A control method and system for high-pressure water level of a waste heat boiler of a gas turbine unit

Abstract

The present application relates to a kind of gas turbine unit waste heat boiler high pressure package water level control method and system, it is related to gas turbine unit waste heat boiler high pressure package water level control performance promotion technical field, it solves the problem that traditional three impulse control strategy based on PID is insufficient when facing such frequent, large-scale operating condition changes, and its technical points are that the present application is a cascade control structure.In outer loop, according to high pressure package feed water flow command signal and high pressure package water level measurement signal, the estimation value of high pressure package water level true signal is obtained by Kalman filtering algorithm, to eliminate the interference of field noise on water level measurement signal, enhance the reliability of high pressure package water level feedback signal.The effect is that this method can effectively guarantee the operation reliability and safety of waste heat boiler high pressure package feed water system under the condition that power grid AGC fluctuates frequently and field noise has significant influence on measurement signal.

Description

Technical Field

[0001] This invention relates to the field of improving the performance of high-pressure boiler water level control in waste heat boilers of gas turbine units, specifically to a method and system for controlling the water level of high-pressure boilers in waste heat boilers of gas turbine units. Background Technology

[0002] In new power systems, the continuous increase in the proportion of new energy sources necessitates that gas turbines, as a key resource for flexible regulation, must frequently respond to grid AGC commands. This process causes drastic and rapid fluctuations in unit power and waste heat boiler operating conditions, severely disrupting the high-pressure coil water level balance. Real-time matching of feedwater flow and evaporation becomes difficult, resulting in frequent and significant fluctuations in water level.

[0003] The high-pressure water level is a core parameter for ensuring the safe, stable, and economical operation of waste heat boilers. Traditional PID-based three-impulse control strategies are inadequate in dealing with such frequent and significant changes in operating conditions. At the same time, the inherent noise interference in the water level measurement signal is easily amplified by the controller, causing ineffective oscillations of the regulating valve and equipment wear. This often forces operators to reduce control parameters, sacrificing system response speed and control accuracy for stability.

[0004] Therefore, with gas turbines deeply involved in grid AGC regulation, how to accurately estimate the true value of the high-voltage transformer water level through advanced filtering technology and integrate it with the active disturbance rejection control architecture to develop an advanced water level control strategy with strong anti-interference and noise suppression capabilities has become a key technology for ensuring unit safety, improving operational flexibility, and supporting the stable and reliable operation of new power systems. Summary of the Invention

[0005] (a) Technical problems to be solved In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler.

[0006] (II) Technical Solution To achieve the above objectives, the main technical solutions adopted by the present invention include: This invention provides a method for controlling the water level of the high-pressure boiler in a gas turbine unit's waste heat boiler, comprising: S1. The control system is a cascaded three-impulse control structure, consisting of a Kalman filter, a linear active disturbance rejection controller, an inner loop, and a steam flow feedforward. S2. Using the Kalman filtering algorithm, the estimated value of the actual water level signal of the high-pressure transformer is obtained by using the high-pressure transformer water flow command signal and the high-pressure transformer water level measurement signal. S3. The linear active disturbance rejection controller generates a water supply flow command signal based on the estimated value of the real water level signal from the high-pressure transformer and the given value signal.

[0007] Preferably, a Kalman filtering algorithm is used to enhance the reliability of the high-pressure transformer water level feedback signal. The input signals of the Kalman filtering module are the high-pressure transformer water flow command signal and the high-pressure transformer water level measurement signal, and the output signal is an estimate of the actual high-pressure transformer water level signal.

[0008] Preferably, the Kalman filtering algorithm includes two parts: prediction and measurement update; Preferably, the prediction steps of the Kalman filtering algorithm are as follows: (1) Calculate the prior state estimate of the high-pressure water level according to equation (1); (1) In the formula, express k Prior estimates of the system state variables at time t. express k Posterior estimate of the system state variables at time -1 The state matrix of the high-pressure water level object model, The control matrix for the high-pressure water level object model. express k The input quantity of the system at time -1, i.e. k -1 is the high-pressure water supply flow command signal.

[0009] (2) Calculate the prior state estimation error covariance matrix according to equation (2); (2) In the formula, for k The prior state estimation error covariance matrix at time step 1. for k -1 time posterior state estimation error covariance matrix Let be the covariance matrix of the process noise.

[0010] Preferably, the measurement update steps of the Kalman filtering algorithm are as follows: (1) Calculate the Kalman gain according to equation (3); (3) In the formula, for k Kalman gain at time step This is the covariance matrix for measuring noise.

[0011] (2) Calculate the posterior state estimate according to equation (4); (4) In the formula, for k Posterior state estimation at time 10:00. This is a high-pressure water level measurement signal that is subject to noise pollution.

[0012] (3) Calculate the output of the Kalman filter according to equation (5). k Estimated value of the actual signal of the high-pressure transformer water level at any given time ; (5) (4) Update the posterior state estimation error covariance matrix according to equation (6); (6) In the formula, for k The covariance matrix of the posterior state estimation error at time step [time]. Represents a unit array.

[0013] Preferably, the outer loop employs a linear active disturbance rejection controller, which includes an extended state observer, a tracking differentiator, and a linear feedback unit.

[0014] Preferably, the extended state observer is used to obtain real-time estimates of the system state and total disturbance, and the input signals of the extended state observer include... k Constantly send water flow command signals and high-pressure water level estimation signal The output includes: high-pressure water level observation signal. High-pressure water level change rate observation signal and total disturbance observation signal The extended state observer adopts a third-order integral cascade structure, which can be expressed as Equation (7). (7) This represents the derivative of the high-pressure water level observation signal. The derivative of the observed signal representing the rate of change of water level in the high-pressure transformer. The derivative of the observed signal representing the total disturbance. This represents the state observer gain coefficient. is the cutoff frequency of the extended state observer.

[0015] Preferably, the tracking differentiator is used to arrange the signal transition process as follows: (8) In the formula, The given signal is generated by the tracking differentiator. To track the derivative of the given signal generated by the differentiator; The rate of change signal of the given value is generated by the tracking differentiator. The derivative for tracking the rate of change of a given value formed by the differentiator; This represents the input given value signal. The bandwidth parameter determines the tracking speed.

[0016] Preferably, the linear feedback unit linearly combines the high-pressure water level observation signal with the transition signal used by the tracking differentiator to arrange the signal according to equation (9) to form the controller output command signal; (9) In the formula, Formed by a linear active disturbance rejection controller k The control command signal at time +1 Indicates the cutoff frequency of the control loop. This is the gain coefficient, selected based on the static gain of the high-voltage transformer.

[0017] Preferably, the linear active disturbance rejection controller control command signal This signal is superimposed on the steam flow signal D to form a feedwater flow command signal. .

[0018] (III) Beneficial Effects The beneficial effects of this invention are: This invention discloses a method for controlling the high-pressure boiler water level in a gas turbine unit's waste heat boiler. Based on a cascade three-impulse control structure, it combines Kalman filtering with advanced linear active disturbance rejection (ADR) control technology. The Kalman filtering algorithm estimates the true high-pressure boiler water level signal, eliminating interference from field noise and enhancing the reliability of the water level feedback signal. The ADR controller performs real-time estimation of system state and total disturbance, designs transient processes, and generates control commands, thereby increasing the high-pressure boiler water level control system's ability to suppress disturbances. This invention achieves accurate estimation of the true high-pressure boiler water level through Kalman filtering technology and integrates it with active disturbance rejection linear ADR, effectively improving the system's anti-interference capabilities, operational flexibility, and safety. Attached Figure Description

[0019] Figure 1 This is a control system diagram of a method for controlling the water level of a high-pressure boiler in a gas turbine unit waste heat boiler according to the present invention. Figure 2 This is a structural diagram of the extended state observer in an active disturbance rejection controller. Detailed Implementation

[0020] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0022] See Figure 1 This embodiment provides a method for controlling the water level of the high-pressure boiler in a gas turbine unit's waste heat boiler, including: S1. The control system is a cascaded three-impulse control structure, consisting of a Kalman filter, a linear active disturbance rejection controller, an inner loop, and a steam flow feedforward. S2. Using the Kalman filtering algorithm, the estimated value of the actual water level signal of the high-pressure transformer is obtained by using the high-pressure transformer water flow command signal and the high-pressure transformer water level measurement signal. S3. The linear active disturbance rejection controller generates a water supply flow command signal based on the estimated value of the real water level signal from the high-pressure transformer and the given value signal.

[0023] Specifically, A Kalman filtering algorithm is employed to enhance the reliability of the high-pressure transformer water level feedback signal. The input signals of the Kalman filtering module are the high-pressure transformer water flow command signal and the high-pressure transformer water level measurement signal, and the output signal is an estimate of the actual high-pressure transformer water level signal.

[0024] Specifically, The Kalman filtering algorithm consists of two parts: prediction and measurement update. Specifically, The prediction steps of the Kalman filter algorithm are as follows: (1) Calculate the prior state estimate of the high-pressure water level according to equation (1); (1) In the formula, express k Prior estimates of the system state variables at time t. express k Posterior estimate of the system state variables at time -1 The state matrix of the high-pressure water level object model, The control matrix for the high-pressure water level object model. A , B The matrix was obtained through a step test of the feedwater flow rate of the high-pressure transformer; express k The input quantity of the system at time -1, i.e. k-1 is the high-pressure water supply flow command signal.

[0025] (2) Calculate the prior state estimation error covariance matrix according to equation (2); (2) In the formula, for k The prior state estimation error covariance matrix at time step 1. for k -1 time posterior state estimation error covariance matrix Let be the covariance matrix of the process noise.

[0026] Specifically, The measurement update steps of the Kalman filtering algorithm are as follows: (1) Calculate the Kalman gain according to equation (3); (3) In the formula, for k Kalman gain at time step To measure the covariance matrix of the noise, C The observation matrix of the high-pressure transformer water level object model is obtained through a step test of the high-pressure transformer water flow rate.

[0027] (2) Calculate the posterior state estimate according to equation (4); (4) In the formula, for k Posterior state estimation at time 10:00. This is a high-pressure water level measurement signal that is subject to noise pollution.

[0028] (3) Calculate the output of the Kalman filter according to equation (5). k Estimated value of the actual signal of the high-pressure transformer water level at any given time ; (5) (4) Update the posterior state estimation error covariance matrix according to equation (6); (6) In the formula, for k The covariance matrix of the posterior state estimation error at time step [time]. Represents a unit array.

[0029] Specifically, The external loop employs a linear active disturbance rejection controller, which includes an extended state observer, a tracking differentiator, and a linear feedback unit.

[0030] Specifically, Reference Figure 2 The extended state observer is used to obtain real-time estimates of the system state and total disturbance. The input signals of the extended state observer include... k Constantly send water flow command signals and high-pressure water level estimation signal The output includes: high-pressure water level observation signal. High-pressure water level change rate observation signal and total disturbance observation signal The extended state observer adopts a third-order integral cascade structure, which can be expressed as Equation (7). (7) This represents the derivative of the high-pressure water level observation signal. The derivative of the observed signal representing the rate of change of water level in the high-pressure transformer. The derivative of the observed signal representing the total disturbance. This represents the state observer gain coefficient. is the cutoff frequency of the extended state observer.

[0031] Specifically, The tracking differentiator, as described in equation (8), is used to arrange the signal transition process: (8) In the formula, The given signal is generated by the tracking differentiator. To track the derivative of the given signal generated by the differentiator; The rate of change signal of the given value is generated by the tracking differentiator. The derivative for tracking the rate of change of a given value formed by the differentiator; This represents the input given value signal. The bandwidth parameter determines the tracking speed.

[0032] Specifically, The linear feedback unit combines the high-pressure water level observation signal with the transition signal used by the tracking differentiator to arrange the signal according to equation (9) to form the controller output command signal.

[0033] (9) In the formula, Formed by a linear active disturbance rejection controller k The control command signal at time +1 Indicates the cutoff frequency of the control loop. This is the gain coefficient, selected based on the static gain of the high-voltage transformer.

[0034] Specifically, Linear active disturbance rejection controller control command signal This signal is superimposed on the steam flow signal D to form a feedwater flow command signal. .

Claims

1. A method for controlling the water level of the high-pressure boiler in a gas turbine unit's waste heat boiler, characterized in that, include: S1. The high-pressure water level measurement signal is processed using the Kalman filtering algorithm; The high-pressure transformer water flow command signal and the high-pressure transformer water level measurement signal are used as inputs, and the estimated value of the true high-pressure transformer water level signal is output to eliminate measurement noise interference. S2. The estimated value of the actual water level signal of the high-pressure transformer and the setpoint signal of the high-pressure transformer water level are input to the linear active disturbance rejection controller; the linear active disturbance rejection controller generates a basic control command signal based on the estimated value of the actual water level signal of the high-pressure transformer, the setpoint signal and the internal system state observation results; S3. The steam flow signal is used as a feedforward signal and superimposed with the basic control command signal generated by the linear active disturbance rejection controller to form the final high-pressure water supply flow command signal, which is used to control the water supply regulating valve.

2. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 1, characterized in that, The Kalman filtering algorithm in step S1 includes a prediction step and a measurement update step: The prediction step is used to calculate the prior estimate of the system state and the prior estimate error covariance; The measurement update step is used to calculate the Kalman gain, and based on the Kalman gain and the new measurement value, update the posterior estimate of the system state, and output the estimated value of the true signal of the high-pressure water level.

3. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 2, characterized in that, The prediction steps specifically include: (1) Calculate the prior state estimate of the high-pressure water level according to equation (1); （1） In the formula, express k Prior estimates of the system state variables at time t. express k Posterior estimate of the system state variables at time -1 The state matrix of the high-pressure water level object model, The control matrix for the high-pressure water level object model. express k The input quantity of the system at time -1, i.e. k -1 is the high-pressure water supply flow command signal; (2) Calculate the prior state estimation error covariance matrix according to equation (2); （2） In the formula, for k The prior state estimation error covariance matrix at time step 1. for k -1 time posterior state estimation error covariance matrix Let be the covariance matrix of the process noise.

4. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 3, characterized in that, The measurement update steps of the Kalman filtering algorithm are as follows: (1) Calculate the Kalman gain according to equation (3); （3） In the formula, for k Kalman gain at time step The covariance matrix of the measurement noise; (2) Calculate the posterior state estimate according to equation (4); （4） In the formula, for k Posterior state estimation at time 10:

00. The high-pressure water level measurement signal is affected by noise pollution; (3) Calculate the output of the Kalman filter according to equation (5). k Estimated value of the actual signal of the high-pressure transformer water level at any given time ; （5） (4) Update the posterior state estimation error covariance matrix according to equation (6); （6） In the formula, for k The covariance matrix of the posterior state estimation error at time step [time]. Represents a unit array.

5. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 4, characterized in that, The linear active disturbance rejection controller includes an extended state observer, a tracking differentiator, and a linear feedback unit.

6. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 5, characterized in that, The extended state observer is used to obtain real-time estimates of the system state and total disturbance. The input signals of the extended state observer include... k Constantly send water flow command signals and high-pressure water level estimation signal The output includes: high-pressure water level observation signal. High-pressure water level change rate observation signal and total disturbance observation signal The extended state observer adopts a third-order integral cascade structure, expressed as Equation (7). （7） This represents the derivative of the high-pressure water level observation signal. The derivative of the observed signal representing the rate of change of water level in the high-pressure transformer. The derivative of the observed signal representing the total disturbance. This represents the state observer gain coefficient. is the cutoff frequency of the extended state observer.

7. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 6, characterized in that, The tracking differentiator is used to schedule the signal transition process as follows: （8） In the formula, The given signal is generated by the tracking differentiator. To track the derivative of the given signal generated by the differentiator; The rate of change signal of the given value is generated by the tracking differentiator. The derivative for tracking the rate of change of a given value formed by the differentiator; This represents the input given value signal. The bandwidth parameter determines the tracking speed.

8. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 7, characterized in that, The linear feedback unit combines the high-pressure water level observation signal and the transition signal used by the tracking differentiator to arrange the signal linearly according to equation (9) to form the controller output command signal; （9） In the formula, Formed by a linear active disturbance rejection controller k The control command signal at time +1 Indicates the cutoff frequency of the control loop. This is the gain coefficient, selected based on the static gain of the high-voltage transformer.

9. The method for controlling the water level of the high-pressure boiler in a gas turbine unit waste heat boiler according to claim 8, characterized in that, Linear active disturbance rejection controller control command signal This signal is superimposed on the steam flow signal D to form a feedwater flow command signal. .

10. A high-pressure boiler water level control system for a gas turbine unit waste heat boiler for implementing the method as described in any one of claims 1 to 9, characterized in that, The system includes: The Kalman filtering module processes the high-pressure transformer water level measurement signal using the Kalman filtering algorithm; it takes the high-pressure transformer water flow command signal and the high-pressure transformer water level measurement signal as inputs and outputs an estimated value of the true high-pressure transformer water level signal to eliminate measurement noise interference. The linear active disturbance rejection control module inputs the estimated value of the actual high-pressure transformer water level signal and the high-pressure transformer water level setpoint signal to the linear active disturbance rejection controller; the linear active disturbance rejection controller generates basic control command signals based on the estimated value of the actual high-pressure transformer water level signal, the setpoint signal, and internal system state observation results. The signal superposition module uses the steam flow signal as a feedforward signal and superimposes it with the basic control command signal generated by the linear active disturbance rejection controller to form the final high-pressure water supply flow command signal, which is used to control the water supply regulating valve.

Images