A disturbance-free switching control method and system for aircraft engine systems

Abstract

A disturbance-free switching control method and system for an aero-engine system, which relates to the field of aero-engine control technology. The control method includes establishing a switching system model of a control system based on multiple working points of the aero-engine; dividing the aero-engine system into a DoS activation mode and a DoS sleep mode based on a DoS attack model with limited frequency and duration according to whether the aero-engine system is attacked by DoS, and establishing a DoS attack interval conversion signal; constructing an elastic observer model and an elastic disturbance-free switching controller model based on the switching system model and the DoS attack interval conversion signal, and bringing them into the switching system model to obtain a closed-loop system model; calculating the elastic observer gain and the elastic disturbance-free switching controller gain based on the time-dependent switching signal of the closed-loop system model. The control signal bumps of the aero-engine system caused by switching and DoS attacks can be reduced or eliminated, thereby solving the disturbance-free switching control problem of the switching system under DoS attacks.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engine control, and in particular to a disturbance-free switching control method and system for an aircraft engine system. Background Art

[0002] Switching system control technology has a wide range of applications in aerospace, mechanical systems, power systems, industrial processes and other fields, and is receiving more and more attention. With the improvement of system performance requirements, the improvement of transient performance has also become a major research hotspot under the premise of ensuring the steady-state performance of the closed-loop system. Due to the switching of the gain of the elastic disturbance-free switching controller, the control input of the switching system has a jump at the moment of gain switching, which will reduce the transient performance of the aircraft engine system and even bring safety hazards. Disturbance-free switching provides a solution to this problem. Its purpose is to reduce or even eliminate the bumps of the control signal through the comprehensive design of components such as elastic observers and elastic disturbance-free switching controllers, so that the switching system receives a smoother input signal, thereby improving the transient performance of the system.

[0003] With the mutual penetration and integration of cyber-physical system technology and switching system technology, the dependence of aircraft engine control systems on the network is increasing. However, the openness and vulnerability of the network have brought new security challenges to aircraft engine control systems. In particular, when attacked by malicious adversaries, the system may not receive or receive incorrect measurement signals and control signals, which will affect the real-time control of the aircraft engine system. As a typical network attack, DoS (Denial of Service) attacks are easily launched by malicious attackers because they do not need to obtain data such as the model of the attacked system. When the measurement component and control component of the aircraft engine system are attacked by DoS, the elastic non-disturbance switching controller cannot receive the measurement signal from the sensor. The traditional elastic non-disturbance switching controller sets the measurement signal to 0 at this time, which will increase the discontinuity of the control signal. How to design a safe and effective control method that can not only deal with the stability problems caused by DoS attacks and ensure the convergence of the closed-loop system, but also reduce the control signal bumps caused by switching and DoS attacks and achieve non-disturbance switching control is an important issue in the current research of aircraft engine control systems. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes a multi-mode disturbance-free switching control method and system for an aircraft engine control system.

[0005] The technical concept of the present invention is: different elastic observer gains and elastic non-disturbance switching controller gains are used in the DoS attack activation interval and the dormant interval, and the weak convergence or divergence of the DoS activation interval is compensated by the stronger convergence of the system in the DoS dormant interval, thereby ensuring the safety of the closed-loop system. At the same time, the elastic non-disturbance switching controller gain is limited to a constant control gain through non-disturbance constraints, thereby reducing the control signal turbulence caused by switching and DoS attacks, and realizing non-disturbance switching control of the aircraft engine speed regulation control system under DoS attacks.

[0006] The technical solution of the present invention is as follows:

[0007] S1: linearizing the aircraft engine control system into multiple subsystems based on multiple operating points of the aircraft engine, and establishing a switching system model of the control system according to the linearized subsystems;

[0008] Specifically based on multiple working points of the aircraft engine, such as Mach number, flight altitude, etc., the aircraft engine control system is linearized into multiple subsystems, thereby establishing a switching system model of the multi-modal aircraft engine control system:

[0009] (2.1),

[0010] in, is an unpredictable system state. is the derivative of the system state, is the control input, is the measured output, , and Respectively dimension, Peacekeeping dimensional real vector space, is the switching signal, is the number of subsystems in the switching system. season ( ), indicating the subsystems are activated, among which and are two adjacent switching moments, , and is the system matrix of appropriate dimension.

[0011] S2: Based on the duration and frequency of the DoS attack, the aircraft engine system is divided into DoS active mode and DoS dormant mode to determine whether it is under DoS attack, and a DoS attack interval conversion signal is established;

[0012] S21: Determine the duration and frequency of the DoS attack;

[0013] The energy of DoS attack is usually limited. With the help of switching system control theory, it can be expressed as a form with limited duration and frequency of DoS attack. and Respectively represent time intervals The DoS activation interval and DoS sleep interval on and Respectively represent time intervals The total length of the DoS activation interval and the DoS sleep interval, that is, =t- ; Indicates time interval The number of DoS attacks on the

[0014] DoS attack duration: There is a constant in the interval and For all the time satisfy

[0015] (2.2),

[0016] DoS attack frequency: There is a constant in the interval and For all the time satisfy

[0017] (2.3),

[0018] S22: Establishing a DoS attack interval conversion signal based on the DoS attack model ;

[0019] in, Indicates that the system is under DoS attack, that is, in the DoS activation range. Indicates that the system is in DoS sleep range; when hour, is 0 or 1, Indicates the time when the transition between the DoS active interval and the dormant interval occurs. express The next transition moment after that.

[0020] S3: constructing a resilient observer model and a resilient disturbance-free switching controller model based on the switching system model and the DoS attack interval conversion signal;

[0021] Based on the switching signal in the switching system model and DoS attack interval conversion signal , the elastic observer model and the elastic disturbance-free switching controller model are constructed as follows:

[0022] The elastic observer model is:

[0023] (2.4),

[0024] In the formula, is the elastic observer state, is the derivative of the elastic observer state, is the elastic observer output, and Respectively Peacekeeping dimensional real vector space, is the measured output received by the elasticity observer, is the switching signal, is the number of subsystems in the switching system, when season ( ), indicating the The subsystem is activated. It is the DoS attack interval conversion signal. Indicates that the system is in the DoS activation range. Indicates that the system is in DoS sleep range. and is the system matrix of appropriate dimension, is the elastic observer gain associated with the switching signal and the DoS attack interval transition signal, and are the gains of the elastic observer to be designed in the DoS sleep interval and the DoS activation interval respectively.

[0025] In addition, when the DoS attack interval conversion signal is in the sleep interval, that is, When ; When the DoS attack interval conversion signal is in the activation interval, that is, When .

[0026] The elastic bumpless switching controller model is:

[0027] (2.5),

[0028] The following conditions are met:

[0029] (2.6),

[0030] In the formula, is the gain of the elastic disturbance-free switching controller related to the switching signal and the DoS attack interval transition signal, and are the gains of the elastic disturbance-free switching controller to be designed in the DoS sleep interval and the DoS activation interval respectively; at the same time, the gain of the elastic disturbance-free switching controller satisfies the disturbance-free switching constraint condition (2.6). It is the reference elastic disturbance-free switching controller gain selected according to the aircraft engine system parameters. It is a parameter that quantifies the performance of disturbance-free switching. It is related to the switching signal and the DoS attack interval conversion signal. Under the condition of meeting the requirements of system steady-state performance, the smaller its value is, the better the disturbance-free switching effect is. In particular, when the parameter , ( )are selected as the same value, the coefficients on the right side of equation (2.6) are simplified to The disturbance-free switching constraint (2.6) limits the difference between the control input of the system and the smoothed reference signal to a certain range, thereby achieving disturbance-free switching.

[0031] S4: bringing the elastic observer model and the elastic disturbance-free switching controller model into the switching system model, deriving the dynamic equations under the DoS activation mode and the DoS sleep mode respectively, and obtaining a closed-loop system model;

[0032] Combining (2.1), (2.4) and (2.5), the closed-loop system equation is

[0033] (2.7),

[0034] in,

[0035] ,

[0036] ,

[0037] ,

[0038] ,

[0039] represents the expanded-dimensional state consisting of the elastic observer state and the observation error, represents the derivative of the expanded state, represents the system matrix in the closed-loop system, is the system matrix in DoS sleep mode, is the system matrix in DoS activated mode, represents the error between the elastic observer state and the system state.

[0040] S5: Calculate the elastic observer gain and the elastic disturbance-free switching controller gain for the closed-loop system model based on the time-dependent switching signal:

[0041] The specific steps include:

[0042] S51: Define time-dependent switching signal:

[0043] remember For switching signal In the time interval The number of switching times on and For all the time satisfy

[0044] (2.8),

[0045] in, is called the average dwell time, and It's called the vibration world.

[0046] S52: constructing a Lyapunov function in the DoS sleep interval and the DoS activation interval based on the time-dependent switching signal and the DoS attack interval conversion signal;

[0047]

[0048] in, ( ) is the Lyapunov function, , , , and is a positive definite matrix.

[0049] S53: Based on the Lyapunov function constructed in the DoS sleep interval and the DoS activation interval and the switching system stability theory, a gain selection model of the elastic observer and the elastic disturbance-free switching controller is established;

[0050] The specific method is to select a positive definite matrix , , ,matrix , , , , , a normal number , , , , , , and the constant for satisfy

[0051] (2.9),

[0052] (2.10),

[0053] (2.11),

[0054] (2.12),

[0055] (2.13),

[0056] (2.14),

[0057] (2.15),

[0058] (2.16),

[0059] (2.17),

[0060] in, , , .

[0061] Inequality (2.9) ensures that during the DoS sleep period, a single subsystem is running Inequality (2.10) ensures that during the DoS activation period, a single subsystem is running Inequalities (2.11) and (2.12) ensure that the disturbance-free switching condition (2.6) holds. Inequalities (2.13) to (2.16) constrain the increments of the multi-Lyapunov function at the switching time and at the start and end of the DoS attack. , inequality (2.17) and the above conditions together ensure that the closed-loop system Exponential convergence on .

[0062] S54: Based on the gain selection model of the elastic observer and the elastic disturbance-free switching controller, the gain selection model is optimized by using the Schur complement principle and the matrix scaling method, and a linear matrix inequality calculation model of the gain of the elastic observer and the elastic disturbance-free switching controller is established;

[0063] The specific method is to optimize conditions (2.9) to (2.17) to obtain the linear matrix inequality model for the gain calculation of the elastic observer and the elastic disturbance-free switching controller: the selected matrix is , , ,matrix , , , , , a normal number , , , , , , and the constant for satisfy

[0064] (2.18),

[0065] (2.19),

[0066] (2.20),

[0067] (2.21),

[0068] (2.22),

[0069] (2.23),

[0070] (2.24),

[0071] (2.25),

[0072] (2.26),

[0073] in , , .

[0074] S55: obtaining the elastic observer gain and the elastic disturbance-free switching controller gain by solving the linear matrix inequality calculation model of the elastic observer and the elastic disturbance-free switching controller gain;

[0075] Among them, the elastic observer gain is: , ,

[0076] Elastic bumpless switching controller gains: , ,

[0077] In the formula, , , , and is the matrix obtained by solving the matrix inequality system (2.18) to (2.26), which satisfies the following equation:

[0078] , , (2.27),

[0079] , , (2.28),

[0080] When the above inequality is solved, during the DoS sleep period, a single subsystem is running Established, during the DoS activation period, a single subsystem is running Established due to , so the stronger convergence of the system in the DoS sleep interval compensates for the weak convergence or divergence in the DoS activation interval, and the security of the closed-loop system is guaranteed. At the same time, the designed elastic disturbance-free switching controller also satisfies the disturbance-free switching condition (2.6), that is, the elastic disturbance-free switching controller gain obtained is , With reference elastic bumpless switching controller gain As close as possible, so as to limit the difference between the control input of the system and the smooth reference signal to a minimum and achieve disturbance-free switching control.

[0081] A disturbance-free switching control system for an aircraft engine system, comprising:

[0082] Switching system module: used to establish a switching system model of the control system based on multiple operating points of the aircraft engine;

[0083] A DoS attack interval conversion signal module is used to classify whether the aircraft engine system is subjected to a DoS attack into a DoS active mode and a DoS dormant mode based on the duration and frequency of the DoS attack, and to establish DoS attack interval conversion signals for the DoS active mode and the DoS dormant mode;

[0084] An elastic observer and an elastic non-disturbance switching controller building module, used to construct an elastic observer model and an elastic non-disturbance switching controller model based on the switching system model and the DoS attack interval conversion signal;

[0085] A closed-loop system model building module is used to bring the elastic observer model and the elastic disturbance-free switching controller model into the switching system model, derive the dynamic equations under the DoS activation mode and the DoS sleep mode respectively, and obtain a closed-loop system model;

[0086] Gain calculation module: used to calculate the elastic observer gain and the elastic disturbance-free switching controller gain for the closed-loop system model based on the time-dependent switching signal.

[0087] The beneficial effects of the present invention are:

[0088] For the aircraft engine control system modeled by switching system, a switching signal related to the dwell time and a resilient disturbance-free switching controller based on elastic observer are collaboratively designed to stabilize the system under DoS attack. The control turbulence caused by switching and DoS attack is normalized, and the designed resilient disturbance-free switching controller can simultaneously reduce or eliminate the control signal turbulence caused by the two factors. The method of the invention optimizes the solvability condition and can solve the disturbance-free switching control problem of the aircraft engine control system under DoS attack. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 A schematic flow chart of a multi-mode disturbance-free switching control method for an aircraft engine system provided by the present invention;

[0090] Figure 2 is a switching signal used in a specific implementation manner of the present invention;

[0091] Figure 3 This is the disturbance-free switching constraint effect achieved in a specific implementation of the present invention, wherein the illustration is a partial enlarged view of the disturbance-free switching effect of 0-1.2 seconds. DETAILED DESCRIPTION

[0092] The present invention takes the aircraft engine speed regulation as an example to illustrate that according to the design requirements of the control system, two speed increments of the core components of the aircraft engine are selected. , is the system status, the increment of fuel consumption is the input quantity. and Respectively represent the speed of the low-pressure rotor and the high-pressure rotor, Indicates fuel consumption.

[0093] See also Figure 1 As shown, a disturbance-free switching control method for an aircraft engine system specifically includes the following steps:

[0094] S1: linearizing the aircraft engine control system into multiple subsystems based on multiple operating points of the aircraft engine, and establishing a system model of the control system based on the linearized subsystems;

[0095] Specifically based on multiple working points of the aircraft engine, such as Mach number, flight altitude, etc., the aircraft engine control system is linearized into multiple subsystems, thereby establishing a switching system model of the multi-modal aircraft engine control system:

[0096] (2.1),

[0097] in, is an unpredictable system state. is the derivative of the system state, is the control input, is the measured output, , and Respectively dimension, Peacekeeping dimensional real vector space, is the switching signal, is the number of subsystems in the switching system. season ( ), indicating the subsystems are activated, among which and are two adjacent switching moments, , and is the system matrix of appropriate dimension.

[0098] In formula 2.1, take As ,Pick As ,Pick As , the nonlinear part of the aircraft engine control system is linearized, and the following switching system model is obtained:

[0099] ,

[0100] In addition, in this embodiment, , and Using literature ( Hanz Richter. Advanced Control of Turbofan Engines[M]. Springer, 2012 ) The parameters of the 90k engine under FC04 and FC06 conditions are as follows:

[0101] ,

[0102] ,

[0103] S2: Based on the duration and frequency of the DoS attack, the aircraft engine system is divided into DoS active mode and DoS dormant mode to determine whether it is under DoS attack, and a DoS attack interval conversion signal is established;

[0104] S21: Determine the duration and frequency of the DoS attack;

[0105] The energy of DoS attack is usually limited. With the help of switching system control theory, it can be expressed as a form with limited duration and frequency of DoS attack. and Respectively represent time intervals The DoS activation interval and DoS sleep interval on and Respectively represent time intervals The total length of the DoS activation interval and the DoS sleep interval, that is, =t- ; Indicates time interval The number of DoS attacks on the

[0106] DoS attack duration: There is a constant in the interval and For all the time satisfy

[0107] (2.2),

[0108] DoS attack frequency: There is a constant in the interval and For all the time satisfy

[0109] (2.3),

[0110] S22: Establishing a DoS attack interval conversion signal based on the DoS attack model ;

[0111] in, Indicates that the system is under DoS attack, that is, in the DoS activation range. Indicates that the system is in DoS sleep range; when hour, is 0 or 1, Indicates the time when the transition between the DoS active interval and the dormant interval occurs. express The next transition moment after that.

[0112] According to the DoS attack to be resisted, in this implementation scheme, , , , , DoS attack meets and .

[0113] S3: constructing a flexible observer model and a flexible disturbance-free switching controller model based on the switching system model and the DoS attack interval conversion signal;

[0114] Based on the switching signal in the switching system model and DoS attack interval conversion signal , the elastic observer model and the elastic disturbance-free switching controller model are constructed as follows:

[0115] The elastic observer model is:

[0116] (2.4),

[0117] In the formula, is the elastic observer state, is the derivative of the elastic observer state, is the elastic observer output, ,and Respectively Peacekeeping dimensional real vector space, is the measured output received by the elasticity observer, is the switching signal, is the number of subsystems in the switching system, when season ( ), indicating the The subsystem is activated. It is the DoS attack interval conversion signal. Indicates that the system is in the DoS activation range. Indicates that the system is in DoS sleep range. and is the system matrix of appropriate dimension, is the elastic observer gain associated with the switching signal and the DoS attack interval transition signal, and are the gains of the elastic observer to be designed in the DoS sleep interval and the DoS activation interval respectively.

[0118] In addition, when the DoS attack interval conversion signal is in the sleep interval, that is, When ; When the DoS attack interval conversion signal is in the activation interval, that is, When .

[0119] The elastic bumpless switching controller model is:

[0120] (2.5),

[0121] The following conditions are met:

[0122] (2.6),

[0123] In the formula, is the gain of the elastic disturbance-free switching controller related to the switching signal and the DoS attack interval transition signal, and are the gains of the elastic disturbance-free switching controller to be designed in the DoS sleep interval and the DoS activation interval respectively. At the same time, the gain of the elastic disturbance-free switching controller satisfies the disturbance-free switching constraint (2.6). It is the reference elastic disturbance-free switching controller gain selected according to the aircraft engine system parameters. It is a parameter that quantifies the performance of disturbance-free switching. It is related to the switching signal and the DoS attack interval conversion signal. Under the condition of meeting the requirements of system steady-state performance, the smaller its value is, the better the disturbance-free switching effect is. In particular, when the parameter , ( )are selected as the same value, the coefficients on the right side of equation (2.6) are simplified to The disturbance-free switching constraint (2.6) limits the difference between the control input of the system and the smoothed reference signal to a certain range, thereby achieving disturbance-free switching.

[0124] Combined with the design experience of aircraft engine elastic disturbance-free switching controller, in this implementation scheme, the reference elastic disturbance-free switching controller gain is set to .

[0125] S4: bringing the elastic observer model and the elastic disturbance-free switching controller model into the switching system model, deriving the dynamic equations under the DoS activation mode and the DoS sleep mode respectively, and obtaining a closed-loop system model;

[0126] Combining (2.1), (2.4) and (2.5), the closed-loop system equation is

[0127] (2.7),

[0128] in,

[0129]

[0130]

[0131]

[0132]

[0133] represents the expanded-dimensional state consisting of the elastic observer state and the observation error, represents the derivative of the expanded state, represents the system matrix in the closed-loop system, is the system matrix in DoS sleep mode, is the system matrix in DoS activated mode, express The derivative of represents the error between the elastic observer state and the system state.

[0134] S5: Calculate the elastic observer gain and the elastic disturbance-free switching controller gain for the closed-loop system model based on the time-dependent switching signal:

[0135] The specific steps include:

[0136] S51: Define time-dependent switching signal:

[0137] remember For switching signal In the time interval The number of switching times on and For all the time satisfy

[0138] (2.8),

[0139] in, is called the average dwell time, and It's called the vibration world.

[0140] For the system switching signal, in this implementation scheme, , , the switching signal satisfies ,like Figure 2 As shown, the gray area represents the DoS activation interval, the white area represents the DoS sleep interval, and the red line represents the switching signal.

[0141] S52: constructing a Lyapunov function in the DoS sleep interval and the DoS activation interval based on the time-dependent switching signal and the DoS attack interval conversion signal;

[0142] ,

[0143] in, ( ) is the Lyapunov function, , , , and is a positive definite matrix.

[0144] S53: Based on the Lyapunov function constructed in the DoS sleep interval and the DoS activation interval and the switching system stability theory, a gain selection model of the elastic observer and the elastic disturbance-free switching controller is established;

[0145] The specific method is to select a positive definite matrix , , ,matrix , , , , , a normal number , , , , , , and the constant for satisfy

[0146] (2.9),

[0147] (2.10),

[0148] (2.11),

[0149] (2.12),

[0150] (2.13),

[0151] (2.14),

[0152] (2.15),

[0153] (2.16),

[0154] (2.17),

[0155] in, , , .

[0156] Inequality (2.9) ensures that during the DoS sleep period, a single subsystem is running Inequality (2.10) ensures that during the DoS activation period, a single subsystem is running Inequalities (2.11) and (2.12) ensure that the disturbance-free switching condition (2.6) holds. Inequalities (2.13) to (2.16) constrain the increments of the multi-Lyapunov function at the switching time and at the start and end of the DoS attack. , inequality (2.17) and the above conditions together ensure that the closed-loop system Exponential convergence on .

[0157] S54: Based on the gain selection model of the elastic observer and the elastic disturbance-free switching controller, the gain selection model is optimized by using the Schur complement principle and the matrix scaling method, and a linear matrix inequality calculation model of the gain of the elastic observer and the elastic disturbance-free switching controller is established;

[0158] The specific method is to optimize conditions (2.9) to (2.17) to obtain the linear matrix inequality model for the gain calculation of the elastic observer and the elastic disturbance-free switching controller: The selected matrix , , ,matrix , , , , , a normal number , , , , , , and the constant for satisfy

[0159] (2.18),

[0160] (2.19),

[0161] (2.20),

[0162] (2.21),

[0163] (2.22),

[0164] (2.23),

[0165] (2.24),

[0166] (2.25),

[0167] (2.26),

[0168] in , , .

[0169] S55: obtaining the elastic observer gain and the elastic disturbance-free switching controller gain by solving the linear matrix inequality calculation model of the elastic observer and the elastic disturbance-free switching controller gain;

[0170] Among them, the elastic observer gain is: , ,

[0171] Elastic bumpless switching controller gains: , .

[0172] In the formula, , , , and is the matrix obtained by solving the matrix inequality system (2.18) to (2.26), which satisfies the following equation:

[0173] , , (2.27),

[0174] , , (2.28),

[0175] When the above inequality is solved, during the DoS sleep period, a single subsystem is running Established, during the DoS activation period, a single subsystem is running Established due to , so the stronger convergence of the system in the DoS sleep interval compensates for the weak convergence or divergence in the DoS activation interval, and the security of the closed-loop system is guaranteed. At the same time, the designed elastic disturbance-free switching controller also satisfies the disturbance-free switching condition (2.6), that is, the elastic disturbance-free switching controller gain obtained is , With reference elastic bumpless switching controller gain As close as possible, so as to limit the difference between the control input of the system and the smooth reference signal to a minimum and achieve disturbance-free switching control.

[0176] Using MATLAB simulation debugging, solve the inequality group (2.18) to (2.26) multiple times until the inequality has a solution and Get a smaller value, and the final set of parameters is , , , , , , , , , and the elastic observer gain:

[0177] , ,

[0178] , ,

[0179] and the elastic bumpless switching controller gain:

[0180] , ,

[0181] , ,

[0182] As above, we get the complete elastic observer and elastic disturbance-free switching controller. The elastic disturbance-free switching controller gain is , , , Near reference elastic bumpless switching controller gain , thereby the control input of the system With a smoothed reference signal The difference between them is limited to a very small value to achieve disturbance-free switching control.

[0183] like Figure 3 As shown, the blue line represents the control input of the system With a smoothed reference signal The difference between In particular, due to and The transition between and The jumps between are small, and the control turbulence caused by DoS attacks is suppressed; in addition, due to and The transition between and The jumps between them are small, and the control jolt caused by switching is suppressed.

[0184] The present invention also discloses a non-disturbance switching control system for an aircraft engine system, comprising:

[0185] Switching system module: used to establish a switching system model of the control system based on multiple operating points of the aircraft engine;

[0186] A DoS attack interval conversion signal module is used to classify whether the aircraft engine system is subjected to a DoS attack into a DoS active mode and a DoS dormant mode based on the duration and frequency of the DoS attack, and to establish DoS attack interval conversion signals for the DoS active mode and the DoS dormant mode;

[0187] An elastic observer and an elastic non-disturbance switching controller building module, used to construct an elastic observer model and an elastic non-disturbance switching controller model based on the switching system model and the DoS attack interval conversion signal;

[0188] A closed-loop system model building module is used to bring the elastic observer model and the elastic disturbance-free switching controller model into the switching system model, derive the dynamic equations under the DoS activation mode and the DoS sleep mode respectively, and obtain a closed-loop system model;

[0189] Gain calculation module: used to calculate the elastic observer gain and the elastic disturbance-free switching controller gain for the closed-loop system model based on the time-dependent switching signal.

[0190] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any addition, subtraction or substitution that can be easily thought of by any person familiar with the technical field within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A disturbance-free switching control method for an aircraft engine system, characterized in that: The method comprises: S1: linearizing the aircraft engine control system into multiple subsystems based on multiple operating points of the aircraft engine, and establishing a switching system model of the control system according to the linearized subsystems; S2: Based on the duration and frequency of the DoS attack, the aircraft engine system is divided into DoS active mode and DoS dormant mode to determine whether it is under DoS attack, and a DoS attack interval conversion signal is established; S3: constructing a resilient observer model and a resilient disturbance-free switching controller model based on the switching system model and the DoS attack interval conversion signal; The elastic observer model is: ， In the formula, is the elastic observer state, is the derivative of the elastic observer state, is the elastic observer output, ,and Respectively Wei, and dimensional real vector space, is the measured output received by the elasticity observer, is the switching signal, is the number of subsystems in the switching system, when season ( ), indicating the The subsystem is activated. It is the DoS attack interval conversion signal. Indicates that the system is in the DoS activation range. Indicates that the system is in DoS sleep range. and is the system matrix of appropriate dimension, is the elastic observer gain associated with the switching signal and the DoS attack interval transition signal; The elastic undisturbed switching controller model is: ， The following conditions are met: ， In the formula, is the gain of the elastic disturbance-free switching controller related to the switching signal and the DoS attack interval transition signal, is the reference elastic disturbance-free switching controller gain selected according to the aircraft engine system parameters. is a parameter that quantifies the bumpless switching performance; S4: bringing the elastic observer model and the elastic disturbance-free switching controller model into the switching system model, deriving the dynamic equations under the DoS activation mode and the DoS sleep mode respectively, and obtaining a closed-loop system model; S5: Calculating the elastic observer gain and the elastic disturbance-free switching controller gain for the closed-loop system model based on the time-dependent switching signal.

2. The disturbance-free switching control method according to claim 1, characterized in that: The step S2 of establishing a DoS attack interval conversion signal step includes: S21: Determine the duration and frequency of the DoS attack; make and Respectively represent time intervals The DoS activation interval and DoS sleep interval on and Respectively represent time intervals The total length of the DoS activation interval and the DoS sleep interval, that is, =t- , Indicates time interval The number of DoS attacks on the DoS attack duration: There is a constant in the interval and For all the time satisfy ， DoS attack frequency: There is a constant in the interval and For all the time satisfy ， S22: Establishing a DoS attack interval conversion signal based on the DoS attack model ; in, Indicates that the system is under DoS attack, that is, in the DoS activation range. Indicates that the system is in DoS sleep range; when hour, is 0 or 1, Indicates the time when the transition between the DoS active interval and the dormant interval occurs. express The next transition moment after that.

3. The disturbance-free switching control method according to claim 1, characterized in that: The switching system model of the control system in step S1 is: ， in, is an unpredictable system state. is the derivative of the system state, is the control input, is the measured output, , and Respectively dimension, Peacekeeping dimensional real vector space, is the switching signal, is the number of subsystems in the switching system, when season ( ), indicating the The subsystem is activated. and are two adjacent switching moments, , and is the system matrix of appropriate dimension.

4. The disturbance-free switching control method according to claim 1, characterized in that: The closed-loop system model in step S4 is: ， in, represents the expanded-dimensional state consisting of the elastic observer state and the observation error, represents the derivative of the expanded state, Represents the system matrix in the closed-loop system.

5. The disturbance-free switching control method according to claim 4, characterized in that: The step of calculating the elastic observer gain and the elastic disturbance-free switching controller gain in step S5 includes: S51: defining a time-dependent switching signal; S52: constructing a Lyapunov function in the DoS sleep interval and the DoS activation interval based on the time-dependent switching signal and the DoS attack interval conversion signal; S53: Based on the Lyapunov function constructed in the DoS sleep interval and the DoS activation interval and the switching system stability theory, a gain selection model of the elastic observer and the elastic disturbance-free switching controller is established; S54: Based on the gain selection model of the elastic observer and the elastic disturbance-free switching controller, the gain selection model is optimized by using the Schur complement principle and the matrix scaling method, and a linear matrix inequality calculation model of the gain of the elastic observer and the elastic disturbance-free switching controller is established; S55: Obtaining the elastic observer gain and the elastic undisturbed switching controller gain by solving the linear matrix inequality calculation model of the elastic observer and the elastic undisturbed switching controller gain.

6. The disturbance-free switching control method according to claim 5, characterized in that: The time-dependent switching signal defined in step S51 is specifically: ， For switching signal In the time interval The number of switching times on and For all the time Satisfies the above formula, where is the average residence time, and To shake up the world; In step S52, the Lyapunov function is constructed in the DoS sleep interval and the DoS activation interval as follows: ， in, ( ) is the Lyapunov function, , , , and is a positive definite matrix.

7. The disturbance-free switching control method according to claim 6, characterized in that: The gain selection model of the elastic observer and the elastic disturbanceless switching controller in step S53 includes: Select positive definite matrix , , ,matrix , , , , , a normal number , , , , , , and the constant for satisfy ， ， ， ， ， ， ， ， ， in, , , , and are the elastic observer gains to be designed in the DoS sleep interval and DoS activation interval, respectively. and are the gains of the elastic disturbance-free switching controller to be designed in the DoS sleep interval and the DoS activation interval respectively.

8. The disturbance-free switching control method according to claim 6, characterized in that: The linear matrix inequality calculation model of the elastic observer and the elastic disturbance-free switching controller gain in step S54 includes: Selected Matrix , , ,matrix , , , , , a normal number , , , , , , and the constant for satisfy ， ， ， ， ， ， ， ， ， in, , , .

9. A disturbance-free switching control system for an aircraft engine system, characterized in that: include: Switching system module: used to establish a switching system model of the control system based on multiple operating points of the aircraft engine; The DoS attack interval conversion signal module is used to classify whether the aircraft engine system is under DoS attack into DoS activation mode and DoS sleep mode based on the duration and frequency of the DoS attack, and establish an interval conversion signal between the DoS activation mode and the DoS sleep mode; The elastic observer and elastic non-disturbance switching controller building module is used to construct an elastic observer model and an elastic non-disturbance switching controller model based on the switching system model and the DoS attack interval conversion signal: The elastic observer model is: ， In the formula, is the elastic observer state, is the derivative of the elastic observer state, is the elastic observer output, ,and Respectively Wei, and dimensional real vector space, is the measured output received by the elasticity observer, is the switching signal, is the number of subsystems in the switching system, when season ( ), indicating the The subsystem is activated. It is the DoS attack interval conversion signal. Indicates that the system is in the DoS activation range. Indicates that the system is in DoS sleep range. and is the system matrix of appropriate dimension, is the elastic observer gain associated with the switching signal and the DoS attack interval transition signal; The elastic undisturbed switching controller model is: ， The following conditions are met: ， In the formula, is the gain of the elastic disturbance-free switching controller related to the switching signal and the DoS attack interval transition signal, is the reference elastic disturbance-free switching controller gain selected according to the aircraft engine system parameters. is a parameter that quantifies the bumpless switching performance; A closed-loop system model building module is used to bring the elastic observer model and the elastic disturbance-free switching controller model into the switching system model, derive the dynamic equations under the DoS activation mode and the DoS sleep mode respectively, and obtain a closed-loop system model; Gain calculation module: used to calculate the elastic observer gain and the elastic disturbance-free switching controller gain for the closed-loop system model based on the time-dependent switching signal.

Images