the extraction and redistribution side.

[0084] The preset safety reserve is determined based on the real-time main steam pressure lower limit margin, reheat pressure lower limit margin, and thermal safety margin.

[0085] For example, the formula for calculating the total amount of energy that can be released by the airflow is: .

[0086] in, To maintain a safety margin, this amount is not simply thermal energy storage, but rather the equivalent of airflow power gain that can be converted into turbine shaft power within a given time window.

[0087] In this embodiment of the invention, a primary frequency modulation capability state vector is also constructed. Used to describe the plastic boundary of the unit at the current load point online: .

[0088] in, Main steam pressure The rate of change of main steam pressure. Main steam flow rate. For reheat pressure, This represents the rate of change of reheat pressure. For reheat flow, The opening degree of the main steam regulating valve. For the opening degree of the reheat / medium pressure regulating valve, The opening degree of the extraction steam regulating valve. , , These are the flow-power coefficients for the high-pressure cylinder, medium-pressure cylinder, and low-pressure cylinder, respectively. For energy storage of usable airflow, This is the safety margin vector. It includes at least the lower limit margin of main steam pressure, the lower limit margin of reheat pressure, the turbine shaft vibration margin, the valve limit margin, the valve rate margin, the superheat margin, the boiler combustion response margin, and the thermal stress margin.

[0089] Thus, by comprehensively calculating the available airflow energy storage on the boiler side and the turbine side, and constructing a primary frequency regulation capability state vector, this invention transforms the frequency regulation capability, which originally relied on experience-based judgment, into a resource boundary that can be quantified online, ensuring that subsequent optimization decisions do not exceed the actual steam supply capacity and safety boundaries of the unit.

[0090] S103. When a frequency modulation event is detected, the frequency modulation process is divided into multiple consecutive stages with different control objectives in terms of timing.

[0091] In this embodiment of the invention, primary frequency regulation refers to the differential adjustment process in which the unit automatically changes its output according to the local frequency deviation. Its key indicators are response speed, peak power, duration, and integral power.

[0092] For example, the triggering condition for a frequency modulation event is: the absolute value of the frequency deviation exceeds the frequency modulation dead zone and the duration exceeds the preset criterion; the exit condition for a frequency modulation event is: the frequency returns to the normal range, the priority of the scheduling instruction changes, or any preset safety constraint condition is triggered.

[0093] like Figure 3 The diagram shown is a schematic of a turbine airflow plastic optimization control structure provided in an embodiment of the present invention. The controller receives the frequency difference. Power deviation Inputs include main steam / reheat pressure and flow rate, valve position and speed, thermal stress, and vibration boundary conditions. The system calculates the available margins for the high-pressure cylinder, intermediate-pressure cylinder, low-pressure cylinder, and extraction steam system, and outputs airflow distribution commands. .

[0094] This instruction includes the main steam valve opening speed, the reheat / intermediate pressure valve coordination speed, the short-term reduction in extraction steam, and the recovery speed during the recovery phase. The control concept is: the fast phase releases the available airflow in the steam chamber and reheat pipes, the steady phase coordinates the extraction steam recovery flow, and the subsequent phase is controlled by a closed-loop relay constrained by boiler pressure and flow.

[0095] In this embodiment of the invention, multiple consecutive stages with different control objectives include a fast-segment release stage, a steady-segment diversion stage, a follow-up stage, and a recovery and recycling stage.

[0096] In some embodiments, during the rapid release phase, with the goal of maximizing the initial adjustment speed, the main steam valve is controlled to open rapidly at an allowable rate, prioritizing the release of available airflow in the high-pressure cylinder front chamber and the main steam pipe.

[0097] For example, the typical time window for the rapid release phase is 0 to 3 / 5 seconds. The main control actions are: the main steam regulating valve opens rapidly at its maximum speed; the high-pressure cylinder chamber and main steam pipe release the available airflow preferentially; and the reheat valve remains preset or follows slightly. The control objective is to maximize the initial regulating speed. And to prevent the main steam pressure from falling below the predicted lower limit.

[0098] Specifically, the opening increment of the main steam regulating valve is constrained by the lower limit of the main steam pressure predicted by the corrected boiler model, to prevent the main steam pressure from collapsing rapidly and causing insufficient subsequent power support.

[0099] In some embodiments, during the steady-state shunting phase, with the optimization goal of increasing the maximum power increment and integral power, the opening of the reheat valve is coordinated and a portion of the regenerative steam extraction is reduced for a short period of time to maintain the power plateau.

[0100] For example, the typical time window for the steady-state diversion phase is 3 / 5 to 15 / 25 seconds. The main control actions are: coordinating the opening of the reheat / intermediate-pressure valves to form a platform using the reheat pipe and intermediate-pressure cylinder passage; and implementing short-term reduction of extraction for some of the regenerated steam. The control objective is to increase the maximum power increment. And frequency-modulated integral power, suppressing the rapid decline after the initial peak.

[0101] Specifically, the platform support capacity of the intermediate-pressure cylinder is calculated based on the reheat pressure and the volume of the reheat pipe, and the opening increment of the reheat / intermediate-pressure regulating valve is determined. At the same time, the first to third stage of reheat extraction steam is selected for short-term reduction of extraction, and the steam that originally entered the reheat system is transferred to the intermediate and low-pressure cylinder to do work, thereby increasing the total power output.

[0102] In some embodiments, during the follow-up phase, with the optimization goal of extending the power support time and ensuring pressure safety, the valve opening is dynamically adjusted according to the boiler combustion response to gradually replenish the extracted steam.

[0103] For example, the typical time window for the follow-up phase is 15 / 25 to 60 seconds, and the main control actions are: boiler combustion / feedwater response follow-up; valve adjustment according to the pressure recovery trend; and gradual steam extraction to replenish the boiler. The control objective is to maximize the power support time. This ensures that the power support time matches the boiler's sustainable steam supply capacity.

[0104] Specifically, once the boiler combustion and feedwater response begin to function, the opening increments of the main steam valve and reheat valve are gradually reduced, while the extraction steam flow is gradually replenished to ensure a smooth transition of power to a level where the boiler can sustainably supply steam, thus preventing a sudden collapse of the power platform.

[0105] In some embodiments, during the recovery phase, with the optimization objective of non-disruptive recovery, the system smoothly recovers to steady-state conditions according to the pressure recovery trend.

[0106] Optionally, during the recovery phase, based on the energy deficit and pressure recovery level, the opening degree of the main steam valve, reheat valve, and extraction steam valve, as well as the extraction steam volume corresponding to the extraction steam valve, are smoothly recovered so that the valve opening degree and extraction steam volume gradually return to the values ​​corresponding to the steady-state operating conditions.

[0107] For example, the recovery phase starts after the frequency difference is restored. The main control actions are: to perform uninterrupted recovery according to the pressure, temperature, and extraction system status to avoid secondary drops and valve repetition. The control objective is to restore steady-state efficiency and eliminate the gas flow energy deficit.

[0108] Specifically, energy deficit refers to the difference between the total energy stored in the gas flow released during a single frequency regulation process and the steam energy replenished by the boiler. Based on the energy deficit and the main steam pressure recovery rate, the recovery rate of each valve is determined to ensure a smooth recovery process and avoid secondary power drops.

[0109] Thus, by dividing the primary frequency modulation process into four consecutive stages with different control objectives and adopting a matching airflow distribution strategy in each stage, the present invention achieves active shaping of the primary frequency modulation power response curve, while taking into account the initial adjustment speed, maximum power increment and power support time.

[0110] S104. Within each stage, with the total amount of energy stored in the releasable airflow as a constraint, perform multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the steam turbine, and output airflow distribution control commands.

[0111] In this embodiment of the invention, the optimization objective of the multi-objective rolling optimization includes at least one of the following: maximizing the integral power of the primary frequency regulation, maximizing the initial adjustment speed, maximizing the maximum power increment, and maximizing the power support time.

[0112] For example, maximizing the integral power of primary frequency regulation means maximizing the integral value of the frequency regulation power output of the unit over time within the assessment window; maximizing the initial adjustment speed means maximizing the rate of increase of the unit power at the beginning of the primary frequency regulation action; maximizing the maximum power increment means maximizing the maximum frequency regulation power increment that the unit can achieve within the safety constraints; and maximizing the power support time means maintaining the unit power above the maximum power increment of the set proportion for the longest time.

[0113] Optionally, multi-objective rolling optimization is defined by an objective function and a set of constraints.

[0114] In this embodiment of the invention, the objective function includes a weighted combination of at least two of the following: primary frequency modulation integral energy term, initial adjustment speed term, maximum power increment term, and power support time term, as well as a penalty term for penalizing behaviors that violate the constraints.

[0115] For example, in the prediction time domain Internally, the controller performs rolling optimization using the following scalar objective function: .

[0116] in, , , , , , , These are the weighting coefficients; Penalties for exceeding or approaching the main steam / reheat pressure limit; For valve actuation amplitude, rate, and saturation penalty; The penalties include thermal stress, temperature difference, turbine vibration, and safety margin. Alternatively, these objectives can be solved using Pareto multi-objective optimization, selecting online the solution that satisfies the power grid assessment weights.

[0117] In this embodiment of the invention, the constraints include at least one of the following: main steam pressure constraint, reheat pressure constraint, valve opening and operating rate constraint, flow rate constraint, and thermal stress constraint.

[0118] Specifically, the constraints include the following categories: 1) Pressure constraint: ; It is used to prevent the main steam and reheat pressure from rapidly collapsing or becoming overpressured. 2) Pressure change rate constraint: ; Used to limit boiler and reheat system shocks; 3) Valve constraints: ; ; It is used to meet the limit, speed limit and wear requirements of DEH actuators; 4) Flow constraints: This is used to prevent abnormal flow, deterioration of final stage humidity, or instability of the extraction system. 5) Thermal safety constraints: ; , Within the permissible zone, it is used to protect the heated surfaces and the turbine body; 6) Performance evaluation constraints: ; It is used to meet the primary frequency regulation response and support assessment.

[0119] like Figure 4 The figure shows a schematic diagram of the primary frequency regulation response curve shaping index provided in an embodiment of the present invention. The figure illustrates a comparison of the typical normalized response of a conventional valve's primary frequency regulation and the airflow shaping control of the present invention. The primary frequency regulation response index defined in this invention includes: Initial adjustment speed The average or maximum power increment from the start of a frequency modulation operation until the set proportional power increment is reached. This is used to characterize the unit's ability to quickly support frequency drops; Maximum available power The maximum power increment achievable under safety constraints is determined not only by the valve position, but also by the main steam pressure, reheat pressure, extraction steam margin, and boiler continuous supply capacity. Power support time Power reached After that, it should not be lower than Time, The value can be 0.8 to 0.95, and can be set according to the power grid assessment rules; Frequency modulation integral power Inside the assessment window This reflects the total contribution of primary frequency modulation to system frequency recovery.

[0120] In one alternative implementation, when the absolute value of the frequency difference is greater than a preset frequency difference threshold and the system frequency drops rapidly, the weights corresponding to the initial adjustment speed and the maximum power increment are increased.

[0121] Among them, the absolute value of frequency difference refers to the absolute value of the difference between the actual frequency and the rated frequency of the system; the system frequency refers to the common frequency of the power system.

[0122] For example, when the absolute value of the frequency difference is greater than 0.1 Hz and the frequency drop rate is greater than 0.05 Hz / s, and The weighting is increased by 20% to 50%, prioritizing rapid unit response and suppressing further frequency decline.

[0123] Specifically, the weight adjustment range is adaptively determined based on the magnitude of the frequency difference and the frequency drop rate. The larger the frequency difference and the faster the frequency drop rate, the higher the weight of the initial adjustment speed and the maximum power increment.

[0124] In another alternative implementation, when the main steam pressure margin is less than a preset pressure threshold, the weights corresponding to power support time and pressure safety are increased, while the weights corresponding to the maximum power increment are decreased.

[0125] Among them, the main steam pressure margin refers to the difference between the current main steam pressure and the lower limit of the main steam pressure.

[0126] For example, when the main steam pressure margin is less than 5% of the rated pressure, and Increase the weight by 30% to 60%, and at the same time... The weight is reduced by 20% to 40%, sacrificing some peak power in exchange for a longer power support time, thus avoiding a rapid collapse of the main steam pressure.

[0127] Specifically, the weight adjustment range is adaptively determined based on the size of the main steam pressure margin. The smaller the pressure margin, the higher the weight of power support time and pressure safety, and the lower the weight of the maximum power increment.

[0128] In another alternative implementation, when the extraction steam system is constrained by heating, the optimization variables of the restricted extraction steam stage are frozen, and the compensation weights of other adjustable channels are increased.

[0129] For example, when a stage of extraction steam is used for heating and the heating flow rate cannot be reduced, the opening increment of the extraction steam regulating valve of that stage is frozen from the optimization variables, while the weights of the reheat / intermediate pressure channel and other adjustable extraction steam stages are increased, and these channels compensate for the power increment that the restricted extraction steam stages cannot provide.

[0130] Specifically, the compensation weight is determined based on the power contribution ratio of the restricted extraction stage to ensure that the total power increment is not significantly affected.

[0131] Thus, by adaptively adjusting the weight coefficients of multi-objective optimization, the present invention enables the control strategy to be dynamically optimized according to the system frequency state, unit operating state and external constraints, thereby achieving the optimal frequency regulation effect under different operating conditions.

[0132] In this embodiment of the invention, the output airflow distribution control commands include: main steam valve opening increment commands, reheat valve opening increment commands, each stage extraction steam valve opening increment commands, and optional bypass valve opening increment commands.

[0133] For example, the control variable is defined as: These represent the increments of the main steam valve, intermediate pressure / reheat valve, reheat regulation, extraction steam valves at each stage, and optional bypass / steam replenishment valve, respectively. Within each control cycle, the controller predicts future... seconds , and Trajectory selection that optimizes the comprehensive objective function. .

[0134] Specifically, the optional bypass valve includes at least one of a high-pressure bypass valve and a low-pressure bypass valve.

[0135] For example, when the main steam pressure margin is sufficient and the available airflow energy storage is ample, the opening degree of the high-pressure bypass valve can be incorporated as an auxiliary adjustment method into the optimization variable to further improve the initial adjustment speed.

[0136] For example, when the reheat pressure margin is sufficient and the flow through the medium and low pressure cylinders needs to be increased, the opening degree of the low pressure bypass valve can be incorporated as an auxiliary adjustment method into the optimization variable to extend the power support time.

[0137] Optionally, if preset execution conditions are met, the airflow distribution control command is executed.

[0138] The preset execution conditions include: the main steam pressure and reheat pressure are within their respective preset allowable pressure ranges; the operating rate of each valve is less than or equal to the preset maximum allowable rate; the flow rate of each channel is within the preset allowable flow rate range; and the thermal stress and turbine vibration amplitude are less than or equal to their corresponding safety thresholds.

[0139] In some embodiments, when the main steam pressure and reheat pressure are respectively within their respective preset allowable pressure ranges, the opening increment command of the main steam valve and reheat valve is executed.

[0140] For example, when the main steam pressure is lower than a preset lower limit or higher than a preset upper limit, the opening increment of the main steam valve is limited to prevent the main steam pressure from exceeding the limit further.

[0141] Specifically, the allowable range of main steam pressure is determined based on the unit's rated parameters and the current load, and is typically 90% to 105% of the rated pressure.

[0142] In other embodiments, when the operating rate of each valve is less than or equal to the preset maximum allowable rate, the opening increment command of the corresponding valve is executed.

[0143] For example, the maximum permissible operating rate of the main steam regulating valve is typically 10% / s to 20% / s, and the maximum permissible operating rate of the reheat regulating valve is typically 5% / s to 15% / s.

[0144] Specifically, the valve actuation rate limit is determined based on the performance of the DEH actuator and the unit safety requirements to prevent excessively rapid valve actuation from damaging the actuator or exacerbating unit vibration.

[0145] In some other embodiments, when the flow rate of each channel is within a preset allowable flow rate range, the airflow distribution command for the corresponding channel is executed.

[0146] For example, the upper limit of the allowable flow rate of the high-pressure cylinder is 105% of the rated flow rate, and the lower limit of the allowable flow rate of the low-pressure cylinder is 30% of the rated flow rate, to prevent the last stage blades from fluttering due to insufficient flow.

[0147] Specifically, the allowable range of flow rate is determined based on the design parameters and safety requirements of the turbine's flow passage.

[0148] In some other embodiments, airflow distribution control commands are executed when the thermal stress and turbine vibration amplitude are less than or equal to the corresponding safety thresholds.

[0149] For example, the thermal stress safety threshold of the high-pressure rotor of a steam turbine is typically 80% of the allowable stress of the material, and the safety threshold for the vibration amplitude of the turbine shaft system is typically 0.05 mm.

[0150] Specifically, the thermal stress is calculated based on the rotor temperature field, and the vibration amplitude is measured in real time by the shaft vibration measuring points.

[0151] Thus, by setting strict execution conditions, the present invention ensures that all airflow distribution commands are executed within the unit's safety boundaries, thereby avoiding unit safety accidents caused by over-adjustment.

[0152] In the plastic optimization adjustment method for primary frequency regulation of supercritical units provided by this invention, a corrected model that can accurately reflect the second-level steam flow inertia is established by replacing the pure integral element in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback. Based on this, the total amount of releaseable gas flow energy is calculated according to the boiler-side state variables output by the corrected model, combined with the pre-generated power conversion relationship and real-time operating margin. This transforms the frequency regulation capability, which originally relied on experience-based judgment, into a resource boundary that can be quantified online, ensuring that the optimization decision does not exceed the actual steam supply capacity of the boiler. Furthermore, when a primary frequency regulation event is detected, the frequency regulation process is divided into multiple continuous stages with different control objectives. In each stage, the main steam valve, reheat valve, and extraction steam valve are subjected to multi-objective rolling optimization with the total amount of releaseable gas flow energy as a constraint. This allows different stages to adopt matching control strategies, ultimately outputting a gas flow distribution control command that balances rapid response and continuous support capabilities. This effectively improves the adjustment speed and power support time of primary frequency regulation of supercritical units and ensures the safe operation of the unit.

[0153] Optionally, historical disturbance data and primary frequency regulation test data of the unit can be used to establish initial values ​​of boiler model parameters and turbine power contribution parameters under different load ranges.

[0154] For example, during the offline phase, historical disturbance data of the unit over the past year, primary frequency regulation test data from more than three tests, and AGC small disturbance test data are collected. Initial parameter values ​​for four typical load ranges of 25%, 50%, 75%, and 100% rated load are established using the least squares method, recursive least squares method, or Bayesian identification method.

[0155] Specifically, the Bayesian identification method uses the offline parameter set as the prior distribution, the real-time measurement data as the likelihood function, and outputs the posterior parameter distribution as the power parameter, which can effectively handle measurement noise and model uncertainty.

[0156] Furthermore, during operation, the boiler model parameters and turbine power contribution parameters are updated using natural frequency disturbances or small-amplitude adjustment actions as excitation.

[0157] For example, the online update uses a recursive least squares method with a forgetting factor, the forgetting factor ranging from 0.95 to 0.995. When the unit is in a stable operating state, the forgetting factor is 0.99 to maintain parameter stability; when the unit is in a state of large-scale adjustment, the forgetting factor is 0.96 to accelerate parameter tracking.

[0158] Specifically, online updates only adjust key parameters, such as boiler time constant and high-pressure cylinder flow-power coefficient, while non-key parameters remain unchanged from their offline identification values, reducing the amount of computation.

[0159] Furthermore, the model credibility index is calculated based on the updated parameters. When the model credibility index is lower than the preset credibility threshold, the system switches to a single-frequency control mode that does not include multi-objective rolling optimization.

[0160] For example, model credibility The values ​​are calculated based on the prediction residuals of the modified boiler model and the identification residuals of the turbine power model, and range from 0 to 1. When, full airflow distribution control is allowed; when At that time, the energy release of the gas storage is limited to 70% of the total amount; when In case of abnormalities at key measuring points, the system degrades to conventional primary frequency regulation control while retaining pressure and valve safety constraints.

[0161] Specifically, the prediction residual refers to the difference between the model's predicted value and the actual measured value. The smaller the residual, the higher the model's reliability.

[0162] Thus, this invention ensures the accuracy of the model under different working conditions by combining offline identification with online updating of parameter calibration methods; at the same time, by setting model reliability indicators, it realizes adaptive switching of control modes, ensuring the safe operation of the system when the model is inaccurate or the measurement points are abnormal.

[0163] Optionally, after receiving the airflow distribution control command, the airflow distribution control command can be superimposed with the original primary frequency modulation command to generate a superimposed command.

[0164] For example, the original frequency adjustment command is generated by the DEH system based on the frequency difference and speed inequality, and the airflow distribution control command of the present invention is superimposed on the original command as a correction.

[0165] Specifically, the magnitude of the airflow distribution correction is determined by the difference between the optimal airflow distribution command output by the multi-objective optimization module and the original DEH primary frequency modulation command, ensuring that the correction will not cause the total command to exceed the safe range.

[0166] Furthermore, the superimposed instructions are sent to the digital electro-hydraulic control system, which processes them according to the preset amplitude limiting parameters, rate limit values, and protection interlocking logic, and outputs the processed instructions to the corresponding actuators.

[0167] For example, the DEH system performs amplitude limiting processing on the superimposed commands to ensure that the valve opening does not exceed the range of 0 to 100%; performs rate limiting processing to ensure that the valve action rate does not exceed the maximum allowable value; and performs protection lockout processing to immediately lock out all regulation commands when the turbine protection is activated.

[0168] Specifically, the controller of the present invention does not directly bypass the original turbine protection. All output commands are passed through the original DEH valve amplitude limiting, speed limiting, manual / automatic status and turbine protection interlock to ensure unit safety.

[0169] Thus, by superimposing a correction amount before the original frequency modulation command, the present invention achieves seamless integration with the existing DEH system without requiring modification of the original protection logic, thereby reducing the difficulty of engineering implementation and security risks.

[0170] Optionally, before each multi-objective rolling optimization is performed, boiler pressure constraint signals, boiler temperature constraint signals, and combustion response constraint signals can also be obtained from the digital distributed control system.

[0171] For example, constraint signals such as the upper and lower limits of main steam pressure, reheat pressure, main steam temperature, reheat temperature, and boiler combustion rate are obtained from the DCS system.

[0172] Specifically, these constraint signals are generated in real time by the boiler control system, reflecting the current operating status and safety boundaries of the boiler.

[0173] Furthermore, boiler pressure constraint signal, boiler temperature constraint signal, and combustion response constraint signal are used as additional constraints for multi-objective rolling optimization.

[0174] For example, when the boiler combustion rate has reached its upper limit, the boiler's continuous power supply capacity constraint is added to the multi-objective rolling optimization to limit the extension of the power support time and avoid unstable boiler combustion.

[0175] Specifically, the additional constraints have a higher priority than the optimization objective, ensuring that the optimized solution always meets the requirements for safe boiler operation.

[0176] Thus, by introducing boiler operation constraint signals from the DCS side, this invention achieves primary frequency regulation optimization control for coordinated turbine and boiler operation, avoiding abnormal boiler operation caused by excessive adjustment on the turbine side.

[0177] In some embodiments, the method further includes the step of adaptively switching the control strategy according to the current operating conditions of the unit, including high load conditions, medium load conditions, low load conditions, heating steam extraction restriction conditions, and low model reliability conditions.

[0178] For example, the control strategies under different operating conditions are as follows: High load / valve position close to the upper limit: main steam flow is high, valve position margin is small, and pressure margin is limited. Priority is given to using extraction steam for short-term reduction and reheat channel to support the power platform, limiting the rapid opening of the main steam regulating valve, and preventing the main steam pressure from collapsing rapidly and the valve from saturating. Medium load / good pressure margin operating condition: There is a certain margin in valve position, pressure and steam extraction. The steam flow of each airflow channel is allocated according to the multi-objective optimal solution to achieve a balance between initial regulation speed and power support time, and to prevent the platform from collapsing after the peak. Low load / significant boiler lag: slow combustion response and potentially small superheat margin. Reduce the maximum power increment target and extend the power support time. Use the boiler's continuous supply capacity as the main constraint to prevent temperature deviation and combustion instability. Heating / extraction steam restricted conditions: Some extraction steam cannot be significantly reduced. Lock the regulating valve of the restricted extraction steam stage and use only the adjustable extraction steam stage and reheat channel for airflow distribution to prevent disturbance of heating parameters; - Low reliability of measuring points or models: Critical measuring points are abnormal or the model error is large. Reduce the order to conventional primary frequency regulation control and retain pressure and valve safety constraints to ensure safety first.

[0179] Specifically, the strategy switching employs a weighted smooth transition and an instruction ramp transition. If the pressure margin decreases under high load, the controller gradually reduces... ,improve and If the margin of the intermediate pressure / reheat channel increases, the participation coefficient of the reheat / intermediate pressure valve is gradually increased; if steam extraction is limited, the corresponding steam extraction valve variable is frozen from the optimization vector and compensated by other channels.

[0180] Thus, by employing a partitioned control strategy and smooth switching logic, this invention enables the same control method to adapt to wide-load operation and various special working conditions, thereby improving the system's adaptability and robustness.

[0181] In some embodiments, when a turbine protection action, a boiler main fuel trip (MFT) warning, pressure exceeding limits, valve jamming, abnormal reheat temperature, or shaft vibration exceeding limits is detected, the original DEH / DCS protection logic is restored.

[0182] For example, when the steam turbine experiences overspeed protection, large axial displacement protection, or the boiler experiences MFT warning, the controller immediately cuts off all airflow distribution correction commands and restores the original DEH primary frequency regulation control logic.

[0183] Specifically, the fail-safe logic has a higher priority than all optimization control logic, ensuring that the unit can quickly switch to safe control mode when an anomaly occurs.

[0184] Thus, by setting up a comprehensive failure protection mechanism, the present invention ensures the safe operation of the unit under any abnormal conditions.

[0185] In this embodiment of the invention, the primary frequency regulation action has a higher priority than the AGC slow regulation. During the primary frequency regulation activation period, the AGC target can be used as a slow variable to enter the follow-up stage; after the frequency is restored, the controller performs power recovery according to the load given by the AGC and the coordination state of the boiler and turbine, and the rate of power recovery is constrained by the main steam pressure recovery trend.

[0186] For example, during the first 30 seconds of a frequency regulation operation, the AGC command is disabled to adjust the unit power; during the subsequent relay phase, the AGC target power is added as a slow variable to the optimization target, so that the unit power gradually approaches the AGC target; after the frequency is restored, the unit power is recovered to the AGC given value at a rate of 0.5% / s to 2% / s according to the main steam pressure recovery rate.

[0187] Specifically, the power recovery rate is adaptively adjusted according to the degree of main steam pressure recovery. The faster the pressure recovery, the higher the recovery rate, thus avoiding slow power recovery from affecting the AGC regulation performance.

[0188] Thus, by organically coordinating with AGC and boiler-generator coordination control, this invention achieves the coordinated unification of primary frequency regulation and grid dispatch commands, which not only meets the frequency security support requirements but also ensures the economic efficiency of the unit's steady-state operation.

[0189] In a specific embodiment, taking a supercritical once-through boiler turbine generator unit as an example, the unit is at 80% rated load, and the main steam pressure and reheat pressure are both within the normal operating range. The DEH has a primary frequency regulation function, and the DCS can provide main steam pressure, main steam temperature, main steam flow, reheat pressure, valve opening degree, and extraction valve position signals.

[0190] For example, the parameters in this embodiment are set as follows: Prediction Time Domain The duration is 60 seconds, and the control cycle is 0.2 to 1 second. Theories for generating target power trajectories based on frequency difference and rotational speed inequality And confined within the available airflow energy storage boundary; The boiler model adopts a two-segment first-order inertia: segment I represents steam generation, and segment II represents superheated transport; parameters are corrected segment by segment according to load. The turbine model is configured as follows: the high-pressure cylinder is used for rapid ramping; the reheat / intermediate-pressure channel is used for the platform; and the low-pressure and extraction steam redistribution are used for integrating electricity. Safety constraints include: lower limit of main steam / reheat pressure, valve speed limit, thermal stress, extraction steam heating constraints, vibration and protection interlock.

[0191] Specifically, the control process is as follows: When the system frequency drops by 0.1Hz and exceeds the primary frequency regulation dead zone, the controller first calculates that the current main steam pressure margin is 8% of the rated pressure and the valve position margin is 15%, determining that the high-pressure cylinder channel can be used for fast-segment release. Simultaneously, based on the reheat pressure margin of 10% of the rated pressure and the reheat pipe volume, the controller calculates that the intermediate-pressure cylinder platform support capacity is 3% of the rated power. If the extraction steam system is not constrained by heating supply, the second and third stage regenerative extraction steam is selected for short-term reduced extraction, which is expected to transfer 2% of the rated power of steam to the intermediate and low-pressure cylinders for work.

[0192] In the first 3 seconds, the main steam regulating valve opens rapidly at the allowable valve speed of 15% / s, with the opening increment limited to within 10%, ensuring that the main steam pressure does not drop below the lower limit. From 3 to 20 seconds, the optimizer gradually increases the opening of the reheat / intermediate pressure regulating valve to 8%, while simultaneously closing the second and third stage extraction steam valves by 15%, keeping the power curve near the target plateau of 6% of the rated power. From 20 to 60 seconds, the boiler combustion and feedwater response take over, and the main steam pressure begins to rise. The controller gradually reduces the extraction steam redistribution amplitude according to the main steam pressure recovery trend, and the extraction steam valve returns to its original opening at 50 seconds. After the frequency returns to the rated value, the main steam valve and reheat valve are restored to the steady-state opening at a rate of 1% / s according to the main steam pressure recovery rate, to avoid a secondary drop in power.

[0193] The expected results are as follows: Compared to primary frequency regulation logic that relies solely on the rapid opening of the main steam valve, this embodiment can increase the initial power ramp-up speed by 20%–30%, extend the power support time by 50%–100%, and increase the frequency regulation integral power within the assessment window by 30%–50% without exceeding pressure and valve safety constraints. The specific improvement is determined by the unit capacity, load point, valve margin, extraction steam constraints, and parameter identification accuracy.

[0194] Optionally, the present invention also serves operation control and simulation evaluation: on the unit side, it can be used as the primary frequency regulation enhancement control logic of DEH / DCS; on the planning and testing side, it can be used as a user-defined model that can replace the traditional inherent model in BPA transient simulation, to predict the primary frequency regulation availability under different loads, different pressure margins and different valve strategies.

[0195] It should be noted that this invention is not limited to units of a specific capacity, but is applicable to supercritical or ultra-supercritical once-through boiler units of 300MW, 600MW, 660MW, and 1000MW levels; for units with heating steam extraction, the heating steam extraction pressure and heating network constraints can be used as hard constraints; for units with energy storage or bypass steam replenishment, the energy storage power or bypass steam replenishment valve can be included in the same optimization vector.

[0196] The following describes the plastic optimization control system for primary frequency regulation of supercritical units provided by the present invention. The plastic optimization control system for primary frequency regulation of supercritical units described below can be referred to in correspondence with the plastic optimization control method for primary frequency regulation of supercritical units described above.

[0197] Figure 5 This is a structural diagram of a flexible optimization control system for primary frequency regulation of a supercritical unit, provided in an embodiment of the present invention. The flexible optimization control system for primary frequency regulation of the supercritical unit includes: a model correction module 501, an energy storage assessment module 502, and a multi-objective optimization module 503.

[0198] The module comprises: a model correction module 501, which replaces the pure integral element representing energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, and outputs the boiler-side state variables; an energy storage evaluation module 502, connected to the model correction module 501, which calculates the total amount of energy that can be released by the gas flow based on the boiler-side state variables and the pre-generated power conversion relationship of each gas flow channel of the turbine, combined with the real-time operating margin; and a multi-objective optimization module 503, connected to the energy storage evaluation module 502, which, when a frequency regulation event is detected, divides the frequency regulation process into multiple consecutive stages with different control objectives in time sequence, and performs multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the turbine within each stage, using the total amount of energy that can be released by the gas flow as a constraint, and outputs the gas flow distribution control command.

[0199] In some embodiments, the plastic optimization regulation system for primary frequency regulation of supercritical units provided in this invention further includes a relationship establishment module; the relationship establishment module is used to: identify the power contribution parameters of each airflow channel of the turbine based on the boiler-side state variables; and establish a mapping relationship between the flow rate change and power output of each channel according to the power contribution parameters, as the power conversion relationship.

[0200] In some embodiments, the model correction module 501 is specifically used to: divide the once-through boiler into two spatially distributed lumped links: a steam generation lumped link and a superheated steam transport lumped link; replace the pure integral link of the original model in each lumped link with a first-order inertial link with pressure-flow feedback, which is used to characterize the dynamic coupling relationship between pressure change and outlet flow change in the lumped link; and based on the first-order inertial link, output the pressure state, pressure change rate and steam flow prediction value of each lumped link as the boiler-side state variables.

[0201] In some embodiments, the model correction module 501 is specifically used to: close the pure integral element into the first-order inertial element by expressing the outlet flow rate of the lumped section as a linear function of the pressure of the lumped section; the first-order inertial element has a static gain and a time constant, the static gain reflecting the strength of the pressure-flow ratio feedback, and the time constant reflecting the inertia of steam release and pressure propagation.

[0202] In some embodiments, the above-mentioned relationship establishment module is specifically used to: divide the controllable objects on the turbine side into high-pressure cylinder channels, reheat and connecting pipe channels, intermediate-pressure cylinder channels, low-pressure cylinder channels and extraction steam channels; and identify the flow-power coefficient, dynamic overshoot coefficient and inertia constant of each channel as the power contribution parameters.

[0203] In some embodiments, the energy storage assessment module 502 is specifically used for: calculating the boiler-side releaseable steam equivalent based on the boiler-side state variables; calculating the releaseable airflow increment on the main steam pipe and steam chamber side, the releaseable airflow increment on the reheat side, and the transferable airflow increment on the extraction steam redistribution side based on the power conversion relationship; subtracting a preset safety reserve from the sum of the boiler-side releaseable steam equivalent, the releaseable airflow increment on the main steam pipe and steam chamber side, the releaseable airflow increment on the reheat side, and the transferable airflow increment on the extraction steam redistribution side to obtain the total releaseable airflow energy storage; wherein, the preset safety reserve is determined based on the real-time main steam pressure lower limit margin, the reheat pressure lower limit margin, and the thermal safety margin.

[0204] In some embodiments, the multi-objective optimization module 503 is further configured to: in the fast-segment release phase, with the optimization objective of maximizing the initial regulation speed, control the main steam valve to open rapidly at the allowable rate, and prioritize the release of available airflow in the high-pressure cylinder front steam chamber and the main steam pipe; in the steady-segment diversion phase, with the optimization objective of increasing the maximum power increment and integral power, coordinate the reheat valve opening and temporarily reduce some of the regenerative extraction steam to maintain the power platform; in the subsequent relay phase, with the optimization objective of extending the power support time and ensuring pressure safety, dynamically adjust the valve opening according to the boiler combustion response and gradually replenish the extraction steam; and in the recovery and recycling phase, with the optimization objective of undisturbed recovery, smoothly recover to steady-state conditions according to the pressure recovery trend.

[0205] In some embodiments, the multi-objective optimization module 503 is specifically used to: in the recovery phase, based on the energy deficit and pressure recovery level, smoothly recover the opening degree of the main steam valve, the reheat valve and the extraction steam valve, as well as the extraction steam quantity corresponding to the extraction steam valve, so that the valve opening degree and extraction steam quantity gradually recover to the values ​​corresponding to the steady-state operating conditions.

[0206] In some embodiments, the multi-objective optimization module 503 is further configured to: increase the weights corresponding to the initial adjustment speed and the maximum power increment when the absolute value of the frequency difference is greater than the preset frequency difference threshold and the system frequency drops rapidly; increase the weights corresponding to the power support time and pressure safety and decrease the weights corresponding to the maximum power increment when the main steam pressure margin is less than the preset pressure threshold; and freeze the optimization variables of the restricted extraction stage and increase the compensation weights of other adjustable channels when the extraction steam system is constrained by heating.

[0207] In some embodiments, the multi-objective optimization module 503 is further configured to: establish initial values ​​of boiler model parameters and turbine power contribution parameters under different load ranges using historical disturbance data of the unit and primary frequency regulation test data; update the boiler model parameters and turbine power contribution parameters during operation using natural frequency disturbances or small-amplitude adjustment actions as excitation; calculate the model reliability index based on the updated parameters; and switch to a primary frequency regulation control mode that does not include the multi-objective rolling optimization when the model reliability index is lower than a preset reliability threshold.

[0208] In some embodiments, the plastic optimization adjustment system for primary frequency regulation of supercritical units provided in this invention further includes an instruction execution module; the instruction execution module is used to: superimpose the airflow distribution control instruction with the original primary frequency regulation instruction to generate a superimposed instruction; send the superimposed instruction to the digital electro-hydraulic control system, which processes the instruction according to preset amplitude limiting parameters, rate limiting values ​​and protection interlocking logic, and outputs the processed instruction to the corresponding actuator.

[0209] In some embodiments, the multi-objective optimization module 503 is further configured to: obtain boiler pressure constraint signal, boiler temperature constraint signal and combustion response constraint signal from the digital distributed control system before each execution of the multi-objective rolling optimization; and use the boiler pressure constraint signal, the boiler temperature constraint signal and the combustion response constraint signal as additional constraints for the multi-objective rolling optimization.

[0210] In some embodiments, the above-mentioned instruction execution module is further configured to: execute the airflow distribution control instruction when preset execution conditions are met; wherein the preset execution conditions include: the main steam pressure and reheat pressure are respectively within their respective preset allowable pressure ranges; the operating rate of each valve is less than or equal to the preset maximum allowable rate; the flow rate of each channel is within the preset allowable flow range; and the thermal stress and turbine vibration amplitude are respectively less than or equal to the corresponding safety thresholds.

[0211] In the flexible optimization control system for primary frequency regulation of supercritical units provided by this invention, a corrected model that accurately reflects the second-level steam flow inertia is established by replacing the pure integral element in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback. Based on this, the total amount of releaseable gas flow energy is calculated according to the boiler-side state variables output by the corrected model, combined with the pre-generated power conversion relationship and real-time operating margin. This transforms the frequency regulation capability, which originally relied on empirical judgment, into a resource boundary that can be quantified online, ensuring that the optimization decision does not exceed the actual steam supply capacity of the boiler. Furthermore, when a primary frequency regulation event is detected, the frequency regulation process is divided into multiple continuous stages with different control objectives. In each stage, the main steam valve, reheat valve, and extraction steam valve are subjected to multi-objective rolling optimization with the total amount of releaseable gas flow energy as a constraint. This allows different stages to adopt matching control strategies, ultimately outputting a gas flow distribution control command that balances rapid response and continuous support capabilities. This effectively improves the regulation speed and power support time of primary frequency regulation of supercritical units and ensures the safe operation of the unit.

[0212] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6 As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communications bus 640, wherein the processor 610, the communications interface 620, and the memory 630 communicate with each other through the communications bus 640.

[0213] The processor 610 can call logic instructions in the memory 630 to execute a plastic optimization adjustment method for primary frequency regulation of a supercritical unit. This method includes: replacing the pure integral element characterizing energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, and outputting the boiler-side state variable; calculating the current total amount of energy stored in the releaseable airflow based on the boiler-side state variable and the pre-generated power conversion relationship of each airflow channel of the turbine, combined with the real-time operating margin; when a primary frequency regulation event is detected, dividing the frequency regulation process into multiple consecutive stages with different control objectives in time sequence, and within each stage, using the total amount of energy stored in the releaseable airflow as a constraint, performing multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the turbine, and outputting airflow distribution control instructions.

[0214] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0215] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the plastic optimization adjustment method for primary frequency regulation of supercritical units provided by the above methods. The method includes: replacing the pure integral element characterizing the energy storage characteristics in the dynamic model of the DC boiler with a first-order inertial element with pressure-flow feedback, and outputting the boiler-side state variable; calculating the current total amount of energy stored in the releaseable airflow based on the boiler-side state variable and the pre-generated power conversion relationship of each airflow channel of the turbine, combined with the real-time operating margin; when a primary frequency regulation event is detected, dividing the frequency regulation process into multiple continuous stages with different control objectives in time sequence, and within each stage, using the total amount of energy stored in the releaseable airflow as a constraint, performing multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the turbine, and outputting airflow distribution control commands.

[0216] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a plastic optimization adjustment method for primary frequency regulation of a supercritical unit provided by the methods described above. The method includes: replacing the pure integral element characterizing the energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element with pressure-flow feedback, and outputting the boiler-side state variable; calculating the current total amount of energy stored in the releaseable airflow based on the boiler-side state variable and the pre-generated power conversion relationship of each airflow channel of the turbine, combined with the real-time operating margin; when a primary frequency regulation event is detected, dividing the frequency regulation process into multiple consecutive stages with different control objectives in time sequence, and within each stage, using the total amount of energy stored in the releaseable airflow as a constraint, performing multi-objective rolling optimization on the main steam valve, reheat valve, and extraction steam valve of the turbine, and outputting airflow distribution control commands.

[0217] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0218] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0219] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for plastically optimized regulation of primary frequency control in a supercritical unit, characterized in that, The method includes: The pure integral element characterizing the energy storage characteristics in the dynamic model of the DC boiler is replaced with a first-order inertial element with pressure-flow feedback, and the boiler-side state variables are output. Based on the boiler-side state variables and the pre-generated power conversion relationships of each airflow channel of the turbine, combined with the real-time operating margin, the total amount of energy that can be released by the airflow is calculated. When a frequency regulation event is detected, the frequency regulation process is divided into multiple consecutive stages with different control objectives in time sequence. Within each stage, the main steam valve, reheat valve and extraction steam valve of the steam turbine are optimized for multiple objectives by using the total amount of releasable airflow energy as a constraint, and airflow distribution control commands are output.

2. The method according to claim 1, characterized in that, The pre-generated power conversion relationships of each airflow channel in the steam turbine are obtained through the following steps: Based on the boiler-side state variables, the power contribution parameters of each airflow channel of the steam turbine are identified; Based on the power contribution parameters, a mapping relationship between the flow rate change and power output of each channel is established, which serves as the power conversion relationship.

3. The method according to claim 1, characterized in that, The process of replacing the pure integral element characterizing energy storage characteristics in the dynamic model of a DC boiler with a first-order inertial element having pressure-flow feedback, and outputting boiler-side state variables, includes: The once-through boiler is divided into two spatially distributed lumped sections: a steam generation lumped section and a superheated steam transport lumped section. The pure integral element of the original model in each lumped segment is replaced with a first-order inertial element with pressure-flow feedback. The first-order inertial element is used to characterize the dynamic coupling relationship between pressure change and outlet flow change in the lumped segment. Based on the first-order inertial element, the pressure state, pressure change rate, and steam flow prediction value of each lumped segment are output as the boiler-side state variables.

4. The method according to claim 3, characterized in that, The replacement of the pure integral element of the original model in each lumped segment with a first-order inertial element with pressure-flow feedback includes: By expressing the outlet flow rate of the lumped section as a linear function of the pressure of the lumped section, the pure integral element is closed into the first-order inertial element; The first-order inertial element has a static gain and a time constant. The static gain reflects the strength of the pressure-flow ratio feedback, and the time constant reflects the inertia of steam release and pressure propagation.

5. The method according to claim 2, characterized in that, The process of identifying the power contribution parameters of each airflow channel of the steam turbine based on the boiler-side state variables includes: The controllable objects on the turbine side are divided into high-pressure cylinder passage, reheat and connecting pipe passage, intermediate-pressure cylinder passage, low-pressure cylinder passage and extraction steam passage; The flow-power coefficient, dynamic overshoot coefficient, and inertia constant of each channel are identified as the power contribution parameters.

6. The method according to claim 1, characterized in that, The calculation of the total releaseable gas flow energy storage based on the boiler-side state variables and the pre-generated power conversion relationships of each gas flow channel of the turbine, combined with real-time operating margin, includes: Based on the boiler-side state parameters, calculate the boiler-side release steam equivalent. Based on the power conversion relationship, the releaseable airflow increment on the main steam pipe and steam chamber side, the releaseable airflow increment on the reheat side, and the transferable airflow increment on the extraction steam redistribution side are calculated respectively. The total energy storage of the releaseable airflow is obtained by subtracting a preset safety reserve from the sum of the boiler-side releaseable steam equivalent, the releaseable airflow increment of the main steam pipe and steam chamber side, the releaseable airflow increment of the reheat side, and the transferable airflow increment of the extraction steam redistribution side. The preset safety reserve is determined based on the real-time main steam pressure lower limit margin, reheat pressure lower limit margin, and thermal safety margin.

7. The method according to claim 1, characterized in that, The multiple successive phases with different control objectives include: a fast-segment release phase, a steady-segment diversion phase, a follow-up phase, and a recovery and recycling phase. The optimization objectives of the multi-objective rolling optimization include at least one of the following: maximizing the integral charge of the primary frequency regulation, maximizing the initial adjustment speed, maximizing the maximum power increment, and maximizing the power support time.

8. The method according to claim 7, characterized in that, The method further includes: During the rapid release phase, with the goal of maximizing the initial adjustment speed, the main steam valve is controlled to open rapidly at the allowable rate, prioritizing the release of available airflow in the high-pressure cylinder front chamber and the main steam pipe. During the steady-state diversion phase, with the optimization goal of increasing the maximum power increment and integral power, the opening of the reheat valve is coordinated and a portion of the regenerative extraction steam is reduced for a short period of time to maintain the power plateau. In the aforementioned relay phase, with the optimization goals of extending the power support time and ensuring pressure safety, the valve opening is dynamically adjusted according to the boiler combustion response to gradually replenish the extracted steam. During the recovery phase, with the goal of undisturbed recovery, the system smoothly recovers to steady-state conditions according to the pressure recovery trend.

9. The method according to claim 8, characterized in that, The recovery and recycling phase, with the optimization objective of undisturbed recovery, smoothly restores the system to steady-state conditions according to the pressure recovery trend, including: During the recovery phase, based on the energy deficit and pressure recovery level, the opening degrees of the main steam valve, the reheat valve, and the extraction steam valve, as well as the extraction steam volume corresponding to the extraction steam valve, are smoothly recovered so that the valve opening degree and extraction steam volume gradually return to the values ​​corresponding to the steady-state operating conditions.

10. The method according to claim 7, characterized in that, The multi-objective rolling optimization is defined by an objective function and a set of constraints; The objective function includes a weighted combination of at least two of the following: primary frequency modulation integral energy term, initial adjustment speed term, maximum power increment term, and power support time term, as well as a penalty term that penalizes behaviors that violate the constraints. The constraints include at least one of the following: main steam pressure constraint, reheat pressure constraint, valve opening and actuation rate constraint, flow rate constraint, and thermal stress constraint.

11. The method according to claim 10, characterized in that, The method further includes: When the absolute value of the frequency difference is greater than the preset frequency difference threshold and the system frequency drops rapidly, the weights corresponding to the initial adjustment speed and the maximum power increment are increased. When the main steam pressure margin is less than the preset pressure threshold, increase the weight of power support time and pressure safety, and decrease the weight of maximum power increment. When the extraction steam system is constrained by heating supply, the optimization variables of the restricted extraction steam stage are frozen, and the compensation weights of other adjustable channels are increased.

12. The method according to claim 1, characterized in that, The method further includes: Using historical disturbance data and primary frequency regulation test data of the unit, initial values ​​of boiler model parameters and turbine power contribution parameters under different load ranges were established; During operation, the boiler model parameters and the turbine power contribution parameters are updated using natural frequency disturbances or small-amplitude adjustment actions as excitations. The model credibility index is calculated based on the updated parameters. When the model credibility index is lower than the preset credibility threshold, the system switches to a single-frequency control mode that does not include the multi-objective rolling optimization.

13. The method according to claim 1, characterized in that, The method further includes: The airflow distribution control command is superimposed with the original primary frequency modulation command to generate a superimposed command; The superimposed instructions are sent to the digital electro-hydraulic control system, which processes them according to preset amplitude limiting parameters, rate limiting values ​​and protection interlocking logic, and outputs the processed instructions to the corresponding actuators.

14. The method according to claim 1, characterized in that, The method further includes: Before each execution of the multi-objective rolling optimization, boiler pressure constraint signal, boiler temperature constraint signal, and combustion response constraint signal are obtained from the digital distributed control system. The boiler pressure constraint signal, the boiler temperature constraint signal, and the combustion response constraint signal are used as additional constraints for the multi-objective rolling optimization.

15. The method according to claim 1, characterized in that, The output airflow distribution control commands include: main steam valve opening increment command, reheat valve opening increment command, each stage extraction steam valve opening increment command, and optional bypass valve opening increment command. The method further includes: The airflow distribution control command is executed when the preset execution conditions are met; The preset execution conditions include: the main steam pressure and reheat pressure are respectively within their respective preset allowable pressure ranges; the operating rate of each valve is less than or equal to the preset maximum allowable rate; the flow rate of each channel is within the preset allowable flow rate range; and the thermal stress and turbine vibration amplitude are respectively less than or equal to the corresponding safety thresholds.

16. A flexible optimization control system for primary frequency regulation of a supercritical unit, characterized in that, The system includes: The model correction module is used to replace the pure integral element representing the energy storage characteristics in the dynamic model of the DC boiler with a first-order inertial element with pressure-flow feedback, and output the boiler-side state variables. The energy storage assessment module, connected to the model correction module, is used to calculate the total amount of energy that can be released by the current airflow based on the boiler-side state variables and the pre-generated power conversion relationship of each airflow channel of the steam turbine, combined with the real-time operating margin. The multi-objective optimization module, connected to the energy storage evaluation module, is used to divide the frequency regulation process into multiple consecutive stages with different control objectives in time sequence when a frequency regulation event is detected. In each stage, the main steam valve, reheat valve and extraction steam valve of the steam turbine are optimized in a multi-objective rolling manner with the total amount of releasable airflow energy as a constraint, and the airflow distribution control command is output.

Images