Anti-interference control method for longitudinal model in ground effect area of ground effect vehicle

Abstract

The invention discloses an anti-interference control method for a longitudinal model in a ground effect area of a ground effect aircraft, and the method comprises the steps: firstly arranging a transition process for system expected input and system actual output through a tracking differentiator, and enabling the transition process to reduce the initial value deviation as much as possible; secondly, a new nonlinear function is introduced to improve the extended observer, so that the anti-interference capability of the system is greatly enhanced, and the robustness of the system is enhanced; finally, a new nonlinear function is introduced, a nonlinear feedback law is improved, a system control law is obtained, and the control precision of the system is greatly improved.

Description

Technical Field

[0001] The invention belongs to the technical field of aviation control, and in particular relates to an anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle. Background Art

[0002] A ground effect vehicle is a special aircraft that uses the ground effect to achieve ultra-low altitude and high-speed flight. Relevant scholars at home and abroad have conducted certain research on it. Some universities and research institutions are actively developing ground effect vehicles. For example, the Measurement and Control Technology and Inertial Navigation Teaching and Research Section of the School of Automation of Harbin Engineering University completed the design and production of a prototype in 2010, and conducted relevant experiments using the ground effect vehicle as a research platform, mainly including mathematical modeling and simulation of ground effect vehicles, navigation system hardware and software design, data fusion algorithm, and flight control system design. In general, at present, my country has formed a ground effect vehicle design technology with independent intellectual property rights, and has successively developed several small ground effect vehicles. However, from the perspective of the development of ground effect vehicle flight control technology, it still uses traditional PID control, traditional ADRC control, linear quadratic regulator control, and dynamic inverse control. Therefore, the relevant control research of ground effect vehicles is relatively lacking compared with ordinary fixed-wing aircraft. Summary of the invention

[0003] Purpose of the Invention

[0004] Aiming at the problem that the related control research of existing ground effect vehicles is relatively insufficient, the present invention provides an anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle. Firstly, a tracking differentiator is used to arrange a transition process between a system expected input and a system actual output, so that the initial value deviation is reduced as much as possible; secondly, a new nonlinear function is introduced to improve an extended observer, which greatly enhances the anti-disturbance capability of the system and enhances the robustness of the system; finally, a new nonlinear function is introduced to improve a nonlinear feedback law, and a system control law is obtained, which greatly improves the control accuracy of the system.

[0005] Inventing technical solutions

[0006] An anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle comprises the following steps:

[0007] Step 1: Linearize the small disturbance at the equilibrium point to establish the longitudinal system model of the ground effect vehicle in the ground effect area;

[0008] Step 2: Obtain the subsystem control model of the pitch angle and the elevator according to the longitudinal system model of the ground effect vehicle in the ground effect area established in step 1;

[0009] Step 3: Design a tracking differentiator of the disturbance rejection control method to track the desired input based on the subsystem control model established in step 2;

[0010] Step 4: Design a tracking differentiator of the anti-disturbance control method to track the actual output of the system according to the subsystem control model established in step 2;

[0011] Step 5: Introduce a new nonlinear function;

[0012] Step 6: Design a third-order extended state observer based on the nonlinear function designed in step 5 and the tracking differentiator designed in step 4 for tracking the actual output of the system, so as to estimate the disturbance of the system in real time and compensate it in real time;

[0013] Step 7: Error e 1 and the error differential e 2 ;

[0014] Step 8: Get the nonlinear feedback law u 0 ;

[0015] Step 9: Obtain the system control law.

[0016] Preferably, in step 1, the longitudinal system model of the ground effect vehicle in the ground effect region is:

[0017]

[0018] Where x = [Vαqθh] T , u=[δ e ], A and B are matrices of appropriate dimensions; x is the state vector, u is the control variable, V is the velocity, α is the angle of attack, q is the pitch rate, θ is the pitch angle, h is the height, δ e For the elevator.

[0019] Preferably, the subsystem control model in step 2 is:

[0020]

[0021] Among them, x 1 、x 2 is the state vector, y is the pitch angle output, b is the magnification, w(t) is the external disturbance, and t is the time; f(x 1 ,x 2 ,w(t),t) represents the total disturbance of the system, including internal and external disturbances.

[0022] Preferably, in the tracking differentiator of step 3:

[0023]

[0024] Among them, fhan(v 1 -θ * ,v 2 ,r,h):

[0025]

[0026] Among them, θ * is the desired pitch angle, v 1 、v 2 Divided into the desired pitch angle tracking signal and its differential, r , h are the tuning parameter and filter factor respectively.

[0027] Preferably, in the tracking differentiator of step 4:

[0028]

[0029] Among them, y is the actual pitch angle output, y 1 ,y 2 They are the tracking signal of the actual pitch angle and its differential respectively.

[0030] Preferably, the nonlinear function of step 5 is:

[0031] (1) When δ<|e|≤γ, the nonlinear function is designed as:

[0032] fal(e,α,δ,γ)=|e| α sign(e) (6)

[0033] (2) When |e|≤δ, the nonlinear function is designed as:

[0034] fal(e,α,δ,γ)=λ 1 sine+λ 2 sin 2 e+λ 3 sin 3 e (7)

[0035] (3) When |e|>γ, the nonlinear function is designed as:

[0036] fal(e,α,δ,γ)=γ α sign(e) (8)

[0037] In order to ensure that the piecewise function is continuous within the domain, the new function expression can be derived and solved at the piecewise point:

[0038]

[0039] Among them, γ, α, δ are positive constants greater than zero, and e is the error.

[0040] Preferably, the third-order extended state observer in step 6 is:

[0041]

[0042] Among them, z 1 、z 2 、z 3 is the output of the extended state observer, z 1 、z 2 is the state observation signal, z 3 is the internal and external disturbances of the estimated system, b 1 , b 2 , b 3 are the design parameters of the state observer and are also the three main parameters adjusted in the system, which are determined by the state of the system. 0 is the estimated value of the amplification factor, e is the state error, y 1 is the tracking signal of the actual pitch angle, and u is the control quantity of the system.

[0043] Preferably, the error e in step 7 is 1 and the error differential e 2 The calculation is as follows:

[0044]

[0045] Preferably, the nonlinear feedback law u 0 The calculation is as follows:

[0046] u 0 =b 01 fal(e 1 ,α 3 ,δ,γ)+b 02 fal(e 2 ,α 4 ,δ,γ) (12)

[0047] Among them, b 01 , b 02 is the proportionality coefficient.

[0048] Preferably, the system control law is:

[0049]

[0050] Among them, b 0 is an estimate of the system's object magnification.

[0051] Advantages of the present invention:

[0052] 1. From the simulation of the present invention Figure 2It can be seen that although PID has an extremely fast output response speed, it has a large overshoot and a long adjustment time. Traditional ADRC has a fast output response speed, a large overshoot and a long adjustment time. Compared with the control effects of PID and traditional ADRC, the improved ADRC solves the contradiction between the system's response speed and overshoot, and the system's output response curve can quickly and well fit the system's expected output.

[0053] 2. From the simulation of the present invention Figure 3-Figure 6 As can be seen in the simulation Figure 3 The pitch rate changes of the traditional ADRC and the improved ADRC are small, which means that the damage to the system is small; Figure 4 Compared with the control effects of PID and traditional ADRC, the control effect of improved ADRC is small in pitch angle error change and high in control accuracy. Figure 5 In the extended state observer, z 3 The control law is disturbed and compensated, so the energy required to improve the control quantity in ADRC is small, which improves the economic benefits. Figure 5 Compared with traditional ADRC, the improved ADRC can accurately estimate the total disturbance. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is the self-interference resistance structure diagram.

[0055] Figure 2 It is the pitch angle change curve.

[0056] Figure 3 It is the pitch angle rate change curve.

[0057] Figure 4 This is the pitch angle error variation curve.

[0058] Figure 5 It is the control quantity change curve.

[0059] Figure 6 It is the disturbance change curve. DETAILED DESCRIPTION

[0060] The present invention is achieved through the following technical solutions.

[0061] An anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle comprises the following steps:

[0062] Step 1: According to the particularity of the mission of the ground effect vehicle in the ground effect area, the complexity of the flight environment and other factors, a small disturbance linearization is performed at the equilibrium point to establish the longitudinal system model of the ground effect vehicle in the ground effect area. The control model is described as follows:

[0063]

[0064] Where x = [Vαqθh] T , u=[δ e ], A and B are matrices of appropriate dimensions. x is the state vector, u is the control variable, V is the velocity, α is the angle of attack, q is the pitch rate, θ is the pitch angle, h is the height, δ e For the elevator.

[0065] Step 2: Based on the control model established in step 1, the subsystem control model of the pitch angle and elevator can be obtained. The control model is described as follows:

[0066]

[0067] Among them, x 1 、x 2 is the state vector, y is the pitch angle output, b is the magnification, w(t) is the external disturbance, and t is the time. 1 ,x 2 ,w(t),t) represents the total disturbance of the system, including internal disturbance and external disturbance.

[0068] Step 3: Based on the control model established in step 2, design a tracking differentiator of the disturbance rejection control method to track the desired input. In the tracking differentiator:

[0069]

[0070] Among them, fhan(v 1 -θ * ,v 2 ,r,h):

[0071]

[0072] Among them, θ * is the desired pitch angle, v 1 、v 2 It is divided into the desired pitch angle tracking signal and its differential, r and h are the tuning parameters and filter factors respectively.

[0073] Step 4: Based on the control model established in step 2, design a tracking differentiator of the anti-disturbance control method to track the actual output of the system. In the tracking differentiator:

[0074]

[0075] Among them, y is the actual pitch angle output, y 1 ,y 2 They are the tracking signal of the actual pitch angle and its differential respectively.

[0076] Step 5: Introduce a new nonlinear function, which is:

[0077] (1) When δ<|e|≤γ, the new function is designed as:

[0078] fal(e,α,δ,γ)=|e| α sign(e) (19)

[0079] (2) When |e|≤δ, the new function is designed as:

[0080] fal(e,α,δ,γ)=λ 1 sine+λ 2 sin 2 e+λ 3 sin 3 e (20)

[0081] (3) When |e|>γ, the new function is designed as:

[0082] fal(e,α,δ,γ)=γ α sign(e) (21)

[0083] In order to ensure that the piecewise function is continuous within the domain, the new function expression can be derived and solved at the piecewise point:

[0084]

[0085] Among them, γ, α, δ are positive constants greater than zero, and e is the error.

[0086] Step 6: Based on the new function designed in step 5 and the tracking differentiator of the actual system output designed in step 4, the following third-order extended state observer is designed to estimate the disturbance of the system in real time and compensate it in real time:

[0087]

[0088] Among them, z 1 、z 2 、z 3 is the output of the extended state observer, z 1 、z 2 is the state observation signal, z 3 is the internal and external disturbances of the estimated system, b 1 , b 2 , b 3are the design parameters of the state observer and are also the three main parameters adjusted in the system, which are determined by the state of the system. 0 is the estimated value of the amplification factor, e is the state error, y 1 is the tracking signal of the actual pitch angle, and u is the control quantity of the system.

[0089] Here, in order to avoid the occurrence of high-frequency oscillation, the fal function is referenced; the extended state observer is a dynamic process, which only uses the input-output information of the original object, and does not use any information of the function that describes the object transfer relationship. 3 The fundamental reason for being able to track the real-time action of the system's acceleration very well is that as long as the system meets the observability conditions, no matter what form the acceleration takes, as long as it is taking effect, its effect will definitely be reflected in the system's output. This is a specific way to extract the real-time action of the system's acceleration from the system's output information.

[0090] In summary, the designed extended state observer can be used to estimate in real time the numerous high-frequency unmodeled dynamics, uncertainties and external disturbances such as wind and wave disturbances to the ground effect vehicle in the system, thereby enhancing the stability and robustness of the ground effect vehicle system.

[0091] Step 7: According to the v in step 3 1 、v 2 and z in step 6 1 、z 2 Make a difference and get the error e 1 and the error differential e 2 :

[0092]

[0093] The feedback mechanism is a mechanism that is not found in classical dynamics systems but is inherent in controlled systems. The feedback mechanism in controlled systems also has the ability to suppress small uncertain disturbances. Different feedback forms have different disturbance suppression capabilities. The efficiency of linear feedback is lower than that of some nonlinear state feedback. Nonlinear state feedback can be used to make the closed-loop system have better dynamic characteristics. This is also the idea of ​​using state feedback to configure nonlinear dynamics in closed loops.

[0094] Step 8: According to fal(e,α,δ,γ) in step 5 and e in step 7 1 、e 2 , we get the nonlinear feedback law u 0 as follows:

[0095] u 0 =b 01 fal(e 1 ,α 3 ,δ,γ)+b02 fal(e 2 ,α 4 ,δ,γ) (25)

[0096] Among them, b 01 , b 02 is the proportionality coefficient.

[0097] Step 9: After designing the nonlinear feedback law, the system control law is obtained according to the principle of active disturbance rejection control:

[0098]

[0099] Among them, b 0 is an estimate of the system's object magnification.

[0100] The process of dynamic compensation linearization is to use the result of the extended state observer estimation to transform the control quantity into the above form for nonlinear control systems, so that the original nonlinear control system becomes a linear control system. In this way, no matter whether the object is deterministic or uncertain, linear or nonlinear, time-varying or time-invariant, after compensation, the system can be transformed into an integrator series controlled system. Therefore, the compensation method of the extended state observer gives us a possibility to deal with the control problems of control systems such as determinism and uncertainty, linearity and nonlinearity, time-varying or time-invariant in a unified way.

[0101] In order to verify the effectiveness of the stabilizer based on the active disturbance rejection control technology designed above, the present invention uses full-process digital simulation to debug and test the active disturbance rejection controller, studies the engineering application of the model, and compares it with traditional PID and ADRC control to obtain Figure 2-Figure 6 .

[0102] The protection scope of the present invention is not limited to the above-mentioned embodiments. Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the scope of the present invention. If these changes and modifications fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include these changes and modifications.

Claims

1. A disturbance rejection control method for a longitudinal model of a ground effect vehicle in a ground effect region, characterized in that: The steps include: Step 1: Linearize the small disturbance at the equilibrium point to establish the longitudinal system model of the ground effect vehicle in the ground effect area; Step 2: Obtain the subsystem control model of the pitch angle and the elevator according to the longitudinal system model of the ground effect vehicle in the ground effect area established in step 1; Step 3: Design a tracking differentiator of the disturbance rejection control method to track the desired input based on the subsystem control model established in step 2; Step 4: Design a tracking differentiator of the anti-disturbance control method to track the actual output of the system according to the subsystem control model established in step 2; Step 5: Introduce a new nonlinear function; Step 6: Design a third-order extended state observer based on the nonlinear function designed in step 5 and the tracking differentiator designed in step 4 for tracking the actual output of the system, so as to estimate the disturbance of the system in real time and compensate it in real time; Step 7: Error e1 and error differential e2; Step 8: Obtain the nonlinear feedback law u0; Step 9: Obtain the system control law.

2. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 1, characterized in that: In step 1, the longitudinal system model of the ground effect vehicle in the ground effect area is: Where x = [V α q θ h] T , u=[δ e ], A and B are matrices of appropriate dimensions; x is the state vector, u is the control variable, V is the velocity, α is the angle of attack, q is the pitch rate, θ is the pitch angle, h is the height, δ e For the elevator.

3. A method for controlling a longitudinal model of a ground effect vehicle in a ground effect region according to claim 2, characterized in that: The subsystem control model in step 2 is: Among them, x1 and x2 are state vectors, y is the pitch angle output, b is the amplification factor, w(t) is the external disturbance, and t is the time; f(x1, x2, w(t), t) represents the total disturbance of the system, including internal and external disturbances.

4. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 3, characterized in that: In the tracking differentiator of step 3: Among them, fhan(v1-θ * ,v2,r,h): Among them, θ * is the desired pitch angle, v1 and v2 are the desired pitch angle tracking signal and its differential, r and h are the tuning parameters and filter factors, respectively.

5. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 4, characterized in that: In the tracking differentiator of step 4: Among them, y is the actual pitch angle output, y1 and y2 are the tracking signal of the actual pitch angle and its differential respectively.

6. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 5, characterized in that: The nonlinear function of step 5 is: (1) When δ<|e|≤γ, the nonlinear function is designed as: fal(e,a,d,c)=|e| α sign(e) (6) (2) When |e|≤δ, the nonlinear function is designed as: fal(e,a,d,g)=λ1sine+λ2sin 2 e+λ3sin 3 e (7) (3) When |e|>γ, the nonlinear function is designed as: fal(e,a,d,c)=c α sign(e) (8) In order to ensure that the piecewise function is continuous within the domain, the new function expression can be derived and solved at the piecewise point: Among them, γ, α, δ are positive constants greater than zero, and e is the error.

7. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 6, characterized in that: The third-order extended state observer in step 6 is: Among them, z1, z2, and z3 are the outputs of the extended state observer, z1 and z2 are state observation signals, z3 is the internal and external disturbances of the estimated system, b1, b2, and b3 are the design parameters of the state observer, and are also the three main parameters adjusted in the system, which are determined by the state of the system, b0 is the estimated value of the amplification factor, e is the state error, y1 is the tracking signal of the actual pitch angle, and u is the control quantity of the system.

8. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 6, characterized in that: The error e1 and the error differential e2 in step 7 are calculated as follows:

9. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle as claimed in claim 8, characterized in that: The nonlinear feedback law u0 is calculated as follows: u0=b 01 fal(e1,a3,d,c)+b 02 fal(e2,a4,d,c) (12) Among them, b 01 , b 02 is the proportionality coefficient.

10. The anti-disturbance control method for a longitudinal model in a ground effect region of a ground effect vehicle according to claim 9, characterized in that: The system control law is: where b0 is an estimate of the system's object magnification.

Images