A composite control method for gas turbine speed based on SMC-PID

By using an LPV model based on weighted fuzzy membership functions and sliding mode-PID composite control, the problems of large overshoot and insufficient anti-interference capability in the speed control of gas turbine power turbines were solved, and high-precision and stable control of gas turbines under variable load conditions was achieved.

CN123064579APending Publication Date: 2026-11-10DATANG INT POWER GENERATION CO LTD BEIJING GAOJING THERMAL POWER BRANCH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610969844.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2026-01-31
Filing Date
2026-07-01
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Traditional PID controllers in gas turbine power turbine speed control suffer from problems such as large overshoot, slow response, and insufficient anti-interference capability under variable load conditions. Existing interference compensation methods are difficult to meet the requirements of real-time performance and simplicity. Furthermore, LPV models in gas turbine modeling suffer from steady-state point bending and local fitting deviation.

Method used

A gas turbine model is established using an LPV model based on weighted fuzzy membership functions. By combining sliding mode control and PID control, an adaptive parameter sliding mode control law is designed. The gas turbine speed is controlled through a dynamic sliding surface model and fuel flow compensation, and an SMC-PID composite controller is constructed.

Benefits of technology

It achieves precise control of the turbine speed of gas turbine power under variable load conditions, shortens the settling time, reduces overshoot, suppresses sliding mode chattering, improves robustness and disturbance rejection capability, and meets stringent stability requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123064579A_ABST
    Figure CN123064579A_ABST
Patent Text Reader

Abstract

This disclosure discloses a composite control method for gas turbine power turbine speed based on SMC-PID, comprising: establishing a cluster of state-space models at multiple steady-state points of the gas turbine; selecting scheduling parameters; employing a scheduling method based on weighted fuzzy membership functions; determining scheduling weights by calculating the correlation between scheduling parameters and steady-state points; establishing an LPV model for the gas turbine; discretizing the classical PID control law with speed as the control objective to obtain a PID control model; constructing a dynamic sliding surface model that integrates power tracking error and its changing trend based on the established LPV model; obtaining the basic fuel quantity from the PID control model; obtaining the fuel flow compensation value from the dynamic sliding surface model; and adding the basic fuel quantity and the fuel flow compensation value to obtain the final fuel flow control signal to control the gas turbine speed. This embodiment shortens the settling time, reduces overshoot, and suppresses sliding mode chattering under typical operating conditions; it controls speed deviation and achieves rapid steady-state, exhibiting superior robustness.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of gas turbine control technology, and in particular to a composite control method for gas turbine power speed based on SMC-PID. Background Technology

[0002] Gas turbines, with their high power density, rapid start-up and shutdown, and low pollutant emissions, have become core power equipment in fields such as power system peak shaving and ship propulsion. The accuracy of turbine speed control directly determines the stability of the unit's output power, fuel economy, and operational safety. However, gas turbine systems exhibit strong nonlinearity, multivariable coupling, and time-varying parameter characteristics. Furthermore, they are susceptible to uncertainties such as fuel flow fluctuations, intake parameter disturbances, and changes in mechanical losses during operation. This leads to traditional PID controllers generally suffering from large overshoot, slow response, and insufficient anti-interference capabilities under varying load conditions, making it difficult to meet the high-precision control requirements under complex operating conditions. To address these issues, researchers have conducted research and proposed improvement schemes from two aspects: optimizing control strategies and improving modeling accuracy, and strengthening disturbance compensation. (1) Optimization of control strategy: Improving control flexibility by combining intelligent algorithms with traditional PID. In terms of optimizing control strategies, Li et al. proposed a hybrid control method based on the Brain Emotion Learning Intelligent Controller (BELBIC) and adaptive multi-input multi-output feedback, which integrates nonlinear feedback linearization technology to improve the dynamic response speed of the gas turbine across the entire load range; Lqbal et al. designed a fuzzy gain scheduling PID controller for a dual-shaft gas turbine controlled by a Woodward governor, using fuzzy logic to dynamically adjust PID parameters and improve the load disturbance adaptability under grid interaction scenarios; Yang et al. optimized fuzzy PID parameters by improving the particle swarm optimization-cuckoo search hybrid algorithm, which significantly reduced the speed fluctuation of the micro gas turbine under varying operating conditions. However, such methods mostly rely on accurate system models or complex parameter optimization processes, and there is still room for improvement in control robustness under model uncertainty or strong disturbance scenarios.

[0003] (2) Improving modeling accuracy and strengthening disturbance compensation: High-precision models are constructed and active disturbance compensation mechanisms are designed to suppress disturbance effects. Hou et al. constructed a high-precision nonlinear model of a gas turbine based on fuzzy modeling methods of entropy clustering and subspace identification, providing a reliable basis for controller design; Lin et al. integrated the start-up model and component characteristic diagram model to establish a nominal nonlinear model of a micro gas turbine and designed a nonlinear active disturbance rejection controller (NADRC) to achieve active suppression of load disturbances; Wen et al. proposed a genetic algorithm to optimize a neural network controller for the complex sudden load of all-electric marine gas turbines, reducing speed overshoot and settling time through online parameter self-tuning. However, existing disturbance compensation methods mostly rely on extended state observers or neural networks, which have problems of observation lag or high computational complexity, making it difficult to meet the dual requirements of real-time performance and simplicity for power turbine speed control.

[0004] Linear variable parameter (LPV) models have been widely used in airborne modeling because they retain the form of linear system models and can describe dynamic characteristics within a variable parameter range through real-time measurable scheduling parameters. Currently, the commonly used scheduling methods for LPV models are linear interpolation and polynomial fitting. While linear interpolation is simple to operate, it can easily lead to model bending at steady-state points, causing instability during steady-state switching. Polynomial fitting (affine-dependent LPV models) only guarantees the minimum L2 norm throughout the entire scheduling process. If the error at some steady-state points is large, it can cause local fitting to tilt towards the error points, severely affecting the overall modeling accuracy and making it difficult to adapt to the modeling requirements of complex gas turbine operating conditions.

[0005] Furthermore, sliding mode control (SMC), as a robust nonlinear control method, possesses advantages such as simple structure, fast response, and strong adaptability to model uncertainties through its disturbance observer (SMDO). It can estimate the total system disturbance in real time and achieve feedforward compensation, demonstrating potential in handling nonlinear, highly disturbed system control problems. However, the application of composite control schemes combining sliding mode control and PID control to balance the engineering practicality of PID with the disturbance suppression capabilities of SMC in the field of gas turbine power turbine speed control still needs improvement. A mature technical solution that can effectively address the synergistic improvement of speed control accuracy and anti-interference performance under varying load conditions has not yet been developed.

[0006] Therefore, it is necessary to improve one or more of the problems existing in the above-mentioned related technical solutions.

[0007] It should be noted that this section is intended to provide background or context for the technical solutions of this disclosure as set forth in the claims. The description herein does not constitute an admission that it is prior art simply because it is included in this section. Summary of the Invention

[0008] The purpose of this invention is to provide a composite control method for the turbine speed of gas turbine power based on SMC-PID, thereby overcoming, at least to some extent, one or more problems caused by the limitations and defects of related technologies.

[0009] This invention provides a composite control method for gas turbine speed based on SMC-PID, comprising: S1. Establish a cluster of state-space models at multiple steady-state points of the gas turbine, select scheduling parameters, adopt a scheduling method based on weighted fuzzy membership function, determine the scheduling weight by calculating the correlation between scheduling parameters and steady-state points, and establish the LPV model of the gas turbine. S2, with rotational speed as the control target, the classic PID control law is discretized to obtain the PID control model; S3. Based on the established LPV model, a dynamic sliding surface model that integrates power tracking error and its changing trend is constructed. A sliding mode control law containing a saturation function is designed, and adaptive parameters are designed for model updating. S4. The basic fuel quantity is obtained according to the PID control model, and the fuel flow compensation value is obtained according to the dynamic sliding surface model. The basic fuel quantity and the fuel flow compensation value are added together to obtain the final fuel flow control signal. The gas turbine operating state is adjusted according to the control signal, thereby controlling the gas turbine speed.

[0010] In this invention, in S1, the scheduling parameters include: low-pressure compressor speed, high-pressure compressor speed, and power turbine speed.

[0011] In this invention, in S1, the weighted fuzzy membership function adopts a normal distribution curve.

[0012] In this invention, in S1, the scheduling weight is determined by calculating the generalized distance between the scheduling parameters and the steady-state point.

[0013] In this invention, the adaptive parameters in S3 are designed based on the Lyapunov stability condition.

[0014] In this invention, the PID control model in S2 is as follows:

[0015] in, u(t) To control the quantity, e(k) For deviation, T The sampling period is k The sampling sequence number represents the power turbine speed error. , n pr To the desired power turbine speed, n p This refers to the actual turbine speed of the gas turbine.

[0016] In this invention, the dynamic sliding surface model in S3 is as follows:

[0017] in, For dynamic sliding surface, c 1. c 2 is a constant coefficient. It is power tracking error.

[0018] In this invention, the sliding mode control law in S3 is as follows:

[0019] in, k The sliding mode control gain determines the speed at which the system state approaches the sliding surface; It is a saturation function.

[0020] The technical solution provided by this invention may include the following beneficial effects: This invention presents a composite control method for gas turbine power speed based on SMC-PID. By optimizing linear variable parameter (LPV) modeling and a sliding mode-PID composite control strategy, this invention overcomes the core challenges of existing gas turbine power turbine speed control under varying load conditions. In terms of modeling, it solves the problems of steady-state point bending and local fitting deviation in traditional methods, fully reproducing the nonlinear and time-varying parameter characteristics across the entire load range, providing a high-fidelity foundation for control design. Regarding control performance, it integrates the advantages of PID steady-state accuracy and sliding mode dynamic response, combined with dynamic weight allocation, to shorten settling time, reduce overshoot, and suppress sliding mode chattering under typical operating conditions, meeting stringent stability requirements. In terms of disturbance rejection, it relies on a sliding mode disturbance observer to achieve fast and accurate disturbance estimation and active suppression, still controlling speed deviation and achieving rapid steady-state control under complex operating conditions, demonstrating superior robustness. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0022] Figure 1 A flowchart illustrating the SMC-PID-based composite control method for gas turbine power speed in an exemplary embodiment of this disclosure is shown. Figure 2 A cross-sectional view of the gas turbine in an exemplary embodiment of this disclosure is shown; Figure 3 This diagram illustrates the principle of the linearization model in an exemplary embodiment of this disclosure. Figure 4 This diagram illustrates the PID gas turbine control structure in an exemplary embodiment of this disclosure. Figure 5 This illustrates a PID-SMC control system in an exemplary embodiment of this disclosure; Figure 6 The simulation results of load shedding in an exemplary embodiment of this disclosure are shown; Figure 7 This illustrates the effect of emergency loading simulation in an exemplary embodiment of this disclosure. Detailed Implementation

[0023] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0024] Furthermore, the accompanying drawings are merely illustrative diagrams of embodiments of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities.

[0025] This example implementation provides a composite control method for gas turbine speed based on SMC-PID. Please refer to [link / reference]. Figure 1 This method may include: S1-S4, as follows: S1. Establish a cluster of state-space models at multiple steady-state points of the gas turbine, select scheduling parameters, adopt a scheduling method based on weighted fuzzy membership function, determine the scheduling weight by calculating the correlation between scheduling parameters and steady-state points, and establish the LPV model of the gas turbine. S2, with rotational speed (e.g., power turbine speed) as the control target, the classic PID control law is discretized to obtain the PID control model; S3. Based on the established LPV model, a dynamic sliding surface model that integrates power tracking error and its changing trend is constructed. A sliding mode control law containing a saturation function is designed, and adaptive parameters are designed for model updating. S4. The basic fuel quantity is obtained according to the PID control model, and the fuel flow compensation value is obtained according to the dynamic sliding surface model. The basic fuel quantity and the fuel flow compensation value are added together to obtain the final fuel flow control signal. The gas turbine operating state is adjusted according to the control signal, thereby controlling the gas turbine speed.

[0026] In this embodiment, the present invention overcomes the core challenge of existing gas turbine power turbine speed control under variable load conditions by optimizing linear variable parameter (LPV) modeling and sliding mode-PID composite control strategy: In terms of modeling, it solves the problems of steady-state point bending and local fitting deviation in traditional methods, and can completely reproduce the nonlinearity and time-varying parameter characteristics of the entire load range, providing a high-fidelity foundation for control design; in terms of control performance, it integrates the advantages of PID steady-state accuracy and sliding mode dynamic response, and with dynamic weight allocation, it shortens the settling time, reduces overshoot, and suppresses sliding mode chattering under typical operating conditions, meeting stringent stability requirements; in terms of disturbance rejection, it relies on the sliding mode disturbance observer to achieve fast and accurate disturbance estimation and active suppression, and can still control speed deviation and achieve fast steady-state under complex operating conditions, with better robustness.

[0027] The specific process of each step in the above embodiments will be described below.

[0028] First, the gas turbine LPV model will be explained.

[0029] The working fluid in a gas turbine is a continuously flowing mixture of gases. A compressor generates high-pressure gas that drives a high-speed impeller. Fuel heat serves as the total energy of the unit, which is converted into effective work by the turbine. It is a typical internal combustion engine. The gas turbine studied in this application is as follows: Figure 2 As shown, according to the airflow process, a gas turbine can be divided into components such as a low-pressure compressor, a high-pressure compressor, a combustion chamber, a high-pressure turbine, a low-pressure turbine, and a power turbine. The compressor source of gas is the external atmospheric environment, which is compressed in stages to form high-pressure gas. The combustion chamber receives the high-temperature, high-pressure air from the compressor outlet and mixes it with the injected fuel for efficient combustion, generating combustion chamber exhaust with even higher temperature and pressure. The turbine receives the combustion chamber exhaust and expands to do work, driving the power turbine to rotate at high speed and output effective work, thereby realizing the conversion of the chemical energy of the fuel into mechanical work, driving external loads or outputting electrical energy. This application first establishes a component-level model of the gas turbine based on the component characteristics and aerodynamic thermodynamics.

[0030] It should be noted that, Figure 2 In the diagram, ② represents the inlet section of the gas turbine. It is the inlet and outlet cross section of the low-pressure compressor. ③ is the inlet section of the high-pressure compressor, and ④ is the outlet section of the high-pressure gas. It is the cross-section of the combustion chamber inlet. This is the inlet section of the high-pressure turbine. It is the outlet section of the high-pressure turbine. This is the inlet section of the low-pressure turbine. HPT is the high-pressure turbine, and IPT is the low-pressure turbine.

[0031] However, the nonlinear component-level mechanistic model of the gas turbine is difficult to meet real-time requirements, making it unsuitable for direct use in real-time systems. The linear variable parameter (LPV) model relies on real-time measurable and bounded scheduling parameters, preserving the form of a linear system model while describing all dynamic characteristics across the entire variable parameter range along the parameter trajectory. It has already achieved widespread application in airborne modeling. Therefore, this application, based on this principle and to facilitate the design of the controller described later, establishes an LPV model of the gas turbine on top of the component-level model.

[0032] The nonlinear model of the gas turbine at a certain steady-state point is shown below: (1) Expanding equation (1) using Taylor and ignoring higher-order terms, we obtain the following form: (2) In the formula, , At the same time, define The state-space model can be obtained as follows: (3) However, because only the first-order terms of the nonlinear Taylor expansion are retained, and higher-order terms are ignored, the state-space model is only valid within the bounded neighborhood of the steady-state point, such as... Figure 3 As shown, the linear model cannot simulate the dynamic characteristics of the original model under large-scale state changes. This makes it impossible for the established linear model to simulate the operating characteristics of the gas turbine under large-scale changes such as acceleration and deceleration. To solve this problem, this application first establishes a set of state-space models at multiple steady-state points and combines these state-space model matrix sets with measurable time-varying scheduling parameters θ. A dependency relationship is established between the state-space model and the scheduling parameter θ. When θ changes in real time along the nonlinear trajectory, the state-space model changes accordingly and can always simulate the operating characteristics of the original nonlinear model, thus possessing the ability to approximate the large-scale dynamics of the nonlinear model. At this point, the linear model can be transformed into the following LPV model form: (4) In the formula, It is a state variable, including the low-pressure compressor speed. High-pressure compressor speed Power turbine speed , These are input variables, including fuel flow rate, load torque, and the inlet guide vane angles of the high and low pressure compressors. These are output variables, including key cross-sectional parameters, turbine shaft speeds, output power, and other parameters. A It is a system matrix. These are scheduling parameters, including the power turbine speed and the low-pressure compressor speed. B It is the input matrix.C It is the output matrix. D It is a feedforward matrix.

[0033] The matrices are represented as follows:

[0034] in, Let be the weighting coefficient corresponding to the i-th steady-state point. , .

[0035]

[0036] in, It is the generalized distance between the current state of the gas turbine and the i-th steady-state point.

[0037] The core of the LPV model is to obtain the fitting relationship between the coefficients of each matrix in the state-space model and the scheduling parameters. When the scheduling parameters are at the linearized steady-state point, the fitted state-space model is approximately the linearized state-space model obtained at that point. Between two steady-state points, a completely new state-space model will be fitted to simulate the small-range dynamic characteristics of the model at the current operating point. Currently, the commonly used scheduling methods are linear interpolation and polynomial fitting. Linear interpolation is relatively simple, but it can cause the LPV model to bend at the steady-state point, easily leading to instability during steady-state point switching. Polynomial fitting, i.e., affine dependency LPV model, only guarantees the minimum L2 norm of the entire scheduling process. If the error is large at some steady-state points, it will cause the local fitting to tilt towards the error points, which will seriously affect the global modeling accuracy.

[0038] To address the aforementioned issues, this application integrates the multinomial fitting scheduling approach with the operational characteristics of gas turbines, employing an LPV model scheduling method based on a weighted fuzzy membership function. The scheduling weights are calculated based on the correlation between scheduling parameters and the steady-state point. The state-space model determines its influence weight based on the generalized distance to the steady-state point, aligning with the actual operational characteristics of gas turbines. Furthermore, this application selects a normally distributed curve as the membership function.

[0039] The design process of the SMC-PID composite controller will be explained next.

[0040] 1. PID control loop design The PID algorithm is a closed-loop control algorithm. It is simple in principle and easy to implement. The classic control law is: (5) In the formula, u(t) To control the quantity, e(t) For deviation, , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively.

[0041] In order to implement the PID algorithm using a computer, formula (5) must be discretized. This application aims to control the speed of the power turbine, and discretizes the classical PID control law to obtain the following PID control model: (6) In the formula, T The sampling period is k For sampling sequence number, For the power turbine speed error, n pr To the desired power turbine speed, n p This represents the actual turbine speed of the gas turbine. The PID control loop designed in this application is as follows: Figure 4 As shown.

[0042] 2. Slip Mode Controller Design Based on the LPV model, this application designs a dynamic sliding surface for fusion error and its variation trend, expressed as follows: (7) In the formula, It is power tracking error. For the desired power, This represents the actual output power. c 1. c 2 is a constant coefficient.

[0043] To ensure the system state converges quickly to the sliding surface and remains stable, a sliding mode control law is designed: (8) In the formula, k The sliding mode control gain determines the speed at which the system state approaches the sliding surface; Let saturation function be defined as: (9) The boundary layer thickness is the saturation function, and its function is to create a linear transition region near the sliding surface, avoiding abrupt changes in the control input and thus reducing chattering. s is the dynamic sliding surface, which is dimensionless.

[0044] To enhance the sliding mode controller's adaptability to system uncertainties, the following adaptive parameter update law is designed: (10) In the formula, , , , The parameters are for the control model designed based on the Lyapunov stability condition.

[0045] 3. SMC-PID Composite Controller Based on the PID control loop and sliding mode controller, an SMC-PID composite controller is obtained, and the results are as follows: Figure 5 As shown. Figure 5 middle, T orque,ref Indicates the desired output torque. T orque This represents the actual output torque of the gas turbine. ΔW f The compensation control quantity output by SMC. W f This indicates the final control quantity.

[0046] Figure 5 In this system, the control system focuses on the turbine speed of the gas turbine, employing a composite control strategy combining PID control and sliding mode control. The PID controller takes the deviation between the turbine speed reference value and the actual speed as input, and outputs fuel flow rate as the control signal to achieve main speed regulation and tracking. Simultaneously, the deviation between the applied disturbance torque reference value and the actual output torque of the gas turbine is input to the sliding mode controller, which calculates a fuel flow rate compensation value. This compensation value is used to suppress the adverse effects of load disturbances and other factors on the system. Finally, the basic fuel flow rate output from the PID controller is added to the compensation value output from the sliding mode controller to obtain the final fuel flow rate control signal, which is then input to the gas turbine. The gas turbine adjusts its operating state according to this fuel flow rate, thereby achieving precise control of the turbine speed. Simultaneously, the gas turbine can effectively resist load disturbances, ensuring stable system operation.

[0047] The gas turbine model established in this application has a maximum power of 30MW, and the steady-state speed of the power turbine is constant at 4000rpm, meaning the maximum torque is approximately 7.26×10⁻⁶. 4 (N·m), to verify the effectiveness of load control, simulations were conducted under load disturbance conditions, specifically for load shedding and emergency loading scenarios of the gas turbine. The results are as follows: Figure 6-7 As shown in the figure, PID represents the response curve of the power turbine speed using only a PID controller; SMC-PID represents the response curve of the power turbine speed using the composite controller proposed in this application. n p This refers to the normalized power turbine speed.

[0048] from Figure 6 As can be seen from a, within the simulation time domain, the gas turbine operates under rated steady-state conditions for the first 4 seconds. At this time, the gas turbine output torque matches the expected external torque, the system has no deviation, and the controller has no output.t At 4s, a negative step of 30% of the rated torque is applied to simulate the load shedding process of the gas turbine. Simultaneously, random noise with a maximum amplitude of 2% is applied to the desired torque throughout the simulation time to simulate load fluctuations during actual gas turbine operation. Figure 6 (b) It can be seen that, in t =4 When load shedding occurs, the power turbine speed shows a sudden surge due to the sudden drop in load torque. Under the traditional PID control strategy, the speed overshoot is significant, and the speed fluctuates within a certain amplitude due to random noise disturbances in the subsequent steady state stage, exhibiting disturbance rejection instability. However, the SMC-PID composite control strategy effectively suppresses the speed surge by outputting torque error compensation in real time through the sliding mode controller. The speed overshoot is reduced by about 11% compared to the traditional PID control, and the speed fluctuation amplitude is lower in the steady state stage, fully demonstrating the superiority of composite control under load shedding conditions.

[0049] from Figure 7 (a) shows that the gas turbine maintains stable operation at 0.7 operating condition for the first 4 seconds, and the actual output torque of the gas turbine matches the expected external torque. At t=4s, an emergency load is applied, and the external load suddenly increases to the maximum load. At the same time, noise with a maximum amplitude of 2% is applied to the expected torque throughout the entire simulation time. Figure 7 (b) It can be seen that when the torque suddenly increases at t=4s, the power turbine speed drops instantly due to the sudden increase in load torque. With traditional PID control, the speed drop is large, reaching 8%, and the speed fluctuates significantly under load disturbance in steady state. In contrast, the speed drop of SMC-PID control is 6.5%, which is about 18% lower than that of traditional PID. It is less affected by load fluctuation in steady state, has stronger anti-interference ability, and has better overall control effect.

[0050] In summary, the beneficial effects of this application include: (1) Break through the limitations of traditional modeling methods and build a high-precision dynamic model: solve the problems of bending of the steady-state point of the LPV model caused by linear interpolation and excessive local fitting deviation of the polynomial fitting method. By optimizing the scheduling parameters and modeling logic, an LPV model that can accurately reflect the strong nonlinearity, multivariable coupling and time-varying parameter characteristics of the gas turbine is established, providing a reliable dynamic characteristic description basis for controller design and improving the adaptability of the model to the full range of operating conditions.

[0051] (2) Integrating the advantages of multiple control methods to improve the control performance under varying operating conditions: Overcoming the shortcomings of traditional PID controllers such as large overshoot and slow response, as well as the shortcomings of single sliding mode control which is prone to chattering and intelligent optimization PID which relies on complex parameter optimization, the robustness of sliding mode control (SMC) is combined with the engineering practicality of PID control to design a composite control structure that has both fast response and steady-state accuracy, so as to achieve accurate tracking of the power turbine speed under varying load conditions.

[0052] (3) Enhance interference suppression capability and ensure system operation stability: In response to uncertain interference such as fuel flow fluctuation and intake parameter disturbance, the real-time interference estimation and feedforward compensation function of the integrated sliding mode interference observer (SMDO) is used to solve the problems of observation lag and high computational complexity of the existing interference compensation method, improve the controller's ability to suppress model uncertainty and external disturbance, and ensure the stable operation of the unit under complex interference scenarios.

[0053] (4) Achieve engineering adaptation of control scheme and expand the value of technology application: While meeting the requirements of high precision and strong robustness control, simplify the calculation process of control algorithm and the difficulty of parameter tuning, avoid over-reliance on complex intelligent algorithms or high precision sensors, form a control scheme with simple structure and easy engineering implementation, and provide reliable speed control technology support for gas turbines in fields such as power system peak shaving and ship propulsion.

[0054] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention 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 the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A composite control method for gas turbine speed based on SMC-PID, characterized in that, include: S1. Establish a cluster of state-space models at multiple steady-state points of the gas turbine, select scheduling parameters, adopt a scheduling method based on weighted fuzzy membership function, determine the scheduling weight by calculating the correlation between scheduling parameters and steady-state points, and establish the LPV model of the gas turbine. S2, with rotational speed as the control target, the classic PID control law is discretized to obtain the PID control model; S3. Based on the established LPV model, a dynamic sliding surface model that integrates power tracking error and its changing trend is constructed. A sliding mode control law containing a saturation function is designed, and adaptive parameters are designed for model updating. S4. The basic fuel quantity is obtained according to the PID control model, and the fuel flow compensation value is obtained according to the dynamic sliding surface model. The basic fuel quantity and the fuel flow compensation value are added together to obtain the final fuel flow control signal. The gas turbine operating state is adjusted according to the control signal, thereby controlling the gas turbine speed.

2. The SMC-PID-based composite control method for gas turbine power speed according to claim 1, characterized in that, In S1, the scheduling parameters include: low-pressure compressor speed, high-pressure compressor speed, and power turbine speed.

3. The SMC-PID-based composite control method for gas turbine power speed according to claim 1, characterized in that, In S1, the weighted fuzzy membership function adopts a normal distribution curve.

4. The SMC-PID-based composite control method for gas turbine power speed according to claim 1, characterized in that, In S1, the scheduling weights are determined by calculating the generalized distance between the scheduling parameters and the steady-state point.

5. The SMC-PID-based composite control method for gas turbine power speed according to claim 1, characterized in that, In S3, the adaptive parameters are designed based on the Lyapunov stability condition.

6. The SMC-PID-based composite control method for gas turbine power speed according to claim 1, characterized in that, In S2, the PID control model is as follows: in, u(t) To control the quantity, e(k) For deviation, T The sampling period is k The sampling sequence number represents the power turbine speed error. , n pr To the desired power turbine speed, n p This refers to the actual turbine speed of the gas turbine.

7. The SMC-PID-based composite control method for gas turbine power speed according to claim 6, characterized in that, In S3, the dynamic sliding surface model is as follows: in, For dynamic sliding surface, c 1. c 2 is a constant coefficient. It is power tracking error.

8. The SMC-PID-based composite control method for gas turbine power speed according to claim 7, characterized in that, In S3, the sliding mode control law is as follows: in, k The sliding mode control gain determines the speed at which the system state approaches the sliding surface; It is a saturation function.