A Model Predictive Control Method and Device for Supercritical Units Based on Error Adaptation and Extended State Kalman Filtering

Abstract

This application discloses a model predictive control method and device for supercritical units based on error adaptation and extended state Kalman filtering, belonging to the field of supercritical units. The method includes: analyzing the multivariable coupling characteristics of the supercritical unit and determining multivariable control requirements; identifying and discretizing the dynamic system of the supercritical unit based on the transfer function method to determine the prediction model of the model predictive controller; designing an extended state Kalman filter model to perform state estimation of the supercritical unit and designing an error adaptation mechanism algorithm model; integrating the error adaptation mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller. The error adaptation and extended state Kalman filtering model for supercritical units in this application can improve the dynamic response speed, tracking accuracy, and anti-interference capability of the supercritical unit control system.

Description

Technical Field

[0001] This application belongs to the field of supercritical units, and in particular relates to a model predictive control method and device for supercritical units based on error adaptation and extended state Kalman filtering. Background Technology

[0002] Supercritical coal-fired power plants remain important in modern energy systems due to their high thermal efficiency and low emissions. However, their boiler-turbine systems exhibit multivariable coupling, large inertia, and significant nonlinearity, posing challenges to control system design. Traditional control strategies, such as PID control, perform reasonably well in simple environments, but often fail to meet the requirements of flexibility and stability when supercritical units face complex operating conditions due to time delays, parameter variations, and external disturbances.

[0003] Model predictive control (MPC), by predicting the future behavior of a system and optimizing control actions, can effectively handle multivariable coupling and nonlinear problems, making it a hot topic in supercritical unit control research. However, MPC performance is highly dependent on model accuracy and real-time optimization capabilities. Traditional state-space modeling and filtering methods are difficult to fully adapt to the nonlinear dynamics of supercritical units, resulting in insufficient control accuracy and anti-disturbance capabilities. Extended State Kalman Filter (ESKF) enhances its ability to handle nonlinearity and uncertainty by introducing disturbance state estimation, but its fixed-parameter design still has room for improvement in dynamic error adjustment and multivariable coordination. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the related art. To this end, this application proposes a model predictive control method and device for supercritical units based on error adaptation and extended state Kalman filtering. By introducing a dynamic error adaptation mechanism and exponential smoothing technology, combined with the disturbance estimation capability of extended state Kalman filtering, the dynamic response speed, tracking accuracy, and anti-interference capability of the supercritical unit control system can be improved.

[0005] In a first aspect, this application provides a model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering, the method comprising:

[0006] S1. Analyze the multivariable coupling characteristics of supercritical units and determine the multivariable control requirements;

[0007] S2. Based on the transfer function method, the dynamic system of the supercritical unit is identified and discretized to determine the prediction model of the model predictive controller.

[0008] S3. Design an extended state Kalman filter model to estimate the state of the supercritical unit, and design an error adaptive mechanism algorithm model.

[0009] S4. Integrate the error adaptive mechanism algorithm model with the extended state Kalman filter model to optimize the prediction model of the model prediction controller.

[0010] The model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering proposed in this application analyzes the multivariate coupling characteristics of supercritical units to determine multivariate control requirements. Based on the transfer function method, the dynamic system of the supercritical unit is identified and discretized to determine the prediction model of the model predictive controller. An extended state Kalman filter model is designed to estimate the state of the supercritical unit. An error adaptation mechanism algorithm model is designed, and the error adaptation mechanism algorithm model and the extended state Kalman filter model are integrated to optimize the prediction model of the model predictive controller. This method can address the challenges of multivariate coupling, large inertia, and nonlinearity by utilizing dynamic error adjustment and disturbance estimation to improve the tracking accuracy and anti-interference capability of main steam pressure, load, and temperature. Simultaneously, it coordinates the control of inlet valve opening, coal feed, and water feed to improve robustness and stability under wide load operation. Furthermore, through state estimation, error adaptation, and predictive optimization, multivariate coordinated control is achieved, effectively mitigating multivariate coupling interference in supercritical units.

[0011] According to one embodiment of this application, the control objects of the supercritical unit include: the opening degree of the steam inlet valve, the coal feed rate and the water feed rate, as well as the main steam pressure, the unit load and the outlet temperature of the water-cooled wall header, which are used to realize the prediction function of the prediction model of the model prediction controller.

[0012] According to one embodiment of this application, implementing the prediction function of the prediction model of the model prediction controller includes:

[0013] A dynamic model of the supercritical unit is established, and the dynamic system of the supercritical unit is identified using the transfer function method. The identified dynamic model is then discretized into difference equations for adaptation to the model predictive controller.

[0014] According to one embodiment of this application, the design of the extended state Kalman filter model for state estimation of the supercritical unit includes:

[0015] Based on the discretization model, a state-space form is constructed.

[0016] According to one embodiment of this application, the discretized model dynamically adjusts the prediction error weights through real-time error analysis and combines exponential smoothing technology to design an error adaptive mechanism algorithm model.

[0017] According to one embodiment of this application, the step of integrating the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model prediction controller includes:

[0018] Predicting future outputs based on difference equations.

[0019] According to one embodiment of this application, the step of integrating the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model prediction controller includes:

[0020] The error adaptive mechanism algorithm model dynamically optimizes control priorities through real-time error analysis to ensure the system's response to load changes;

[0021] The model predictive controller based on the extended state Kalman filter model uses its predictive capabilities to regulate the main steam valve opening, coal feed rate, and water feed rate.

[0022] The state estimation results of the extended state Kalman filter model are input into the model prediction controller;

[0023] The predictive control function of the model predictive controller can identify external disturbances and internal uncertainties in advance and optimize the control strategy.

[0024] Secondly, this application provides a model predictive control device for supercritical units based on error adaptation and extended state Kalman filtering, the device comprising:

[0025] The first processing module is used to analyze the multivariable coupling characteristics of supercritical units and determine multivariable control requirements.

[0026] The second processing module is used to identify and discretize the dynamic system of the supercritical unit based on the transfer function method, and determine the prediction model of the model predictive controller.

[0027] The third processing module is used to design an extended state Kalman filter model to perform state estimation of the supercritical unit and to design an error adaptive mechanism algorithm model.

[0028] The fourth processing module is used to integrate the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model prediction controller.

[0029] The supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering proposed in this application analyzes the multivariate coupling characteristics of the supercritical unit to determine the multivariate control requirements. Based on the transfer function method, the dynamic system of the supercritical unit is identified and discretized to determine the prediction model of the model predictive controller. An extended state Kalman filter model is designed to estimate the state of the supercritical unit. An error adaptation mechanism algorithm model is designed, and the error adaptation mechanism algorithm model and the extended state Kalman filter model are integrated to optimize the prediction model of the model predictive controller. This device addresses the challenges of multivariate coupling, large inertia, and nonlinearity by utilizing dynamic error adjustment and disturbance estimation to improve the tracking accuracy and anti-interference capability of main steam pressure, load, and temperature. Simultaneously, it coordinates the control of inlet valve opening, coal feed, and water feed to improve robustness and stability under wide load operation. Furthermore, through state estimation, error adaptation, and predictive optimization, multivariate coordinated control is achieved, effectively mitigating the multivariate coupling interference of the supercritical unit.

[0030] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering as described in the first aspect above.

[0031] Fourthly, this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering as described in the first aspect above.

[0032] The above-described one or more technical solutions in the embodiments of this application have at least one of the following technical effects:

[0033] By analyzing the multivariate coupling characteristics of supercritical units, multivariate control requirements are determined. Based on the transfer function method, the dynamic system of the supercritical unit is identified and discretized to determine the prediction model of the model predictive controller. An extended state Kalman filter model is designed for state estimation of the supercritical unit. An error adaptive mechanism algorithm model is designed, and the error adaptive mechanism algorithm model and the extended state Kalman filter model are integrated to optimize the prediction model of the model predictive controller. This approach addresses the challenges of multivariate coupling, large inertia, and nonlinearity by utilizing dynamic error adjustment and disturbance estimation to improve the tracking accuracy and anti-interference capability of main steam pressure, load, and temperature. Simultaneously, it coordinates the control of inlet valve opening, coal feed, and water feed, improving robustness and stability under wide load operation. Furthermore, through state estimation, error adaptation, and predictive optimization, multivariate coordinated control is achieved, effectively mitigating multivariate coupling interference in supercritical units.

[0034] Furthermore, based on the error adaptive mechanism, the control priority is dynamically optimized through real-time error analysis to ensure the system's rapid response to load changes. The model predictive controller based on extended state Kalman filtering, with its predictive capabilities, can precisely adjust the main steam valve opening, coal feed rate, and water feed rate. This optimized coordination improves the system's control accuracy and dynamic response capability. By inputting the state estimation results of the extended state Kalman filter into the model predictive controller, disturbances caused by multivariate coupling in supercritical units can be effectively suppressed, improving the coordination of main steam pressure, load, and temperature control, thereby enhancing the overall stability of the system. Simultaneously, by inputting the state estimation results of the extended state Kalman filter model into the model predictive controller, and with the error adaptive mechanism combined with exponential smoothing technology, control actions are quickly adjusted when disturbances occur, enhancing the system's anti-interference capability and operational robustness.

[0035] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0036] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0037] Figure 1 This is a model predictive control structure diagram of a supercritical unit based on error adaptation and extended state Kalman filtering provided in the embodiments of this application;

[0038] Figure 2 This is a flowchart illustrating the model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering provided in this application embodiment.

[0039] Figure 3 This is one of the comparison diagrams of the control effects of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application;

[0040] Figure 4 This is the second comparison diagram of the control effect of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application;

[0041] Figure 5 This is one of the comparison diagrams of the control effect of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application under disturbance conditions;

[0042] Figure 6This is the second comparison diagram of the control effect of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application under disturbance conditions;

[0043] Figure 7 This is the third comparison diagram of the control effect of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application under disturbance conditions;

[0044] Figure 8 This is one of the comparison diagrams between the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application and the actual control results of the power plant;

[0045] Figure 9 This is the second comparison diagram between the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application and the actual control results of the power plant;

[0046] Figure 10 This is the third comparison diagram between the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application and the actual control results of the power plant;

[0047] Figure 11 This is a schematic diagram of the structure of the supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering provided in the embodiments of this application;

[0048] Figure 12 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0049] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0050] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0051] The following description, in conjunction with the accompanying drawings, details the supercritical unit model predictive control method based on error adaptive and extended state Kalman filtering, the supercritical unit model predictive control device based on error adaptive and extended state Kalman filtering, the electronic equipment, and the readable storage medium provided in this application, through specific embodiments and application scenarios.

[0052] Among them, the model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering can be applied to the terminal, specifically executed by the hardware or software in the terminal.

[0053] The terminal includes, but is not limited to, portable communication devices such as mobile phones or tablets. It should also be understood that, in some embodiments, the terminal may not be a portable communication device, but rather a desktop computer.

[0054] The supercritical unit model predictive control method based on error adaptive and extended state Kalman filtering provided in this application embodiment can be executed by an electronic device or a functional module or entity within an electronic device that can implement the supercritical unit model predictive control method based on error adaptive and extended state Kalman filtering. The electronic devices mentioned in this application embodiment include, but are not limited to, mobile phones, tablets, computers, cameras, and wearable devices. The following description uses an electronic device as the execution subject to illustrate the supercritical unit model predictive control method based on error adaptive and extended state Kalman filtering provided in this application embodiment.

[0055] like Figure 2 As shown, the predictive control method for supercritical unit models based on error adaptation and extended state Kalman filtering includes steps S1, S2, S3, and S4.

[0056] Step S1: Analyze the multivariable coupling characteristics of the supercritical unit and determine the multivariable control requirements;

[0057] In this step, supercritical units are a highly efficient and environmentally friendly power generation technology, mainly used in thermal power generation, such as... Figure 1 The coal-fired power unit shown.

[0058] Supercritical units can include major equipment such as boilers, steam turbines, and generators.

[0059] This method is applicable to supercritical units of any power, for example, it can be applied to 350MW supercritical units, etc., and is not limited here.

[0060] The multivariable coupling characteristics of supercritical units can include the dynamic correlation characteristics between each control variable and the control objective.

[0061] For example, rapid adjustment of the steam inlet valve opening directly affects the load and pressure, while the slow dynamics of coal feed and water feed indirectly affect the temperature and pressure through combustion and heat transfer.

[0062] Multivariate control requirements can be determined based on the characteristics of multivariate coupling, which refers to how to control the above-mentioned control objects when controlling the changes of the above-mentioned control objectives to the desired values.

[0063] In actual implementation, the difference between fast and slow dynamics in the multivariable coupling characteristics leads to interference between loops. Therefore, it is necessary to clarify the multivariable control requirements and design a control strategy that can balance fast tracking and slow coordination based on the multivariable control requirements, so as to simultaneously meet the requirements of fast response and stability.

[0064] In some embodiments, the control objects of the supercritical unit include: the opening degree of the steam inlet valve, the coal feed rate and the water feed rate, the main steam pressure, the unit load and the outlet temperature of the water-cooled wall header, etc., which are used to realize the prediction function of the prediction model of the model predictive controller (MPC).

[0065] In this embodiment, the controlled object may include control variables and control objectives, wherein, for example... Figure 1 As shown, control variables may include the opening degree of the steam inlet valve, the amount of coal fed and the amount of water fed, etc., and control objectives may include the main steam pressure, the unit load and the outlet temperature of the water-cooled wall header, etc.

[0066] Step S2: Identify and discretize the dynamic system of the supercritical unit based on the transfer function method to determine the prediction model of the model predictive controller;

[0067] In this step, the dynamic system of the supercritical unit can be understood as the multivariable coupling characteristics in step S1, or the multivariable control requirements determined based on the multivariable coupling characteristics.

[0068] The transfer function can be expressed as:

[0069]

[0070] Where a1, a2, ..., an-1 and an are the coefficients of the denominator polynomial; b1, b2, ..., bn-1 and bn are the coefficients of the numerator polynomial; n is the order of the denominator polynomial, and s represents the order of the system.

[0071] Discretization can transform the continuous-time model of the dynamic system of a supercritical unit, obtained from transfer function identification, into a discrete-time model suitable for digital control, thereby optimizing the control strategy.

[0072] In actual implementation, the discrete-time model after discretization can be determined as the prediction model of the model predictive controller. This will be explained in detail in the following embodiments, and will not be elaborated here.

[0073] The predictive model of the model predictive controller is used to predict the future system output and future system input of the supercritical unit.

[0074] Step S3: Design an extended state Kalman filter model to estimate the state of the supercritical unit, and design an error adaptive mechanism algorithm model (EEA).

[0075] In this step, the Extended State Kalman Filter (ESKF) model is used to perform state estimation for the supercritical unit.

[0076] State estimation can include estimating the system state and disturbances of a supercritical unit.

[0077] Among them, the system state is a state vector that records the historical state of the system, captures the large inertia characteristics, and facilitates the prediction of future behavior. It is usually composed of output variables and their delay terms. The disturbance is a disturbance vector that represents the impact of unmodeled dynamics and external disturbances on the system, such as the nonlinearity of the combustion process and the complexity of heat transfer, changes in fuel quality, sudden load changes, and environmental factors.

[0078] In actual operation, ESKF can estimate system state and disturbances to predict outputs at multiple future time steps (such as main steam pressure, load, and temperature), thereby optimizing control inputs (such as valve opening and coal feed rate) and providing accurate initial state conditions for MPC. MPC can then use these initial state conditions as a starting point to predict the future system outputs and inputs of the supercritical unit.

[0079] In some embodiments, the error adaptive mechanism algorithm model can be used to adjust the magnitude of residual correction during the state estimation process of ESKF.

[0080] The residual can be the difference between the current system state and disturbance estimated by ESKF and the actual current system state and disturbance.

[0081] In actual implementation, the residual correction is used to correct the current system state and disturbance estimated by ESKF based on the actual current system state and disturbance.

[0082] In some embodiments, the error adaptive mechanism algorithm model can also be integrated into the objective function, wherein the objective function is a function used to measure the deviation between the future system output of the supercritical unit predicted by the model predictive controller and the expected system output of the supercritical unit, as well as the cost of control actions.

[0083] In actual implementation, the objective function can be optimized and solved based on the optimization control algorithm, thereby obtaining the future system input of the supercritical unit under the condition that it matches the expected system output.

[0084] The future system output or expected system output may include the control objectives of the supercritical unit, and the future system input may include the control variables of the supercritical unit.

[0085] In some embodiments, the optimization control algorithm can be a quadratic programming algorithm.

[0086] Step S4: Integrate the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller.

[0087] In this step, the integrated error adaptive mechanism algorithm model and the extended state Kalman filter model can optimize the accuracy of the future system input of the supercritical unit determined by the model predictive controller.

[0088] In actual implementation, the error adaptive mechanism algorithm model dynamically adjusts the process noise covariance through real-time residual analysis, estimates the extended state (system state and disturbances of supercritical units) using ESKF, and uses ESKF estimation and disturbance correction to predict future output, optimize control increment, balance output deviation and control cost, and improve control accuracy and robustness.

[0089] The supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in this application analyzes the multivariate coupling characteristics of the supercritical unit to determine multivariate control requirements. Based on the transfer function method, the dynamic system of the supercritical unit is identified and discretized to determine the prediction model of the model predictive controller. An extended state Kalman filter model is designed to estimate the state of the supercritical unit. An error adaptation mechanism algorithm model is designed, and the error adaptation mechanism algorithm model and the extended state Kalman filter model are integrated to optimize the prediction model of the model predictive controller. This method can address the challenges of multivariate coupling, large inertia, and nonlinearity by utilizing dynamic error adjustment and disturbance estimation to improve the tracking accuracy and anti-interference capability of main steam pressure, load, and temperature. Simultaneously, it coordinates the control of inlet valve opening, coal feed, and water feed to improve robustness and stability under wide load operation. Furthermore, through state estimation, error adaptation, and predictive optimization, multivariate coordinated control is achieved, effectively mitigating multivariate coupling interference in the supercritical unit.

[0090] The following detailed description of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering is based on specific embodiments.

[0091] In some embodiments, implementing the prediction function of the prediction model of the model prediction controller may include:

[0092] A dynamic model of the supercritical unit is established. The dynamic system of the supercritical unit is identified using the transfer function method. The identified dynamic model is then discretized into difference equations for use in adapting the model predictive controller.

[0093] In this embodiment, based on the historical operating data of the unit, a backpropagation algorithm can be used to establish a dynamic model of the controlled unit system using a BP neural network as the simulation process, with the control variables as inputs and the control target as the output.

[0094] The transfer function method is used to identify the dynamic system of a supercritical unit, and the dynamic model in transfer function form is discretized into difference equations, which serve as the prediction model for MPC to predict future system output. This transforms the continuous-time model into a discrete-time model suitable for digital control, thereby optimizing the control strategy.

[0095] The representation of the transfer function has been explained in the above steps and will not be repeated here.

[0096] In actual implementation, for a multi-input multi-output system, the discretized output is:

[0097]

[0098] Where k is the discrete time step, representing the current sampling time; y i (k) is the value of the i-th output at time k, such as the control target: main steam pressure, unit load, and water-cooled wall header outlet temperature; y i (kp) is the value of the i-th output at discrete time step kp; u j (kq) is the value of the j-th input variable at discrete time step kq, such as the steam inlet valve opening, coal feed rate, and water feed rate; a i,p and b ij,q These are model coefficients, reflecting the dynamic characteristics of the system; n i The i-th output represents the autoregressive order; nu represents the total number of input variables; m i,j Let n denote the delay order of the j-th input to the i-th output; where n i m i,j i, j, k, q and p are all integers greater than or equal to 0.

[0099] According to the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application, a dynamic model of the supercritical unit is established, the dynamic system of the supercritical unit is identified by the transfer function method, and the identified dynamic model is discretized into difference equations for use in adapting the model predictive controller, so that the model predictive controller can predict the future system input and output of the supercritical unit based on the difference equations.

[0100] In some embodiments, designing an extended state Kalman filter model for state estimation of a supercritical unit may include:

[0101] Based on the discretization model, a state-space form is constructed.

[0102] In this embodiment, through the state-space form, i.e. the state-space model, ESKF can estimate the system state and disturbances, and use them to predict the outputs (such as main steam pressure, load, and temperature) at multiple future time steps, thereby optimizing the control inputs (such as valve opening and coal feed rate) and providing accurate initial state conditions for MPC.

[0103] By introducing disturbance terms, the state-space model can capture unmodeled dynamics and external disturbances (such as sudden load changes and fuel quality variations), thereby improving the stability of supercritical unit systems under wide load operation and disturbance conditions.

[0104] In some embodiments, a perturbation term can be introduced based on the above discretized model to construct the state-space form as follows:

[0105] x e,k+1 =A e x e,k +B e u k +ξ k

[0106] y k =C e x e,k

[0107] Where k is the discrete time step, representing the current sampling time; x e,k This is a state vector containing system states, such as output and its delay terms; u k It is the input vector, including valve opening degree, coal feed rate, and water feed rate; ξ k It is a disturbance, capturing unmodeled dynamics and external disturbances; y k It is the output vector, including main steam pressure, load, and temperature; A e B e and C e It is the state-space matrix, obtained by transforming the discretized model.

[0108] In the context of supercritical units, the state-space model describes the dynamic behavior of the supercritical unit system, such as how control targets like main steam pressure and load change with control inputs like valve opening and coal feed rate.

[0109] Therefore, based on the state-space model, ESKF estimation of system state and disturbance can be designed to improve the reliability of the system state and disturbance estimation results of ESKF.

[0110] Among them, the system state is a state vector that records the historical state of the system, captures the large inertia characteristics, and facilitates the prediction of future behavior. It is usually composed of output variables and their delay terms. The disturbance is a disturbance vector that represents the impact of unmodeled dynamics and external disturbances on the system, such as the nonlinearity of the combustion process and the complexity of heat transfer, changes in fuel quality, sudden load changes, and environmental factors.

[0111] In actual implementation, ESKF estimates not only the system state but also disturbances. The prediction steps of ESKF are as follows:

[0112]

[0113] P k|k-1 =AP k-1|k-1 A T +Q k

[0114] Where k is the discrete time step, representing the current sampling time. The prediction of the extended state, i.e. in, It is a prediction of the system state. It is the prediction of the perturbation; A is the extended state transition matrix. Among them, A x It is the original state transition matrix, describing the dynamics of the control objective as it changes with historical values. x and I w It is an identity matrix, with dimensions matching the state and the perturbation, respectively, assuming the perturbation changes slowly; B x This is the original input matrix, defining the effect of control commands on the control objective; 0 indicates no direct impact on the disturbance. k|k-1 It is an extended prediction covariance matrix that reflects the uncertainty of state and disturbance predictions; Among them, Q x It is the system state noise covariance, Q w It is the perturbation noise covariance, which reflects the randomness of fluctuations.

[0115] In actual execution, ESKF predicts the system state and disturbance at the current sampling time. After collecting the actual system state and disturbance at the current sampling time, ESKF also updates the predicted system state and disturbance at the current sampling time based on the actual system state and disturbance at the current sampling time.

[0116] In some embodiments, the ESKF update steps are as follows:

[0117]

[0118]

[0119]

[0120] P k|k =(IK k C)P k|k-1

[0121] Where k is the discrete time step, representing the current sampling time, and e k It is the measurement residual, representing the difference between the actual measured value and the expected output based on the predicted state; C = [C x [0] is the extended output matrix, where C x Mapping the system state to the output is defined as how historical values ​​of the control target are mapped to measured values; 0 indicates that disturbances do not directly affect the measurement. k It is the innovation covariance, which measures the statistical uncertainty of the measurement residual, and is equal to the covariance of the predicted output plus the measurement noise covariance; P k|k-1 R is the extended prediction covariance matrix, representing the uncertainty in state and disturbance predictions; R is the measurement noise covariance matrix, representing the statistical characteristics of sensor errors; K k β is the Kalman gain matrix, which determines the weight of the measurement residual on the state update based on sensor accuracy and prediction error; β is the adaptive weighting coefficient, which adjusts the magnitude of the residual correction and can be determined based on an error adaptive mechanism; fal(e k (α, δ) is an adaptive correction function that depends on the residual e k Adaptive factor α and threshold δ are used to enhance robustness; I is the identity matrix, whose dimensions match the expanded state to ensure the correctness of matrix operations in covariance updates; For the prediction of the updated extended state; P represents the prediction of the extended state. k|k It is an extended update of the covariance matrix, representing The uncertainty reflects the reduced uncertainty in the revised control target estimate.

[0122] The supercritical unit model predictive control method based on error adaptive and extended state Kalman filtering provided in the embodiments of this application constructs a state-space form based on a discretized model, which can capture unmodeled dynamics and external disturbances, and improve the stability of the supercritical unit system under wide load operation and disturbance conditions; and can design ESKF to estimate the system state and disturbances based on the state-space model, thereby improving the accuracy of ESKF in predicting the system state and disturbances at the current sampling time.

[0123] In some embodiments, the discretized model dynamically adjusts the prediction error weights through real-time error analysis and combines exponential smoothing techniques to design an error adaptive mechanism algorithm model.

[0124] In this embodiment, the core of the error adaptive mechanism lies in the real-time adjustment of dynamic weights, and the adjustment rules are as follows:

[0125] Q k (k)=S q ·(1+α1|e(k)|)

[0126]

[0127] Where k is the discrete time step, representing the current sampling time, and Q... k (k) is the unsmoothed dynamic covariance; S q α is the basic covariance; e(k) is the real-time error, i.e., the measurement residual in the ESKF above; α1 and β1 are the amplification factor and the smoothing factor, respectively. It is the smoothed dynamic covariance.

[0128] It should be noted that the amplification factor α1 is greater than 1 and the smoothing factor β1 is greater than 0 and less than 1. As can be seen from the above adjustment rules, when the amplification factor α1 and / or the smoothing factor β1 are set to 1, it is considered that the error adaptive mechanism algorithm model has not added the amplification factor α1 and / or the smoothing factor β1. Therefore, the amplification factor α1 and the smoothing factor β1 are not set to 1.

[0129] In actual implementation, α1 and β1 can be set based on the experimental test results. For example, if the experimental test results show that α1 and β1 are 1.7 and 0.8 respectively, and the adjustment effect is better, then α1 and β1 can be set to 1.7 and 0.8 respectively. Of course, in other embodiments, α1 and β1 can also be set to other values, which is not limited here.

[0130] Among them, the prediction error weight refers to the dynamic adjustment part of the prediction error covariance. It can affect the Kalman gain by adjusting the process noise covariance, so that it trusts the measurement value more when the prediction error is large, and makes the prediction more suitable for disturbances.

[0131] Exponential smoothing is used to improve stability and reduce the impact of noise while preserving the trend of the disturbance.

[0132] According to the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application, the discretized model dynamically adjusts the prediction error weights through real-time error analysis, which makes the ESKF prediction results more suitable for disturbances. In addition, by combining the error adaptation mechanism algorithm model with exponential smoothing technology, the prediction results can be reduced from drastic fluctuations caused by measurement noise and instantaneous sensor errors while preserving the disturbance trend. This reduces the occurrence of changes in prediction direction due to single errors, improves stability, and reduces the impact of noise.

[0133] In some embodiments, integrating the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller may include:

[0134] Predicting future outputs based on difference equations.

[0135] In this embodiment, the Model Predictive Controller (MPC) uses difference equations to generate future prediction outputs, and the prediction output process is shown in the following equation:

[0136]

[0137]

[0138] ΔU=[Δu k|k ,…,Δu k+Nc-1|k ] T

[0139] Where k is the discrete time step, representing the current sampling time, and y k+1|k This represents the output predicted at time k+i from time k; u k+i-q|k This represents the input instruction predicted at time k+iq at time k; a p and b q These are the coefficients of the difference equation, describing the system dynamics; n and m are the orders of the difference equation. ΔU is the state vector, containing historical inputs and outputs; Np is the control increment vector; Nc is the prediction window; F and Φ are prediction matrices, representing the free response and forced response, respectively; Y is the prediction output vector, with a length equal to the prediction window Np.

[0140] According to the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application, the future output is predicted based on the difference equation. This allows the ESKF-based MPC to use the difference equation, initialize through historical state, predict the future output in a single step, iteratively calculate the multi-step predicted output, and combine it with the disturbance estimation of ESKF for correction to obtain the predicted output vector, thereby adapting to external disturbances and improving prediction accuracy.

[0141] In some embodiments, integrating the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller may include:

[0142] The error adaptive mechanism algorithm model dynamically optimizes control priorities through real-time error analysis to ensure the system's response to load changes;

[0143] The model predictive controller based on the extended state Kalman filter model can regulate the opening of the main steam valve, the coal feed rate and the water feed rate by means of its predictive ability.

[0144] The state estimation results of the extended state Kalman filter model are input into the model predictive controller;

[0145] The predictive control function of the model predictive controller can identify external disturbances and internal uncertainties in advance and optimize the control strategy.

[0146] In this embodiment, the error adaptive mechanism dynamically adjusts the process noise covariance through real-time residual analysis, estimates the extended state using ESKF, and uses ESKF estimation and disturbance correction to predict future output, thereby optimizing control increment, balancing output deviation and control cost, and improving control accuracy and robustness.

[0147] In actual implementation, after MPC realizes the future predicted output based on the difference equation, it can solve the objective function in the form of quadratic programming based on the future predicted output to obtain the future system input. The objective function is constructed by output deviation term and control cost term to measure the deviation between the predicted output and the target value and the cost of control actions.

[0148] The quadratic programming form is used to optimize the solution of the objective function.

[0149] In some embodiments, the objective function and the quadratic programming form are as follows:

[0150] J = (YY) r ) T S q (YY r )+ΔU T S r ΔU

[0151]

[0152] ω=Φ T S q Φ+r w S r

[0153]

[0154] Where k is the discrete time step, representing the current sampling time; J is the state vector; J is the objective function, measuring the prediction output deviation and control cost; Y is the prediction output vector, with a length of the prediction window Np; Y r The reference output vector is the set target value; ΔU is the control increment vector with a length equal to the control window Nc; ω is the quadratic coefficient matrix; FF is the linear coefficient matrix; r w These are scalar weights that adjust the relative importance of the control increment; where ω and FF are calculated from the prediction matrix, state, and weights; F and Φ are the prediction matrices; S q It is the output error weight matrix, S r To control the incremental weight matrix.

[0155] In the supercritical unit scenario, ESKF estimates the state initialization, predicts the output and combines it with disturbance correction, uses quadratic programming to solve for the optimal control increment vector, adjusts the control input, and achieves load tracking, pressure and temperature stability.

[0156] In some embodiments, control constraints can be set to meet the safe operation requirements of supercritical units:

[0157] Δu min ≤Δu(k)≤Δu max

[0158]

[0159] Where k is the discrete time step, representing the current sampling time; the control quantity refers to the control input u(k) and its increment Δu(k); in supercritical units, the control input usually includes valve opening (%), coal feed rate (kg / s), and water feed rate (kg / s), used to regulate load, pressure, and temperature; the control increment Δu (Δu(k) = u(k) - u(k-1)) represents the change in the control input, and the constraints ensure that the control action is smooth and within a safe range.

[0160] The constraint values ​​set can be determined based on the physical limits of the equipment and the requirements for operational stability.

[0161] The above constraints can be optimized by embedding MPC through quadratic programming to ensure that the control increment ΔU and control input u(k) are within a safe range, reducing equipment overload or excessive action. At the same time, in conjunction with the disturbance estimation and error adaptive mechanism of ESKF, the stability and robustness of control are improved.

[0162] In actual implementation, after integrating the error adaptive mechanism, the objective function becomes:

[0163]

[0164] Where J is the objective function, measuring the prediction output deviation and control cost; k is the discrete time step, representing the current sampling time; y is the prediction output vector, with a length of the prediction window Np; y r It is the reference output vector, i.e., the set target value; ΔU is the control increment vector, with a length equal to the control window Nc; S r It is the control increment weight matrix, Q K (i) is the dynamic covariance at time i.

[0165] like Figure 3 and Figure 4 As shown, this invention compares and analyzes the command disturbance response performance of EEA-ESKF-MPC, EA-ESKF-MPC, ESKF-MPC, and traditional MPC in terms of main steam pressure, unit load, and header temperature. Among them, EEA-ESKF-MPC is the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering of this application; EA-ESKF-MPC is the case where the amplification factor and smoothing factor in the error adaptation mechanism algorithm model are both set to 1, that is, the error adaptation mechanism algorithm model does not add amplification factor and smoothing factor; ESKF-MPC is a supercritical unit model predictive control method based only on extended state Kalman filtering.

[0166] Traditional MPC has a slow response and large deviation; ESKF-MPC improves stability through perturbation estimation; EA-ESKF-MPC is sensitive to adjustment and fluctuates significantly because it does not add amplification and smoothing factors; in contrast, EEA-ESKF-MPC combines the perturbation estimation of ESKF and the adaptive adjustment advantages of EA-ESKF-MPC, and optimizes through amplification and smoothing factors. Its response curve may recover to the set value more quickly and with smaller fluctuations, showing higher anti-interference accuracy and stability.

[0167] like Figure 5 , Figure 6 and Figure 7 As shown, in order to evaluate the performance of different control strategies during the load ramping process, the present invention conducted load variation tests within the operating range of 70MW to 320MW.

[0168] exist Figure 5 Compared to other methods, the pressure response of EEA-ESKF-MPC significantly reduces overshoot, and its curve exhibits smooth transition characteristics, demonstrating its excellent ability to maintain system stability under dynamic changes.

[0169] Figure 6 This further highlights the advantages of EEA-ESKF-MPC in load tracking. Its response curve closely matches the set value with minimal deviation, maintaining precise control even when the load is rapidly increasing, demonstrating its high adaptability to changes in the set value.

[0170] same, Figure 7 This demonstrates the superiority of the EEA-ESKF-MPC in temperature control, with its response reducing fluctuations, maintaining stricter stability, and ensuring reliable adjustment of temperature parameters during ramp-up.

[0171] This comparison shows that EEA-ESKF-MPC, through comprehensive optimization of dynamic error adjustment, amplification factor and smoothing technology, achieves excellent control accuracy, fast response and stability in ramp-up tests. It can effectively cope with the complex dynamic requirements of multivariable coupling and large inertia of supercritical units, and provides an efficient and reliable control scheme for variable load operation.

[0172] like Figure 8 , Figure 9 and Figure 10 As shown, this invention compares and analyzes the tracking performance of EEA-ESKF-MPC and existing power plant controllers based on the operating data of a 350MW supercritical unit in terms of main steam pressure, unit load and header temperature.

[0173] Depend on Figure 8 , Figure 9 and Figure 10 It can be seen that the existing PID controller in the power plant may exhibit large fluctuations and tracking lag under real operating conditions; while the response curve of EEA-ESKF-MPC is closer to the set value and has smaller fluctuations, demonstrating its potential to improve coordination and stability in practical applications.

[0174] According to the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in the embodiments of this application, the control priority is dynamically optimized through real-time error analysis based on the error adaptation mechanism to ensure the system's rapid response to load changes. Furthermore, the model predictive controller based on extended state Kalman filtering, with its predictive capability, can achieve precise adjustment of the main steam valve opening, coal feed rate, and water feed rate. This optimized coordination can improve the system's control accuracy and dynamic response capability. By inputting the state estimation results of the extended state Kalman filter into the model predictive controller, disturbances caused by multivariate coupling in the supercritical unit can be effectively suppressed, improving the coordination of main steam pressure, load, and temperature control, thereby enhancing the overall stability of the system. Simultaneously, the error adaptation mechanism, combined with exponential smoothing technology, rapidly adjusts control actions when disturbances occur, enhancing the system's anti-interference capability and operational robustness.

[0175] The supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering provided in this application can be executed by a supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering. This application uses the execution of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering by the supercritical unit model predictive control device as an example to illustrate the supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering provided in this application.

[0176] This application also provides a model predictive control device for supercritical units based on error adaptation and extended state Kalman filtering.

[0177] like Figure 11 As shown, the supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering includes: a first processing module 1110, a second processing module 1120, a third processing module 1130 and a fourth processing module 1140.

[0178] The first processing module 1110 is used to analyze the multivariable coupling characteristics of the supercritical unit and determine the multivariable control requirements.

[0179] The second processing module 1120 is used to identify and discretize the dynamic system of the supercritical unit based on the transfer function method, and determine the prediction model of the model predictive controller.

[0180] The third processing module 1130 is used to design an extended state Kalman filter model to perform state estimation of the supercritical unit and to design an error adaptive mechanism algorithm model.

[0181] The fourth processing module 1140 is used to integrate the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller.

[0182] The supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering provided in this application analyzes the multivariate coupling characteristics of the supercritical unit to determine multivariate control requirements. It identifies and discretizes the dynamic system of the supercritical unit using the transfer function method to determine the predictive model of the model predictive controller. An extended state Kalman filter model is designed to estimate the state of the supercritical unit. An error adaptation mechanism algorithm model is designed, and the error adaptation mechanism algorithm model and the extended state Kalman filter model are integrated to optimize the predictive model of the model predictive controller. This device addresses the challenges of multivariate coupling, large inertia, and nonlinearity by utilizing dynamic error adjustment and disturbance estimation to improve the tracking accuracy and anti-interference capability of main steam pressure, load, and temperature. Simultaneously, it coordinates the control of inlet valve opening, coal feed, and water feed to improve robustness and stability under wide load operation. Furthermore, through state estimation, error adaptation, and predictive optimization, it achieves multivariate coordinated control, effectively mitigating multivariate coupling interference in the supercritical unit.

[0183] In some embodiments, the third processing module 1130 can also be used for:

[0184] Based on the discretization model, a state-space form is constructed.

[0185] In some embodiments, the fourth processing module 1140 can also be used for:

[0186] Predicting future outputs based on difference equations.

[0187] In some embodiments, the fourth processing module 1140 can also be used for:

[0188] The error adaptive mechanism algorithm model dynamically optimizes control priorities through real-time error analysis to ensure the system's response to load changes;

[0189] The model predictive controller based on the extended state Kalman filter model can regulate the opening of the main steam valve, the coal feed rate and the water feed rate by means of its predictive ability.

[0190] The state estimation results of the extended state Kalman filter model are input into the model predictive controller;

[0191] The predictive control function of the model predictive controller can identify external disturbances and internal uncertainties in advance and optimize the control strategy.

[0192] In some embodiments, the device may further include a fifth processing module for:

[0193] A dynamic model of a supercritical unit is established. The dynamic system of the supercritical unit is identified using the transfer function method. The identified dynamic model is then discretized into difference equations to adapt to the model predictive controller and realize the predictive function of the predictive model of the model predictive controller.

[0194] The supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering in this application embodiment can be an electronic device or a component within an electronic device, such as an integrated circuit or chip. The electronic device can be a terminal or other devices besides a terminal. For example, the electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc. It can also be a server, network attached storage (NAS), personal computer (PC), television (TV), ATM, or self-service machine, etc. This application embodiment does not specifically limit the specific implementation.

[0195] The supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit it.

[0196] The supercritical unit model predictive control device based on error adaptation and extended state Kalman filtering provided in this application embodiment can achieve... Figure 2 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.

[0197] In some embodiments, such as Figure 12As shown, this application embodiment also provides an electronic device 1200, including a processor 1201, a memory 1202, and a computer program stored in the memory 1202 and executable on the processor 1201. When the program is executed by the processor 1201, it implements the various processes of the above-described embodiment of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0198] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0199] This application also provides a non-transitory computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described embodiments of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering, and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0200] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0201] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described predictive control method for supercritical unit models based on error adaptation and extended state Kalman filtering.

[0202] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0203] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described embodiments of the supercritical unit model predictive control method based on error adaptation and extended state Kalman filtering, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0204] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0205] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0206] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the related technology, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0207] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0208] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0209] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A supercritical unit model predictive control method based on error self-adaption and extended state Kalman filtering, characterized in that, include: S1. Analyze the multivariable coupling characteristics of supercritical units and determine the multivariable control requirements; S2. Based on the transfer function method, the dynamic system of the supercritical unit is identified and discretized to determine the prediction model of the model predictive controller. S3. Design an extended state Kalman filter model to estimate the state of the supercritical unit, and design an error adaptive mechanism algorithm model. S4. Integrate the error adaptive mechanism algorithm model with the extended state Kalman filter model to optimize the prediction model of the model prediction controller; The integration of the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller includes: the error adaptive mechanism algorithm model dynamically optimizes the control priority through real-time error analysis to ensure the system's response to load changes; The model predictive controller based on the extended state Kalman filter model uses its predictive capabilities to regulate the opening of the main steam valve, the coal feed rate, and the water feed rate; the state estimation results of the extended state Kalman filter model are input into the model predictive controller; the predictive control function of the model predictive controller can identify external disturbances and internal uncertainties in advance and optimize the control strategy.

2. The model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering according to claim 1, characterized in that, The control objects of the supercritical unit include: the opening degree of the steam inlet valve, the coal feed rate and the water feed rate, as well as the main steam pressure, the unit load and the outlet temperature of the water-cooled wall header, which are used to realize the prediction function of the prediction model of the model prediction controller.

3. The model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering according to claim 2, characterized in that, The implementation of the prediction function of the prediction model of the model prediction controller includes: A dynamic model of the supercritical unit is established, and the dynamic system of the supercritical unit is identified using the transfer function method. The identified dynamic model is then discretized into difference equations for adaptation to the model predictive controller.

4. The model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering according to claim 1, characterized in that, The design of the extended state Kalman filter model is used to perform state estimation for the supercritical unit, including: Based on the discretization model, a state-space form is constructed.

5. The model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering according to claim 4, characterized in that, The discretized model dynamically adjusts the prediction error weights through real-time error analysis and combines exponential smoothing technology to design an error adaptive mechanism algorithm model.

6. The model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering according to any one of claims 1-5, characterized in that, The integration of the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller includes: Predicting future outputs based on difference equations.

7. A model predictive control device for supercritical units based on error adaptation and extended state Kalman filtering, characterized in that, include: The first processing module is used to analyze the multivariable coupling characteristics of supercritical units and determine multivariable control requirements. The second processing module is used to identify and discretize the dynamic system of the supercritical unit based on the transfer function method, and determine the prediction model of the model predictive controller. The third processing module is used to design an extended state Kalman filter model to perform state estimation of the supercritical unit and to design an error adaptive mechanism algorithm model. The fourth processing module is used to integrate the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller; the integration of the error adaptive mechanism algorithm model and the extended state Kalman filter model to optimize the prediction model of the model predictive controller includes: the error adaptive mechanism algorithm model dynamically optimizes the control priority through real-time error analysis to ensure the system's response to load changes; The model predictive controller based on the extended state Kalman filter model uses its predictive capabilities to regulate the opening of the main steam valve, the coal feed rate, and the water feed rate; the state estimation results of the extended state Kalman filter model are input into the model predictive controller; the predictive control function of the model predictive controller can identify external disturbances and internal uncertainties in advance and optimize the control strategy.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the model predictive control method for supercritical units based on error adaptation and extended state Kalman filtering as described in any one of claims 1-6.

Images