CN102749852A - Fault-tolerant anti-interference control method for multisource interference system - Google Patents

Abstract

The invention relates to a fault-tolerant anti-interference control method for a multisource interference system. Aiming at the multisource interference system containing time-varying faults, modeling interference and non-modeling random interference, a fault-tolerant anti-interference controller is designed. The fault-tolerant anti-interference control method comprises the following steps of: firstly, designing a fault diagnosis observer to estimate and counteract the time-varying faults in the system; secondly, designing an interference observer to estimate and counteract the modeling interference in the multisource interference system; thirdly, designing a robust H infinity state feedback controller to inhibit the non-modeling random interference, fault estimation errors and interference estimation errors in the multisource interference system; and finally, designing the fault-tolerant anti-interference controller based on the fault diagnosis observer, the interference observer and the robust H infinity state feedback controller. The method has the advantages of high anti-interference performance, significant fault-tolerant performance, high working reliability and the like and can be used for altitude control subsystems in the fields of aviation and aerospace.

Description

The fault-tolerant anti-interference control method of multi-source EVAC

Technical field

The present invention relates to a kind of fault-tolerant anti-interference control method; Particularly a kind of fault-tolerant anti-interference control method of multi-source EVAC; This method can be used for the fault diagnosis and the fault-tolerant anti-interference control of multi-source EVAC, like the attitude RACS of aerospace systems such as satellite, guided missile and aircraft.

Background technology

In recent years, along with development of aviation and aerospace technology, the structure of aircraft and mission requirements are complicated day by day, and are also increasingly high to the requirement of control accuracy and degree of stability, cause the fault probability of happening of aircraft also more and more high.For example, the scholar is arranged through 764 spacecrafts that succeed in sending up in the period of the 1990-2001 have been carried out statistics and analysis both at home and abroad, the result finds that 121 break down, and accounts for 15.8% of spacecraft sum.The reliability of aircraft work, a research emphasis that becomes aerospace field in the maintainability and the validity of rail flight.And being reliability, maintainability and the validity of raising system, fault-tolerant control and fault detection and diagnosis opened up a new way.In addition; Be accompanied by the complicated day by day of Flight Vehicle Structure and mission requirements; The factor that influences attitude of flight vehicle control accuracy and degree of stability is more and more, mainly reduce following some: external environment condition disturbance torque, the vibration of the inner topworks of celestial body flywheel, the friction of flywheel, the unloading of jet momentum, sensor are measured the uncertainty of noise and system model etc.

To above-mentioned a series of problems, when change energy bounded fault was not still disturbed when existing in the system, Chinese scholars had proposed a lot of effective methods and has obtained certain effect.But consider that interference is at all times, ubiquitous in the real system, single consideration fault will be brought a series of problem, particularly can not guarantee the attitude control accuracy and the degree of stability of system.On this basis, the situation of accident barrier and interference becomes a research direction when existing simultaneously in the system, and Chinese scholars has proposed a series of solution, for example H _∞Optimisation technique, based on the H of internal model structure _∞Controller etc.But existing method is on the basis of design observer suspected fault; Regarding interference as the norm bounded quantity suppresses; Such disposal route has the inferior position of following two aspects: first; Regard all interference in the system as the norm bounded quantity and suppress, the interference that is about to exist in the system is regarded an integral body as and is handled, but has ignored that internal system is divided information that modeling disturbs or through the obtainable interfere information of physical measurement means; Do not make full use of system resource, be difficult to realize high-accuracy posture control; The second, regard all interference in the system as the norm bounded quantity and suppress, will certainly increase the conservative property of system, be difficult to realize high-accuracy posture control equally.

Summary of the invention

Technology of the present invention is dealt with problems and is: to the multi-source EVAC; Overcome the deficiency of prior art; A kind of fault-tolerant anti-interference control method with Interference Cancellation and rejection is provided; Solve Interference Cancellation, interference inhibition and the fault diagnosis and the fault-tolerant control problem of multi-source EVAC, improve the control accuracy and the degree of stability of system.

Technical solution of the present invention is: a kind of fault-tolerant anti-interference control method of multi-source EVAC is characterized in that may further comprise the steps:

At first, the time accident in estimation of design error failure diagnostic observations device and the bucking-out system hinders; Next, but the modeling in estimation of design interference observer and the counteracting multi-source EVAC is disturbed; Once more, design robust H _∞State feedback controller suppress in the multi-source EVAC can not the modeling random disturbance, Fault Estimation sum of errors Interference Estimation error; At last, based on fault diagnosis observer, interference observer and robust H _∞State feedback controller designs fault-tolerant anti-interference controller; Concrete steps are following:

The first step is built the kinetic model that contains the multi-source EVAC, and is write as state-space expression

, build its system dynamics model, and it is following to be write as state-space expression to the multi-source EVAC that contains that but the modeling of the barrier of accident is sometimes disturbed and can not the modeling random disturbance:

$\stackrel{\·}{x}\left(t\right)+\mathrm{Ef}(\stackrel{\·}{x},t)=\mathrm{Ax}\left(t\right)+B_1(u\left(t\right)+d_1\left(t\right)+F\left(t\right))+B_2{d}_2\left(t\right)$

Wherein, x (t) is a multi-source EVAC state variable, and u (t) is control input, d ₁(t) but be that modeling is disturbed, F (t) for the time accident hinder d ₂(t) be can not the modeling random disturbance, A, E, B ₁And B ₂Be the matrix of known dimension,

For system's nonlinear terms and satisfy the Lipschitz condition, external model description disturbance d ₁(t) by following external disturbance model ∑ ₁Expression:

${\mathrm{\Σ}}_1:\left\{\begin{array}{c}{d}_1\left(t\right)=\mathrm{Vw}\left(t\right)\\ \stackrel{\·}{w}\left(t\right)=\mathrm{Ww}\left(t\right)+B_3\mathrm{\δ}\left(t\right)\end{array}\right.$

Wherein, w (t) but be the state variable of modeling interference model, but V is the output matrix of modeling interference model, but W representes system's battle array of modeling interference model, δ (t) be energy bounded can not the modeling random disturbance, B ₃Be gain battle array that can not the modeling random disturbance.

Second step, design error failure diagnostic observations device

Time accident in the multi-source EVAC hinders F (t); Design error failure diagnostic observations device is estimated in real time it, and is tried to achieve estimated value

and then obtain Fault Estimation error

The 3rd step, the design interference observer

But disturb d to the modeling in the multi-source EVAC ₁(t), the design interference observer is estimated in real time it, and is tried to achieve estimated value

And then obtain the Interference Estimation error Estimated value for w (t).

The 4th step, design robust H _∞State feedback controller

To in the multi-source EVAC can not modeling random disturbance d ₂(t), Fault Estimation error e _F(t) and the Interference Estimation error e _w(t), design robust H _∞State feedback controller suppresses it, and controller architecture is following:

u _f(t)＝Mx(t)

Wherein, u _f(t) be robust H _∞The STATE FEEDBACK CONTROL input, M is a state feedback controller gain battle array undetermined.

In the 5th step, design fault-tolerant anti-interference controller

Design fault-tolerant anti-interference controller, the time accident in the system is hindered F (t) but and modeling interference d ₁(t) offset, can not modeling random disturbance d ₂(t), Fault Estimation error e _F(t) and the Interference Estimation error e _w(t) suppress, fault-tolerant anti-interference controller structure is following:

$u\left(t\right)=u_f\left(t\right)-\hat{F}\left(t\right)-{\hat{d}}_1\left(t\right)$

Then the multi-source EVAC can be expressed as:

$\stackrel{\·}{x}\left(t\right)+\mathrm{Ef}(\stackrel{\·}{x},t)=(A+B_{1}M)x\left(t\right)+B_{1}V{e}_w\left(t\right)+B_1{e}_F\left(t\right)+B_2{d}_2\left(t\right)$

But the system estimation error equation of arrangement modeling interference model with the time accident system estimation error equation that hinders following:

${\stackrel{\·}{e}}_w\left(t\right)=(W+{\mathrm{LB}}_{1}V)e_w\left(t\right)+{\mathrm{LB}}_1{e}_F\left(t\right)+{\mathrm{LB}}_2{d}_2\left(t\right)+B_3\mathrm{\δ}\left(t\right)$

${\stackrel{\·}{e}}_F\left(t\right)={\mathrm{KB}}_{1}V{e}_w\left(t\right)+{\mathrm{KB}}_1{e}_F\left(t\right)+{\mathrm{KB}}_2{d}_2\left(t\right)+\stackrel{\·}{F}\left(t\right)$

But join list state the multi-source EVAC, the time accident barrier system estimation error equation and the system estimation error equation that modeling is disturbed obtain closed-loop system:

$\left\{\begin{array}{c}\left(\begin{array}{c}\stackrel{\·}{x}\left(t\right)\\ {\stackrel{\·}{e}}_w\left(t\right)\\ {\stackrel{\·}{e}}_F\left(t\right)\end{array}\right)=\left(\begin{array}{ccc}A+B_{1}M& B_{1}V& B_1\\ 0& W+{\mathrm{LB}}_{1}V& {\mathrm{LB}}_1\\ 0& {\mathrm{KB}}_{1}V& {\mathrm{KB}}_1\end{array}\right)\left(\begin{array}{c}x\left(t\right)\\ e_w\left(t\right)\\ e_F\left(t\right)\end{array}\right)+\left(\begin{array}{c}{B}_2\\ {\mathrm{LB}}_2\\ {\mathrm{KB}}_2\end{array}\right)d_2\left(t\right)+\left(\begin{array}{c}0\\ B_3\\ 0\end{array}\right)\mathrm{\δ}\left(t\right)+\left(\begin{array}{c}0\\ 0\\ 1\end{array}\right)\stackrel{\·}{F}\left(t\right)-\left(\begin{array}{c}E\\ 0\\ 0\end{array}\right)f\left(\stackrel{\·}{x}\left(t\right)\right)\\ z_{\∞}\left(t\right)=\left[\begin{array}{ccc}{C}_0& C_1& C_2\end{array}\right]\left(\begin{array}{c}x\left(t\right)\\ e_w\left(t\right)\\ e_F\left(t\right)\end{array}\right)\end{array}\right.$

Wherein, z _∞(t) be H _∞Performance is with reference to output, [C ₀C ₁C ₂] be H _∞The adjustable output matrix of performance.

In the 6th step, gain matrix is found the solution

Utilize convex optimized algorithm to find the solution the fault-tolerant anti-interference controller gain battle array of multi-source EVAC; Given initial value x (0), e _w(0) and e _F(0), adjustable output matrix [C ₀C ₁C ₂], non-linear weight parameter λ disturbs inhibition degree γ ₁, γ ₂And γ ₃, find the solution following protruding optimization problem:

$\min \left(\begin{array}{ccc}\left(\begin{array}{cc}{x}^T\left(0\right)& e^T\left(0\right)\end{array}\right)& \left(\begin{array}{cc}{P}_1& \\ & P_2\end{array}\right)& {\left(\begin{array}{cc}{x}^T\left(0\right)& e^T\left(0\right)\end{array}\right)}^T\end{array}\right)$

$\mathrm{\Φ}=\left[\begin{array}{cccccccc}{\mathrm{\Φ}}_{11}& B_{1}G& -E& 0& B_2& 0& {\mathrm{\Φ}}_{18}& P_1{C}_0^T\\ *& {\mathrm{\Φ}}_{22}& 0& P_2{H}_1& R_2{B}_2& P_2{H}_2& \mathrm{\λ}{\left(U{B}_{1}G\right)}^T& C^T\\ *& *& -{\mathrm{\λ}}^{2}I& 0& 0& 0& \mathrm{\λ}{\left(\mathrm{UE}\right)}^T& 0\\ *& *& *& -{\mathrm{\γ}}_1^2& 0& 0& 0& 0\\ *& *& *& *& -{\mathrm{\γ}}_2^2& 0& \mathrm{\λ}{\left({\mathrm{UB}}_2\right)}^T& 0\\ *& *& *& *& *& -{\mathrm{\γ}}_3^{2}I& 0& 0\\ *& *& *& *& *& *& -I& 0\\ *& *& *& *& *& *& *& -I\end{array}\right]$

Wherein: e (0)=[e _w(0) e _F(0)] ^T, Φ ₁₁=(AP ₁+ B ₁R ₁)+(AP ₁+ B ₁R ₁) ^T, Φ ₂₂=(P ₂W ₁+ R ₂B ₁G)+(P ₂W ₁+ R ₂B ₁G) ^T, Φ ₁₈=λ (AP ₁+ B ₁R ₁) ^T, C=[C ₁C ₂], G=[EI], H ₁=[B ₃0] ^T, H ₂=[0 1] ^TSymbol * representes the symmetry blocks of appropriate section in the symmetric matrix, find the solution P ₁, P ₂, R ₁And R ₂, then interference observer and fault diagnosis observer gain battle array does $\left[\begin{array}{c}L\\ K\end{array}\right]=P_2^{-1}{R}_2,$ State feedback controller gain battle array is

Fault diagnosis observer structure in the said step 2 is following:

$\left\{\begin{array}{c}\hat{F}\left(t\right)=\mathrm{\τ}-p\left(x\right)\\ \stackrel{\·}{\mathrm{\τ}}={\mathrm{KB}}_1(\mathrm{\τ}-p\left(x\right))+K[\mathrm{Ax}\left(t\right)+B_{1}u\left(t\right)+B_1{\hat{d}}_1\left(t\right)-\mathrm{Ef}\left(\stackrel{\·}{x}\left(t\right)\right)]\end{array}\right.$

Wherein, K is a fault diagnosis observer gain matrix undetermined, and ε (t) is an auxiliary variable.

Interference observer structure in the said step 3 is following:

$\left\{\begin{array}{c}{\hat{d}}_1\left(t\right)=V\hat{w}\left(t\right)\\ \hat{w}\left(t\right)=v\left(t\right)-\mathrm{Lx}\left(t\right)\\ \stackrel{\·}{v}\left(t\right)=(W+{\mathrm{LB}}_{1}V)(v\left(t\right)-\mathrm{Lx}\left(t\right))+L[\mathrm{Ax}\left(t\right)+B_{1}u\left(t\right)+B_1\hat{F}\left(t\right)-\mathrm{Ef}(\stackrel{\·}{x},t)]\end{array}\right.$

Wherein, v (t) is an auxiliary variable, and L is an interference observer gain matrix undetermined, but V is the output matrix of modeling interference model, but W representes system's battle array of modeling interference model.

The present invention's advantage compared with prior art is:

(1) the fault-tolerant anti-interference control method of multi-source EVAC of the present invention is a kind of composite layered anti-interference control method; The feedforward part of controller is made up of fault diagnosis observer and interference observer; Be used for estimating that but the time accident with bucking-out system hinders and the modeling interference, the feedback fraction of controller is by robust H _∞State feedback controller constitutes, and the controller of design makes system have more meticulous fault diagnosis and fault-tolerant control ability.

(2) strong robustness of the present invention to disturbing; But the time accident barrier modeling that has rate of change bounded is at the same time disturbed and can not the modeling random disturbance etc. under the multi-source situation about disturbing; Accident barrier when the fault diagnosis observer in the said method can be estimated and offset; Interference observer can be estimated and but robust H is disturbed in the counteracting modeling _∞State feedback controller suppresses can not the modeling random disturbance, Fault Estimation sum of errors Interference Estimation error, has overcome existing method and has regarded interference as the big problem of conservative property that the norm bounded quantity suppresses to bring.

Description of drawings

Fig. 1 is the design flow diagram of a kind of fault-tolerant anti-interference control method based on the multi-source EVAC of the present invention.

Embodiment

As shown in Figure 1, the concrete performing step of the present invention as follows (following with the attitude of satellite confirm with control system be the concrete realization that example is come illustration method):

1, builds the kinetic model that contains the multi-source EVAC, and write as state-space expression

When the Eulerian angle between micro-nano satellite body coordinate system and the orbital coordinate system are very little, can obtain following satellite linearization attitude dynamics and kinematics model:

$\left\{\begin{array}{c}{J}_1\stackrel{\·\·}{\mathrm{\φ}}-n(J_1-J_2+J_3)\stackrel{\·}{\mathrm{\ψ}}+4n^2(J_2-J_3)\mathrm{\φ}=u_1+T_{d1}\\ J_2\stackrel{\·\·}{\mathrm{\θ}}+3n^2(J_1-J_3)\mathrm{\θ}=u_2+F\left(t\right)+T_{d2}\\ J_3\stackrel{\·\·}{\mathrm{\ψ}}+n(J_1-J_2+J_3)\stackrel{\·}{\mathrm{\φ}}+n^2(J_2-J_1)\mathrm{\ψ}=u_3+T_{d3}\end{array}\right.$

In the following formula, J ₁, J ₂, J ₃Be respectively three moment of inertia, n is a satellite orbit angular velocity, and φ, θ, ψ are respectively three Eulerian angle between satellite body coordinate system and the orbital coordinate system;

Be respectively three Eulerian angle speed;

Be respectively three Eulerian angle acceleration; u ₁, u ₂, u ₃Be respectively three control moments; F (t) for the time accident barrier, T _D1, T _D2, T _D3Be respectively three disturbance torque (comprising) because the disturbance torque that sensor or topworks's fault are brought;

The uncertain main uncertainty of micro-nano satellite model from moment of inertia, CONSIDERING THE EFFECTS OF ROTATION inertia is uncertain, from the attitude dynamics model, extracts inertia matrix, and following formula can be converted into following form:

$(M+\mathrm{\ΔM})\stackrel{\·\·}{p}\left(t\right)+(C+\mathrm{\ΔC})\stackrel{\·}{p}\left(t\right)+(S+\mathrm{\ΔS})p\left(t\right)=B_u(u\left(t\right)+d_1\left(t\right)+F\left(t\right))+B_w{d}_2\left(t\right)$

State variable p (t)=[φ, θ, ψ] wherein ^TBe three Eulerian angle,

Be three Eulerian angle speed,

Be three Eulerian angle acceleration, d ₁(t) but be that modeling is disturbed, d ₂(t) for energy bounded can not modeling (be L at random ₂Norm

Bounded) disturb B _uBe control input allocation matrix, B _wFor can not the modeling random disturbance input allocation matrix, M, C, S are known moment of inertia, Δ M, Δ C, Δ S be because the uncertain moment of inertia of disturbed belt, $M=\left[\begin{array}{ccc}{I}_1& 0& 0\\ 0& I_2& 0\\ 0& 0& I_3\end{array}\right],$ $C=\left[\begin{array}{ccc}0& 0& -\mathrm{\ω}(I_1-I_2+I_3)\\ 0& 0& 0\\ \mathrm{\ω}(I_1-I_2+I_3)& 0& 0\end{array}\right],$ $S=\left[\begin{array}{ccc}4{\mathrm{\ω}}^2(I_2-I_3)& 0& 0\\ 0& 3{\mathrm{\ω}}^2(I_1-I_3)& 0\\ 0& 0& -{\mathrm{\ω}}^2(I_1-I_2)\end{array}\right]$ $B_u=B_w=\left[\begin{array}{ccc}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1\end{array}\right],$ Following formula put in order converts state-space model into and be shown below:

$\stackrel{\·}{x}\left(t\right)+\mathrm{Ef}(\stackrel{\·}{x},t)=\mathrm{Ax}\left(t\right)+B_1(u\left(t\right)+d_1\left(t\right)+F\left(t\right))+B_2{d}_2\left(t\right)$

Wherein, multi-source EVAC state variable $x\left(t\right)={\left[\begin{array}{ccc}{\∫}_0^t{e}_q\left(\mathrm{\τ}\right)\mathrm{D\τ}& e_q\left(t\right)& {\stackrel{\·}{e}}_q\end{array},\left(t\right)\right]}^T,$ e _q(t)=p (t)-p _p(t), p _p(t) be the reference locus signal, u (t) is control input, A, E, B ₁And B ₂Be the matrix of known dimension, $A=\left[\begin{array}{ccc}0& 1& 0\\ 0& 0& 1\\ 0& -M^{-1}S& -M^{-1}C\end{array}\right],$ $E=\left[\begin{array}{ccc}0& 0& 0\\ 0& 0& 0\\ M^{-1}\mathrm{\Δ\; S}& M^{-1}\mathrm{\Δ\; C}& M^{-1}\mathrm{\Δ\; M}\end{array}\right],$ $B_1=\left[\begin{array}{c}0\\ 0\\ M^{-1}{B}_u\end{array}\right],$ $B_2=\left[\begin{array}{c}0\\ 0\\ M^{-1}{B}_w\end{array}\right],$ Nonlinear terms f (

(t)) satisfy the Lipschitz condition, promptly have known Lipschitz parametric array U ∈ R ^{3 * 3}Make and set up like lower inequality:

$\left|\right|f\left({\stackrel{\·}{x}}_1\left(t\right)\right)-f\left({\stackrel{\·}{x}}_2\left(t\right)\right)\left|\right|\≤\left|\right|U({\stackrel{\·}{x}}_1\left(t\right)-{\stackrel{\·}{x}}_2\left(t\right))\left|\right|$

Wherein, Be any two states in the system state set, external model description disturbance d ₁(t) by following external disturbance model ∑ ₁Expression:

${\mathrm{\Σ}}_1:\left\{\begin{array}{c}{d}_1\left(t\right)=\mathrm{Vw}\left(t\right)\\ \stackrel{\·}{w}\left(t\right)=\mathrm{Ww}\left(t\right)+B_3\mathrm{\δ}\left(t\right)\end{array}\right.$

Wherein, w (t) but be the state variable of modeling interference model, but V is the output matrix of modeling interference model, but W representes system's battle array of modeling interference model, δ (t) for energy bounded can not modeling (be L at random ₂Norm

Bounded) disturb B ₃Be energy BOUNDED DISTURBANCES gain battle array that can not the modeling interference model.

2, design error failure diagnostic observations device

Time accident in the multi-source EVAC hinders F (t), and design error failure diagnostic observations device is:

$\left\{\begin{array}{c}\hat{F}\left(t\right)=\mathrm{\τ}-p\left(x\right)\\ \stackrel{\·}{\mathrm{\τ}}={\mathrm{KB}}_1(\mathrm{\τ}-p\left(x\right))+K[\mathrm{Ax}\left(t\right)+B_{1}u\left(t\right)+B_1{\hat{d}}_1\left(t\right)-\mathrm{Ef}\left(\stackrel{\·}{x}\left(t\right)\right)]\end{array}\right.$

Wherein, ε (t) is an auxiliary variable; K is a fault diagnosis observer gain matrix undetermined; Try to achieve through subsequent step 6, and then obtain Fault Estimation error

3, design interference observer

But disturb d to modeling in the multi-source EVAC ₁(t), the design interference observer is:

$\left\{\begin{array}{c}{\hat{d}}_1\left(t\right)=V\hat{w}\left(t\right)\\ \hat{w}\left(t\right)=v\left(t\right)-\mathrm{Lx}\left(t\right)\\ \stackrel{\·}{v}\left(t\right)=(W+{\mathrm{LB}}_{1}V)(v\left(t\right)-\mathrm{Lx}\left(t\right))+L[\mathrm{Ax}\left(t\right)+B_{1}u\left(t\right)+B_1\hat{F}\left(t\right)-\mathrm{Ef}(\stackrel{\·}{x},t)]\end{array}\right.$

Wherein, For but d is disturbed in modeling ₁(t) estimated value,

Be the estimated value of w (t), v (t) is an auxiliary variable, and L is an interference observer gain matrix undetermined, tries to achieve through subsequent step 6, and then obtains the Fault Estimation error $e_w\left(t\right)=w\left(t\right)-\hat{w}\left(t\right).$

4, design robust H _∞State feedback controller

To in the multi-source EVAC can not modeling random disturbance d ₂(t), Fault Estimation error e _F(t) and the Interference Estimation error e _w(t) design robust H _∞State feedback controller suppresses it, and controller architecture is following:

u _f(t)＝Mx(t)

Wherein, u _f(t) be state feedback controller, M is a state feedback controller gain battle array undetermined.

5, design fault-tolerant anti-interference controller

Based on interference observer, fault diagnosis observer and robust H _∞State feedback controller, it is following to design fault-tolerant anti-interference controller:

$u\left(t\right)=u_f\left(t\right)-{\hat{d}}_1\left(t\right)-\hat{F}\left(t\right)$

Then the multi-source EVAC can be expressed as:

$\stackrel{\·}{x}\left(t\right)+\mathrm{Ef}(\stackrel{\·}{x},t)=(A+B_{1}M)x\left(t\right)+B_{1}V{e}_w\left(t\right)+B_1{e}_F\left(t\right)+B_2{d}_2\left(t\right)$

But the system estimation error equation that the system estimation error equation of arrangement modeling interference model and the time accident of rate of change bounded hinder is following:

${\stackrel{\·}{e}}_w\left(t\right)=(W+{\mathrm{LB}}_{1}V)e_w\left(t\right)+{\mathrm{LB}}_1{e}_F\left(t\right)+{\mathrm{LB}}_2{d}_2\left(t\right)+B_3\mathrm{\δ}\left(t\right)$

${\stackrel{\·}{e}}_F\left(t\right)={\mathrm{KB}}_{1}V{e}_w\left(t\right)+{\mathrm{KB}}_1{e}_F\left(t\right)+{\mathrm{KB}}_2{d}_2\left(t\right)+\stackrel{\·}{F}\left(t\right)$

But the system estimation error equation that joins accident barrier when listing the system estimation sum of errors of stating multi-source EVAC modeling interference model obtains closed-loop system:

$\left\{\begin{array}{c}\left(\begin{array}{c}\stackrel{\·}{x}\left(t\right)\\ {\stackrel{\·}{e}}_w\left(t\right)\\ {\stackrel{\·}{e}}_F\left(t\right)\end{array}\right)=\left(\begin{array}{ccc}A+B_{1}M& B_{1}V& B_1\\ 0& W+{\mathrm{LB}}_{1}V& {\mathrm{LB}}_1\\ 0& {\mathrm{KB}}_{1}V& {\mathrm{KB}}_1\end{array}\right)\left(\begin{array}{c}x\left(t\right)\\ e_w\left(t\right)\\ e_F\left(t\right)\end{array}\right)+\left(\begin{array}{c}{B}_2\\ {\mathrm{LB}}_2\\ {\mathrm{KB}}_2\end{array}\right)d_2\left(t\right)+\left(\begin{array}{c}0\\ B_3\\ 0\end{array}\right)\mathrm{\δ}\left(t\right)+\left(\begin{array}{c}0\\ 0\\ 1\end{array}\right)\stackrel{\·}{F}\left(t\right)-\left(\begin{array}{c}E\\ 0\\ 0\end{array}\right)f\left(\stackrel{\·}{x}\left(t\right)\right)\\ z_{\∞}\left(t\right)=\left[\begin{array}{ccc}{C}_0& C_1& C_2\end{array}\right]\left(\begin{array}{c}x\left(t\right)\\ e_w\left(t\right)\\ e_F\left(t\right)\end{array}\right)\end{array}\right.$

Wherein, z _∞(t) be H _∞Performance is with reference to output, [C ₀C ₁C ₂] be H _∞The adjustable output matrix of performance.

6, gain matrix is found the solution

Utilize convex optimized algorithm to find the solution the fault-tolerant anti-interference controller gain battle array of multi-source EVAC; Given initial value x (0), e _w(0) and e _F(0), adjustable output matrix [C ₀C ₁C ₂], non-linear weight parameter λ disturbs inhibition degree γ ₁, γ ₂And γ ₃, find the solution following protruding optimization problem:

$\min \left(\begin{array}{ccc}\left(\begin{array}{cc}{x}^T\left(0\right)& e^T\left(0\right)\end{array}\right)& \left(\begin{array}{cc}{P}_1& \\ & P_2\end{array}\right)& {\left(\begin{array}{cc}{x}^T\left(0\right)& e^T\left(0\right)\end{array}\right)}^T\end{array}\right)$

$\mathrm{\Φ}=\left[\begin{array}{cccccccc}{\mathrm{\Φ}}_{11}& B_{1}G& -E& 0& B_2& 0& {\mathrm{\Φ}}_{18}& P_1{C}_0^T\\ *& {\mathrm{\Φ}}_{22}& 0& P_2{H}_1& R_2{B}_2& P_2{H}_2& \mathrm{\λ}{\left(U{B}_{1}G\right)}^T& C^T\\ *& *& -{\mathrm{\λ}}^{2}I& 0& 0& 0& \mathrm{\λ}{\left(\mathrm{UE}\right)}^T& 0\\ *& *& *& -{\mathrm{\γ}}_1^2& 0& 0& 0& 0\\ *& *& *& *& -{\mathrm{\γ}}_2^2& 0& \mathrm{\λ}{\left({\mathrm{UB}}_2\right)}^T& 0\\ *& *& *& *& *& -{\mathrm{\γ}}_3^{2}I& 0& 0\\ *& *& *& *& *& *& -I& 0\\ *& *& *& *& *& *& *& -I\end{array}\right]$

Wherein: e (0)=[e _w(0) e _F(0)] ^T, Φ ₁₁=(AP ₁+ B ₁R ₁)+(AP ₁+ B ₁R ₁) ^T, Φ ₂₂=(P ₂W ₁+ R ₂B ₁G)+(P ₂W ₁+ R ₂B ₁G) ^T, Φ ₁₈=λ (AP ₁+ B ₁R ₁) ^T, C=[C ₁C ₂], G=[E I], H ₁=[B ₃0] ^T, H ₂=[0 1] ^TSymbol * representes the symmetry blocks of appropriate section in the symmetric matrix, find the solution P ₁, P ₂, R ₁And R ₂, then interference observer and fault diagnosis observer gain battle array does $\left[\begin{array}{c}L\\ K\end{array}\right]=P_2^{-1}{R}_2,$ State feedback controller gain battle array is

The content of not doing in the instructions of the present invention to describe in detail belongs to this area professional and technical personnel's known prior art.

Claims (3)

1. the fault-tolerant anti-interference control method of a multi-source EVAC is characterized in that may further comprise the steps: at first, design error failure diagnostic observations device estimate and bucking-out system in time accident barrier; Next, but the modeling in estimation of design interference observer and the counteracting multi-source EVAC is disturbed; Once more, design robust H _∞State feedback controller suppress in the multi-source EVAC can not the modeling random disturbance, Fault Estimation sum of errors Interference Estimation error; At last, based on fault diagnosis observer, interference observer and robust H _∞State feedback controller designs fault-tolerant anti-interference controller; Concrete steps are following:

The first step is built the kinetic model that contains the multi-source EVAC, and is write as state-space expression

, build its system dynamics model, and it is following to be write as state-space expression to the multi-source EVAC that contains that but the modeling of the barrier of accident is sometimes disturbed and can not the modeling random disturbance:

$\stackrel{\·}{x}\left(t\right)+\mathrm{Ef}(\stackrel{\·}{x},t)=\mathrm{Ax}\left(t\right)+B_1(u\left(t\right)+d_1\left(t\right)+F\left(t\right))+B_2{d}_2\left(t\right)$

Wherein, x (t) is a multi-source EVAC state variable, and u (t) is control input, d ₁(t) but be that modeling is disturbed, F (t) for the time accident hinder d ₂(t) be can not the modeling random disturbance, A, E, B ₁And B ₂Be the matrix of known dimension, For system's nonlinear terms and satisfy the Lipschitz condition, external model description disturbance d ₁(t) by following external disturbance model ∑ ₁Expression:

${\mathrm{\Σ}}_1:\left\{\begin{array}{c}{d}_1\left(t\right)=\mathrm{Vw}\left(t\right)\\ \stackrel{\·}{w}\left(t\right)=\mathrm{Ww}\left(t\right)+B_3\mathrm{\δ}\left(t\right)\end{array}\right.$

Wherein, w (t) but be the state variable of modeling interference model, but V is the output matrix of modeling interference model, but W representes system's battle array of modeling interference model, δ (t) be energy bounded can not the modeling random disturbance, B ₃Be gain battle array that can not the modeling random disturbance;

Second step, design error failure diagnostic observations device

Time accident in the multi-source EVAC hinders F (t); Design error failure diagnostic observations device is estimated in real time it, and is tried to achieve estimated value

and then obtain Fault Estimation error

The 3rd step, the design interference observer

But disturb d to the modeling in the multi-source EVAC ₁(t), the design interference observer is estimated in real time it, and is tried to achieve estimated value And then obtain the Interference Estimation error

Estimated value for w (t);

The 4th step, design robust H _∞State feedback controller

To in the multi-source EVAC can not modeling random disturbance d ₂(t), Fault Estimation error e _F(t) and the Interference Estimation error e _w(t), design robust H _∞State feedback controller suppresses it, and controller architecture is following:

u _f(t)＝Mx(t)

Wherein, u _f(t) be robust H _∞The STATE FEEDBACK CONTROL input, M is a state feedback controller gain battle array undetermined;

In the 5th step, design fault-tolerant anti-interference controller

Design fault-tolerant anti-interference controller, the time accident in the system is hindered F (t) but and modeling interference d ₁(t) offset, can not modeling random disturbance d ₂(t), Fault Estimation error e _F(t) and the Interference Estimation error e _w(t) suppress, fault-tolerant anti-interference controller structure is following:

$u\left(t\right)=u_f\left(t\right)-\hat{F}\left(t\right)-{\hat{d}}_1\left(t\right)$

Then the multi-source EVAC can be expressed as:

$\stackrel{\·}{x}\left(t\right)+\mathrm{Ef}(\stackrel{\·}{x},t)=(A+B_{1}M)x\left(t\right)+B_{1}V{e}_w\left(t\right)+B_1{e}_F\left(t\right)+B_2{d}_2\left(t\right)$

But the system estimation error equation of arrangement modeling interference model with the time accident system estimation error equation that hinders following:

${\stackrel{\·}{e}}_w\left(t\right)=(W+{\mathrm{LB}}_{1}V)e_w\left(t\right)+{\mathrm{LB}}_1{e}_F\left(t\right)+{\mathrm{LB}}_2{d}_2\left(t\right)+B_3\mathrm{\δ}\left(t\right)$

${\stackrel{\·}{e}}_F\left(t\right)={\mathrm{KB}}_{1}V{e}_w\left(t\right)+{\mathrm{KB}}_1{e}_F\left(t\right)+{\mathrm{KB}}_2{d}_2\left(t\right)+\stackrel{\·}{F}\left(t\right)$

But join list state the multi-source EVAC, the time accident barrier system estimation error equation and the system estimation error equation that modeling is disturbed obtain closed-loop system:

$\left\{\begin{array}{c}\left(\begin{array}{c}\stackrel{\·}{x}\left(t\right)\\ {\stackrel{\·}{e}}_w\left(t\right)\\ {\stackrel{\·}{e}}_F\left(t\right)\end{array}\right)=\left(\begin{array}{ccc}A+B_{1}M& B_{1}V& B_1\\ 0& W+{\mathrm{LB}}_{1}V& {\mathrm{LB}}_1\\ 0& {\mathrm{KB}}_{1}V& {\mathrm{KB}}_1\end{array}\right)\left(\begin{array}{c}x\left(t\right)\\ e_w\left(t\right)\\ e_F\left(t\right)\end{array}\right)+\left(\begin{array}{c}{B}_2\\ {\mathrm{LB}}_2\\ {\mathrm{KB}}_2\end{array}\right)d_2\left(t\right)+\left(\begin{array}{c}0\\ B_3\\ 0\end{array}\right)\mathrm{\δ}\left(t\right)+\left(\begin{array}{c}0\\ 0\\ 1\end{array}\right)\stackrel{\·}{F}\left(t\right)-\left(\begin{array}{c}E\\ 0\\ 0\end{array}\right)f\left(\stackrel{\·}{x}\left(t\right)\right)\\ z_{\∞}\left(t\right)=\left[\begin{array}{ccc}{C}_0& C_1& C_2\end{array}\right]\left(\begin{array}{c}x\left(t\right)\\ e_w\left(t\right)\\ e_F\left(t\right)\end{array}\right)\end{array}\right.$

Wherein, z _∞(t) be H _∞Performance is with reference to output, [C ₀C ₁C ₂] be H _∞The adjustable output matrix of performance;

In the 6th step, gain matrix is found the solution

Utilize convex optimized algorithm to find the solution the fault-tolerant anti-interference controller gain battle array of multi-source EVAC; Given initial value x (0), e _w(0) and e _F(0), adjustable output matrix [C ₀C ₁C ₂], non-linear weight parameter λ disturbs inhibition degree γ ₁, γ ₂And γ ₃, find the solution following protruding optimization problem:

$\min \left(\begin{array}{ccc}\left(\begin{array}{cc}{x}^T\left(0\right)& e^T\left(0\right)\end{array}\right)& \left(\begin{array}{cc}{P}_1& \\ & P_2\end{array}\right)& {\left(\begin{array}{cc}{x}^T\left(0\right)& e^T\left(0\right)\end{array}\right)}^T\end{array}\right)$

$\mathrm{\Φ}=\left[\begin{array}{cccccccc}{\mathrm{\Φ}}_{11}& B_{1}G& -E& 0& B_2& 0& {\mathrm{\Φ}}_{18}& P_1{C}_0^T\\ *& {\mathrm{\Φ}}_{22}& 0& P_2{H}_1& R_2{B}_2& P_2{H}_2& \mathrm{\λ}{\left(U{B}_{1}G\right)}^T& C^T\\ *& *& -{\mathrm{\λ}}^{2}I& 0& 0& 0& \mathrm{\λ}{\left(\mathrm{UE}\right)}^T& 0\\ *& *& *& -{\mathrm{\γ}}_1^2& 0& 0& 0& 0\\ *& *& *& *& -{\mathrm{\γ}}_2^2& 0& \mathrm{\λ}{\left({\mathrm{UB}}_2\right)}^T& 0\\ *& *& *& *& *& -{\mathrm{\γ}}_3^{2}I& 0& 0\\ *& *& *& *& *& *& -I& 0\\ *& *& *& *& *& *& *& -I\end{array}\right]$

Wherein: e (0)=[e _w(0) e _F(0)] ^T, Φ ₁₁=(AP ₁+ B ₁R ₁)+(AP ₁+ B ₁R ₁) ^T, Φ ₂₂=(P ₂W ₁+ R ₂B ₁G)+(P ₂W ₁+ R ₂B ₁G) ^T, Φ ₁₈=λ (AP ₁+ B ₁R ₁) ^T, C=[C ₁C ₂], G=[E I], H ₁=[B ₃0] ^T, H ₂=[0 1] ^TSymbol * representes the symmetry blocks of appropriate section in the symmetric matrix, find the solution P ₁, P ₂, R ₁And R ₂, then interference observer and fault diagnosis observer gain battle array does $\left[\begin{array}{c}L\\ K\end{array}\right]=P_2^{-1}{R}_2,$ State feedback controller gain battle array is

2. the fault-tolerant anti-interference control method of a kind of multi-source EVAC according to claim 1 is characterized in that: the fault diagnosis observer structure in the said step 2 is following:

$\left\{\begin{array}{c}\hat{F}\left(t\right)=\mathrm{\τ}-p\left(x\right)\\ \stackrel{\·}{\mathrm{\τ}}={\mathrm{KB}}_1(\mathrm{\τ}-p\left(x\right))+K[\mathrm{Ax}\left(t\right)+B_{1}u\left(t\right)+B_1{\hat{d}}_1\left(t\right)-\mathrm{Ef}\left(\stackrel{\·}{x}\left(t\right)\right)]\end{array}\right.$

Wherein,

K is a fault diagnosis observer gain matrix undetermined, and ε (t) is an auxiliary variable.

3. the fault-tolerant anti-interference control method of a kind of multi-source EVAC according to claim 1 is characterized in that: the interference observer structure in the said step 3 is following:

$\left\{\begin{array}{c}{\hat{d}}_1\left(t\right)=V\hat{w}\left(t\right)\\ \hat{w}\left(t\right)=v\left(t\right)-\mathrm{Lx}\left(t\right)\\ \stackrel{\·}{v}\left(t\right)=(W+{\mathrm{LB}}_{1}V)(v\left(t\right)-\mathrm{Lx}\left(t\right))+L[\mathrm{Ax}\left(t\right)+B_{1}u\left(t\right)+B_1\hat{F}\left(t\right)-\mathrm{Ef}(\stackrel{\·}{x},t)]\end{array}\right.$

Wherein, v (t) is an auxiliary variable, and L is an interference observer gain matrix undetermined, but V is the output matrix of modeling interference model, but W representes system's battle array of modeling interference model.

Images