CN102736517A - Direct adaptive reconstruction control method for three-degree-of-freedom helicopter - Google Patents

Abstract

The invention relates to a direct adaptive reconstruction control method for a three-degree-of-freedom helicopter. A model structure of a direct adaptive reconstruction control law is improved, and a series of calculation steps for controlled variables of the three-degree-of-freedom helicopter are performed, so that the response of the three-degree-of-freedom helicopter to failure conditions is quickened, and the flight safety of the helicopter is ensured.

Description

A kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method

Technical field

The present invention relates to a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method.

Background technology

Helicopter flight control system is the important component part of helicopter, and very crucial effect is played in the flying quality and the security of helicopter.Because the helicopter flight control system parts are more, the possibility that breaks down is higher, and therefore, the fault-tolerant control technology of studying its flight control system is vital.When fault takes place, through reconstruct control, guarantee that helicopter recovers smooth flight in a short period of time, improve the security of flight.

The purpose of the reconstruct of flight control system control is, makes helicopter have higher survivability in flight course as when actuator failures or other unknown failure occurring.Adaptive Reconfigurable Control is divided into two kinds of methods again: direct adaptive control method and indirect self-adaptive control method; The indirect self-adaptive control method need be carried out identification earlier to the parameter of controlled device, and a plurality of controlling schemes need be provided, and promptly at first the flight parameter direct-on-line is estimated, confirms controller parameter according to its result.On the contrary, the direct adaptive control method need not definite systematic parameter, has following advantage: technological comparative maturity, and algorithm is simple, can handle one big type of fault, for example the uncertainty of uncertain, the structure of systematic parameter and external interference etc.; Through the parameter that the systematic error response comes online adjustment control, guaranteed the rapidity of fault-tolerant control; For actuator failures, do not need independent fault diagnosis and recognition module, the uncertainty that can avoid so cause and cause the reduction of fault-tolerant control performance (or index).In the existing Adaptive Reconfigurable Control method, for the controlled quentity controlled variable employing model of helicopter Try to achieve, wherein u is the controlled quentity controlled variable of Three Degree Of Freedom helicopter, and x is the state vector of Three Degree Of Freedom helicopter, and r is the input vector of Three Degree Of Freedom helicopter reference model,

Be Fault Compensation vector, K _xBe state vector feedback gain matrix, K _rBe the feedforward gain matrix; But adopt this model system reply failure situations rapidity not enough; This just directly has influence on the capacity of self-regulation of helicopter when the reply fault; And the capacity of self-regulation of helicopter when the reply fault directly has influence on the flight safety problem of himself, and the speed of therefore calculating for the helicopter controlled quentity controlled variable has determined the flight safety problem of helicopter.

Summary of the invention

Technical matters to be solved by this invention provides a kind of convenient in application, and can effectively improve Three Degree Of Freedom helicopter controlled quentity controlled variable arithmetic speed to Three Degree Of Freedom helicopter direct adaptive reconstruct control method.

The present invention adopts following technical scheme in order to solve the problems of the technologies described above: the present invention has designed a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method; Comprise direct adaptive reconstruct control law; Comprising the controlled quentity controlled variable u of Three Degree Of Freedom helicopter, the feedforward gain matrix K _r, Three Degree Of Freedom helicopter reference model input vector r, Fault Compensation vector

The state vector feedback gain matrix K that also comprises reference model ₁, the feedforward gain matrix K ₂, K ₂=K _r, output error gain matrix K ₃, the state vector x of Three Degree Of Freedom helicopter reference model _m, the output error e of Three Degree Of Freedom helicopter _y, the model structure of said direct adaptive reconstruct control law is following:

$u=K_1{x}_m+K_{2}r+K_3{e}_y+\hat{f}$

Wherein, K ₁, K ₂, K ₃With

Satisfy following condition:

${\stackrel{\·}{K}}_1=-{\mathrm{\Γ}}_1{B}_p^{T}P{\mathrm{ex}}_m^T,$ ${\stackrel{\·}{K}}_2=-{\mathrm{\Γ}}_2{B}_p^{T}P{\mathrm{er}}^T,$ ${\stackrel{\·}{K}}_3=-{\mathrm{\Γ}}_3{B}_p^{T}P{\mathrm{ee}}_y^T,$ $\stackrel{\·}{\hat{f}}=-{\mathrm{\Γ}}_4{B}_p^T\mathrm{Pe}$

Wherein, weight vector Γ _iBe the diagonal angle positive definite matrix, i=1 ..., 4, P is an equation

The positive definite symmetric solution, Q is any positive definite symmetric matrices here, e is the state error of Three Degree Of Freedom helicopter, B _pThe real matrix of corresponding dimension during for Three Degree Of Freedom helicopter actual motion, A _eIt is any stable system matrix.

As a kind of optimal technical scheme of the present invention: the output error e of said Three Degree Of Freedom helicopter _yError e according to definition _y=y-y _mTry to achieve, wherein, y is the output vector of Three Degree Of Freedom helicopter model, obtains y through angle-measuring equipment _mOutput vector for Three Degree Of Freedom helicopter reference model.

As a kind of optimal technical scheme of the present invention: the output vector y of said Three Degree Of Freedom helicopter reference model _mAccording to reference model y _m=Cx _mTry to achieve, C is the real matrix of Three Degree Of Freedom helicopter model corresponding dimension when its equilibrium position, obtains according to Three Degree Of Freedom helicopter model.

As a kind of optimal technical scheme of the present invention: the state error e of said Three Degree Of Freedom helicopter is according to the error e=x-x of definition _mTry to achieve, wherein, x is the state vector of Three Degree Of Freedom helicopter model, obtains through angle-measuring equipment.

As a kind of optimal technical scheme of the present invention: the state vector x of said Three Degree Of Freedom helicopter reference model _mAccording to reference model

Try to achieve, wherein, A _mBe any stable system matrix, for add the system state matrix that makes system stability after the feedback of status, B to Three Degree Of Freedom helicopter model _mFor the real matrix of corresponding dimension, get B _m=B, A, B are the real matrix of Three Degree Of Freedom helicopter model corresponding dimension when its equilibrium position, obtain according to Three Degree Of Freedom helicopter model.

As a kind of optimal technical scheme of the present invention: the input vector r of said Three Degree Of Freedom helicopter reference model sets according to the desired output of Three Degree Of Freedom helicopter model.

As a kind of optimal technical scheme of the present invention: the real matrix B of corresponding dimension during said Three Degree Of Freedom helicopter actual motion _p=B.

As a kind of optimal technical scheme of the present invention: the state vector x of said Three Degree Of Freedom helicopter model is following:

$x^T=[\mathrm{\ϵ},p,\mathrm{\λ},\frac{\∂}{\∂t}\mathrm{\ϵ},\frac{\∂}{\∂t}p,\frac{\∂}{\∂t}\mathrm{\λ}]$

Wherein, ε, p, λ, are expressed as up-down angle, roll angle, crab angle, up-down angle angular velocity, roll angle angular velocity and crab angle angular velocity respectively.

As a kind of optimal technical scheme of the present invention: the output vector y of said Three Degree Of Freedom helicopter model is following:

y ^T＝[ε,p,λ]

Wherein, ε, p, λ are expressed as up-down angle, roll angle and crab angle respectively.

As a kind of optimal technical scheme of the present invention: said reference model is a digital model.

On end, a kind of Three Degree Of Freedom helicopter direct adaptive reconstruct control method that is directed against according to the invention adopts above technical scheme compared with prior art, has following technique effect:

(1) make the application of adaptive control laws become convenient; Improved computing velocity for Three Degree Of Freedom helicopter controlled quentity controlled variable; Better capacity of self-regulation when making the Three Degree Of Freedom helicopter have the reply fault has guaranteed the flight safety of Three Degree Of Freedom helicopter;

(2) fault parameter can not survey and with extraneous unknown disturbance situation under; Take all factors into consideration uncertain factors such as helicopter modeling error; Through the control law form being improved the bilinear problem that has solved classic method; And, guaranteed the rapidity of fault-tolerant control through the parameter that the systematic error response comes online adjusting control law;

(3), do not need independent fault diagnosis and recognition module, the uncertainty that can avoid so cause and the reduction of the fault-tolerant control performance (or index) that causes for actuator failures.

Description of drawings

Fig. 1 is for being directed against the module process flow diagram of Three Degree Of Freedom helicopter emulation among the present invention;

Error variation diagram when Fig. 2 takes place for a kind of real system fault to Three Degree Of Freedom helicopter direct adaptive reconstruct control law that does not add the present invention's design;

Error when Fig. 3 takes place for a kind of real system fault to Three Degree Of Freedom helicopter direct adaptive reconstruct control law that adds the present invention's design changes.

Embodiment

Do further detailed explanation below in conjunction with the Figure of description specific embodiments of the invention.

The present invention has designed a kind of Three Degree Of Freedom helicopter direct adaptive reconstruct control method that is directed against, and comprises direct adaptive reconstruct control law, comprising the controlled quentity controlled variable u of Three Degree Of Freedom helicopter, and the feedforward gain matrix K _r, Three Degree Of Freedom helicopter reference model input vector r, Fault Compensation vector

The state vector feedback gain matrix K that also comprises reference model ₁, the feedforward gain matrix K ₂, K ₂=K _r, output error gain matrix K ₃, the state vector x of Three Degree Of Freedom helicopter reference model _m, the output error e of Three Degree Of Freedom helicopter _y, the model structure of said direct adaptive reconstruct control law is following:

$u=K_1{x}_m+K_{2}r+K_3{e}_y+\hat{f}$

Wherein, K ₁, K ₂, K ₃With Satisfy following condition:

${\stackrel{\·}{K}}_1=-{\mathrm{\Γ}}_1{B}_p^{T}P{\mathrm{ex}}_m^T,$ ${\stackrel{\·}{K}}_2=-{\mathrm{\Γ}}_2{B}_p^{T}P{\mathrm{er}}^T,$ ${\stackrel{\·}{K}}_3=-{\mathrm{\Γ}}_3{B}_p^{T}P{\mathrm{ee}}_y^T,$ $\stackrel{\·}{\hat{f}}=-{\mathrm{\Γ}}_4{B}_p^T\mathrm{Pe}$

Wherein, weight vector Γ _iBe the diagonal angle positive definite matrix, i=1 ..., 4, P is an equation

The positive definite symmetric solution, Q is any positive definite symmetric matrices here, e is the state error of Three Degree Of Freedom helicopter, B _pThe real matrix of corresponding dimension during for Three Degree Of Freedom helicopter actual motion, A _eIt is any stable system matrix.

As a kind of optimal technical scheme of the present invention: the output error e of said Three Degree Of Freedom helicopter _yError e according to definition _y=y-y _mTry to achieve, wherein, y is the output vector of Three Degree Of Freedom helicopter model, obtains y through angle-measuring equipment _mOutput vector for Three Degree Of Freedom helicopter reference model.

As a kind of optimal technical scheme of the present invention: the output vector y of said Three Degree Of Freedom helicopter reference model _mAccording to reference model y _m=Cx _mTry to achieve, C is the real matrix of Three Degree Of Freedom helicopter model corresponding dimension when its equilibrium position, obtains according to Three Degree Of Freedom helicopter model.

As a kind of optimal technical scheme of the present invention: the state error e of said Three Degree Of Freedom helicopter is according to the error e=x-x of definition _mTry to achieve, wherein, x is the state vector of Three Degree Of Freedom helicopter model, obtains through angle-measuring equipment.

As a kind of optimal technical scheme of the present invention: the state vector x of said Three Degree Of Freedom helicopter reference model _mAccording to reference model

Try to achieve, wherein, A _mBe any stable system matrix, for add the system state matrix that makes system stability after the feedback of status, B to Three Degree Of Freedom helicopter model _mFor the real matrix of corresponding dimension, get B _m=B, A, B are the real matrix of Three Degree Of Freedom helicopter model corresponding dimension when its equilibrium position, obtain according to Three Degree Of Freedom helicopter model.

As a kind of optimal technical scheme of the present invention: the input vector r of said Three Degree Of Freedom helicopter reference model sets according to the desired output of Three Degree Of Freedom helicopter model.

As a kind of optimal technical scheme of the present invention: the real matrix B of corresponding dimension during said Three Degree Of Freedom helicopter actual motion _p=B.

As a kind of optimal technical scheme of the present invention: said reference model is a digital model.

Wherein, according to the Liapunov stability method of proof, be the assurance system stability, and satisfy

The model of said Three Degree Of Freedom helicopter direct adaptive reconstruct control law $u=K_1{x}_m+K_{2}r+K_3{e}_y+\hat{f}$ Middle K ₁, K ₂, K ₃With

Satisfy respectively ${\stackrel{\·}{K}}_1=-{\mathrm{\Γ}}_1{B}_p^{T}P{\mathrm{Ex}}_m^T,$ ${\stackrel{\·}{K}}_2=-{\mathrm{\Γ}}_2{B}_p^{T}P{\mathrm{Er}}^T,$ ${\stackrel{\·}{K}}_3=-{\mathrm{\Γ}}_3{B}_p^{T}P{\mathrm{Ee}}_y^T,$ $\stackrel{\·}{\hat{f}}=-{\mathrm{\Γ}}_4{B}_p^T\mathrm{Pe}$ Condition, be to adopt following method of proof to obtain:

Set the characteristic information of a certain actuator failures, comprise fault type, fault start time, fault concluding time and fault size, said fault is carried out the software injection through the fault injection module, and fault model is:

$\stackrel{\·}{x}=A_{p}x+B_p\mathrm{\Λu}+B_p\mathrm{\Λ}(I-\mathrm{\σ})\stackrel{\‾}{u}+B_{p}d,$

y＝Cx

Here σ=diag{ σ ₁σ ₂σ _m), Λ=diag{ λ ₁λ ₂λ _m); σ _iBe the maneuverability coefficient of actuator, be used for judging that can actuator be controlled, if σ _i=1, then actuator is normal, and under stuck (LIP) and saturated (HOF) situation σ _i=0; λ _iBe that actuator is renderd a service coefficient, λ under damage (LOE) situation _i∈ [0,1], i=1 wherein, 2 ..., m; Said fault is with unknown matrix Λ ∈ R ^{M * m}, σ ∈ R ^{M * m}And unknown vector

Describe,

The stuck position of expression actuator, unknown vector d is external interference and modeling error, C is the real matrix of Three Degree Of Freedom helicopter body corresponding dimension when its equilibrium position, A _pThe real matrix of corresponding dimension during for Three Degree Of Freedom helicopter actual motion, I is a unit matrix, order

Obtain: $\stackrel{\·}{x}=A_{p}x+B_p\mathrm{\Λ\; u}+B_{p}f.$

Select reference model:

${\stackrel{\·}{x}}_m=A_m{x}_m+B_{m}r,y_m={\mathrm{Cx}}_m---\left(3.6\right)$

x _m, r is respectively the state vector of Three Degree Of Freedom helicopter reference model and the input vector of Three Degree Of Freedom helicopter reference model, A _m, B _mReal matrix for the corresponding dimension of Three Degree Of Freedom helicopter reference model.

Model structure according to following Three Degree Of Freedom helicopter direct adaptive reconstruct control law:

$u=K_1{x}_m+K_{2}r+K_3{e}_y+\hat{f}---\left(3.8\right)$

The state error e of wherein said Three Degree Of Freedom helicopter and the output error e of said Three Degree Of Freedom helicopter _yFollowing gained:

e＝x-x _m e _y＝y-y _m (3.9)

Adopt the model structure (3.8) of Three Degree Of Freedom helicopter direct adaptive reconstruct control law, obtain the dynamic equation of the state error of Three Degree Of Freedom helicopter:

$\stackrel{\·}{e}=\stackrel{\·}{x}-{\stackrel{\·}{x}}_m$

$=A_{p}x+B_p\mathrm{\Λu}+B_{p}f-A_m{x}_m-B_{m}r$

$=(A_p+B_p\mathrm{\Λ}{K}_{3}C)e+(A_p-A_m+B_p\mathrm{\Λ}{K}_1)x_m+(B_p\mathrm{\Λ}{K}_2-B_m)r+B_p(\mathrm{\Λ}\hat{f}+f)$

(3.10)

Based on the model reference matching condition, do as giving a definition:

$A_p+B_p\mathrm{\&Lambda;}{\mathrm{<figure-callout\; id="3"\; label="K"\; filenames="120619144811.png,120621085146.png"\; state="\{\{state\}\}">K</figure-callout>}}_3^{*}C=A_e$ $A_p+B_p\mathrm{\&Lambda;}{K}_1^{*}C=A_m$

$B_p\mathrm{\&Lambda;}{\mathrm{<figure-callout\; id="2"\; label="K"\; filenames="120619144811.png"\; state="\{\{state\}\}">K</figure-callout>}}_2^{*}=B_m$ $\mathrm{\&Lambda;}{f}^{*}+f=0$

(3.11)

Here f ^*The value of each gain matrix when representation model matees fully, A _eBe any stable system matrix, A _p, B _PThe real matrix of corresponding dimension during for Three Degree Of Freedom helicopter actual motion, A _mReal matrix for the corresponding dimension of Three Degree Of Freedom helicopter reference model.Direct adaptive reconstruct control law online updating K ₁, K ₂, K ₃With

Value, the state that makes system and output under the prerequisite of Liapunov stability, the state of track reference model and output.

Define following error matrix and vector:

${\tilde{K}}_1=K_1-K_1^{*}$ ${\tilde{K}}_2=K_2-{\mathrm{<figure-callout\; id="2"\; label="K"\; filenames="120619144811.png"\; state="\{\{state\}\}">K</figure-callout>}}_2^{*}$

${\tilde{K}}_3=K_3-{\mathrm{<figure-callout\; id="3"\; label="K"\; filenames="120619144811.png,120621085146.png"\; state="\{\{state\}\}">K</figure-callout>}}_3^{*}$ $\tilde{f}=\hat{f}-f^{*}$

(3.12)

With (3.11) and (3.12) formula substitution equations (3.10), obtain:

$\stackrel{\&CenterDot;}{e}=A_{e}e+B_p\mathrm{\&Lambda;}({\tilde{K}}_1{x}_m+{\tilde{K}}_{2}r+{\tilde{K}}_3{e}_y+\tilde{f})$

(3.13)

Consider following Lyapunov function:

$V=\frac{1}{2}[e^T\mathrm{Pe}+\mathrm{Tr}\left({\tilde{K}}_1^T{\mathrm{\&Gamma;}}_1^{-1}\mathrm{\&Lambda;}{\tilde{K}}_1\right)+\mathrm{Tr}\left({\tilde{K}}_2^T{\mathrm{\&Gamma;}}_2^{-1}\mathrm{\&Lambda;}{\tilde{K}}_2\right)+\mathrm{Tr}\left({\tilde{K}}_3^T{\mathrm{\&Gamma;}}_3^{-1}\mathrm{\&Lambda;}{\tilde{K}}_3\right)+{\tilde{f}}^T{\mathrm{\&Gamma;}}_4^{-1}\mathrm{\&Lambda;}\tilde{f}]$

(3.18)

Its differential form is following:

$\stackrel{\&CenterDot;}{V}=e^{T}P\stackrel{\&CenterDot;}{e}+\mathrm{tr}\left({\tilde{K}}_1^T{\mathrm{\&Gamma;}}_1^{-1}\mathrm{\&Lambda;}{\stackrel{\&CenterDot;}{\tilde{K}}}_1\right)+\mathrm{tr}\left({\tilde{K}}_2^T{\mathrm{\&Gamma;}}_2^{-1}\mathrm{\&Lambda;}{\stackrel{\&CenterDot;}{\tilde{K}}}_2\right)+\mathrm{tr}\left({\tilde{K}}_3^T{\mathrm{\&Gamma;}}_3^{-1}\mathrm{\&Lambda;}{\stackrel{\&CenterDot;}{\tilde{K}}}_3\right)+{\tilde{f}}^T{\mathrm{\&Gamma;}}_4^{-1}\mathrm{\&Lambda;}\stackrel{\&CenterDot;}{\tilde{f}}$

$=-\frac{1}{2}{e}^T\mathrm{Qe}+\mathrm{tr}[{\tilde{K}}_1^T{\mathrm{\&Gamma;}}_1^{-1}\mathrm{\&Lambda;}({\stackrel{\&CenterDot;}{\tilde{K}}}_1+{\mathrm{\&Gamma;}}_1{B}_p^T{\mathrm{Pex}}_m^T)+{\tilde{K}}_2^T{\mathrm{\&Gamma;}}_2^{-1}\mathrm{\&Lambda;}({\stackrel{\&CenterDot;}{\tilde{K}}}_2+{\mathrm{\&Gamma;}}_2{B}_p^T{\mathrm{Per}}^T)+{\tilde{K}}_3^T{\mathrm{\&Gamma;}}_3^{-1}\mathrm{\&Lambda;}({\stackrel{\&CenterDot;}{K}}_3$

$+{\mathrm{\&Gamma;}}_3{B}_p^T{\mathrm{Pee}}_y^T\left)\right]+{\tilde{f}}^T{\mathrm{\&Gamma;}}_4^{-1}(\stackrel{\&CenterDot;}{\hat{f}}+{\mathrm{\&Gamma;}}_4{B}_p^T\mathrm{Pe})$

If choose: ${\stackrel{\&CenterDot;}{K}}_1=-{\mathrm{\&Gamma;}}_1{B}_p^T{\mathrm{Pex}}_m^T$

${\stackrel{\&CenterDot;}{K}}_2=-{\mathrm{\&Gamma;}}_2{B}_p^{T}P{\mathrm{er}}^T$

${\stackrel{\&CenterDot;}{K}}_3=-{\mathrm{\&Gamma;}}_3{B}_p^{T}P{\mathrm{ee}}_y^T$

$\stackrel{\&CenterDot;}{\hat{f}}=-{\mathrm{\&Gamma;}}_4{B}_p^T\mathrm{Pe}$

Then: $\stackrel{\&CenterDot;}{V}=-\frac{1}{2}{e}^T\mathrm{Qe}<0$

So just can guarantee the stable of the Three Degree Of Freedom Helicopter System overall situation.

As a kind of optimal technical scheme of the present invention: the state vector x of said Three Degree Of Freedom helicopter model is following:

$x^T=[\mathrm{\&epsiv;},p,\mathrm{\&lambda;},\frac{\&PartialD;}{\&PartialD;t}\mathrm{\&epsiv;},\frac{\&PartialD;}{\&PartialD;t}p,\frac{\&PartialD;}{\&PartialD;t}\mathrm{\&lambda;}]$

Wherein, ε, p, λ,

are expressed as up-down angle, roll angle, crab angle, up-down angle angular velocity, roll angle angular velocity and crab angle angular velocity respectively.

As a kind of optimal technical scheme of the present invention: the output vector y of said Three Degree Of Freedom helicopter model is following:

y ^T＝[ε,p,λ]

Wherein, ε, p, λ are expressed as up-down angle, roll angle and crab angle respectively.

Said Three Degree Of Freedom helicopter body satisfies following kinetic model when its equilibrium position:

$\frac{\&PartialD;}{\&PartialD;t}x=\mathrm{Ax}+\mathrm{Bu},y=\mathrm{Cx}+\mathrm{Du}$

Wherein A, B, C, D are the real matrix of Three Degree Of Freedom helicopter body corresponding dimension when its equilibrium position, can obtain through Three Degree Of Freedom helicopter model.

When carrying out emulation to the Three Degree Of Freedom helicopter; Comprise helicopter model, fault injection module, controller module, angle-measuring equipment, data acquisition module, ground monitoring module and actuator; As shown in Figure 1; Process of simulation comprises the steps that wherein the model of Three Degree Of Freedom helicopter direct adaptive reconstruct control law does $u=K_1{x}_m+K_{2}r+K_3{e}_y+\hat{f}:$

Step 1. is set the characteristic information of a certain actuator failures, comprising: fault type, fault start time, fault concluding time and the said fault of fault size are carried out the software injection through the fault injection module, and fault model is:

$\stackrel{\&CenterDot;}{x}=A_{p}x+B_p\mathrm{\&Lambda;u}+B_p\mathrm{\&Lambda;}(I-\mathrm{\&sigma;})\stackrel{\&OverBar;}{u}+B_{p}d,$

y＝Cx

Here σ=diag{ σ ₁σ ₂σ _m), Λ=diag{ λ ₁λ ₂λ _m), σ _iBe the maneuverability coefficient of actuator, be used for judging that can actuator be controlled, if σ _i=1, then actuator is normal, and under stuck (LIP) and saturated (HOF) situation σ _i=0; λ _iBe that actuator is renderd a service coefficient, λ under damage (LOE) situation _i∈ [0,1], i=1 wherein, 2 ..., m; Said fault is with unknown matrix Λ ∈ R ^{M * m}, σ ∈ R ^{M * m}And unknown vector

Describe, The stuck position of expression actuator, unknown vector d is external interference and modeling error, I is a unit matrix, order

Obtain:

Step 2. controller module calculates the controlled quentity controlled variable of Three Degree Of Freedom helicopter according to Three Degree Of Freedom helicopter direct adaptive reconstruct control law;

Step 3. fault injection module is in the fault start time of setting, and the fault characteristic information according to setting carries out signal Processing to the controlled quentity controlled variable that is calculated by controller module, produces the pseudoinstruction signal, outputs to actuator, realizes the actuator failures simulation;

Step 4. angle-measuring equipment obtains the attitude information of helicopter model in real time, and is transferred to data acquisition module, and data acquisition module is to the measured helicopter model attitude information of controller module output;

Step 5. controller module upgrades the parameter in the Three Degree Of Freedom helicopter direct adaptive reconstruct control law to the helicopter model attitude information that obtains, and recalculates the controlled quentity controlled variable of Three Degree Of Freedom helicopter according to the parameter after upgrading, and returns step 3;

In the middle of process of simulation, adopt the ground monitoring module can obtain the attitude information of each time of helicopter model in real time simultaneously, failure message and controlled quentity controlled variable information can stop emulation at any time.

A kind of Three Degree Of Freedom helicopter direct adaptive reconstruct control law that is directed against of the present invention's design is in the middle of the practical implementation process:

1. assumed fault occurs in 60s, the loss of voltage 20% of the afterbody of helicopter (back to) motor, and representation for fault is following form:

$F\left(V_b\right)=\left\{\begin{array}{cc}{V}_b& 0s\&le;t<60s\\ 0.8*V_b& t\&GreaterEqual;60s\end{array}\right.;$

2. the error variation when a kind of real system fault to Three Degree Of Freedom helicopter direct adaptive reconstruct control law front and back of system's adding the present invention design takes place is like Fig. 2, shown in Figure 3;

3. add the model reference adaptive reconfigurable controller, system in stable condition weakens before not adding controller with the vibration of reference state error, and the overshoot of error reduces, the performance of system be improved significantly.

Combine accompanying drawing that embodiment of the present invention has been done detailed description above, but the present invention is not limited to above-mentioned embodiment, in the ken that those of ordinary skills possessed, can also under the prerequisite that does not break away from aim of the present invention, makes various variations.

Claims (10)

1. one kind is directed against Three Degree Of Freedom helicopter direct adaptive reconstruct control method, comprises direct adaptive reconstruct control law, comprising the controlled quentity controlled variable u of Three Degree Of Freedom helicopter, and the feedforward gain matrix K _r, Three Degree Of Freedom helicopter reference model input vector r, Fault Compensation vector

It is characterized in that: the state vector feedback gain matrix K that also comprises reference model ₁, the feedforward gain matrix K ₂, K ₂=K _r, output error gain matrix K ₃, the state vector x of Three Degree Of Freedom helicopter reference model _m, the output error e of Three Degree Of Freedom helicopter _y, the model structure of said direct adaptive reconstruct control law is following:

$u=K_1{x}_m+K_{2}r+K_3{e}_y+\hat{f}$

Wherein, K ₁, K ₂, K ₃With

Satisfy following condition:

${\stackrel{\&CenterDot;}{K}}_1=-{\mathrm{\&Gamma;}}_1{B}_p^{T}P{\mathrm{ex}}_m^T,$ ${\stackrel{\&CenterDot;}{K}}_2=-{\mathrm{\&Gamma;}}_2{B}_p^{T}P{\mathrm{er}}^T,$ ${\stackrel{\&CenterDot;}{K}}_3=-{\mathrm{\&Gamma;}}_3{B}_p^{T}P{\mathrm{ee}}_y^T,$ $\stackrel{\&CenterDot;}{\hat{f}}=-{\mathrm{\&Gamma;}}_4{B}_p^T\mathrm{Pe}$

Wherein, weight vector Γ _iBe the diagonal angle positive definite matrix, i=1 ..., 4, P is an equation

The positive definite symmetric solution, Q is any positive definite symmetric matrices here, e is the state error of Three Degree Of Freedom helicopter, B _pThe real matrix of corresponding dimension during for Three Degree Of Freedom helicopter actual motion, A _eIt is any stable system matrix.

2. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the output error e of said Three Degree Of Freedom helicopter _yError e according to definition _y=y-y _mTry to achieve, wherein, y is the output vector of Three Degree Of Freedom helicopter model, obtains y through angle-measuring equipment _mOutput vector for Three Degree Of Freedom helicopter reference model.

3. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 2, it is characterized in that: the output vector y of said Three Degree Of Freedom helicopter reference model _mAccording to reference model y _m=Cx _mTry to achieve, C is the real matrix of Three Degree Of Freedom helicopter model corresponding dimension when its equilibrium position, obtains according to Three Degree Of Freedom helicopter model.

4. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the state error e of said Three Degree Of Freedom helicopter is according to the error e=x-x of definition _mTry to achieve, wherein, x is the state vector of Three Degree Of Freedom helicopter model, obtains through angle-measuring equipment.

5. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the state vector x of said Three Degree Of Freedom helicopter reference model _mAccording to reference model

Try to achieve, wherein, A _mBe any stable system matrix, for add the system state matrix that makes system stability after the feedback of status, B to Three Degree Of Freedom helicopter model _mFor the real matrix of corresponding dimension, get B _m=B, A, B are the real matrix of Three Degree Of Freedom helicopter model corresponding dimension when its equilibrium position, obtain according to Three Degree Of Freedom helicopter model.

6. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the input vector r of said Three Degree Of Freedom helicopter reference model sets according to the desired output of Three Degree Of Freedom helicopter model.

7. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the real matrix B of corresponding dimension during said Three Degree Of Freedom helicopter actual motion _p=B.

8. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the state vector x of said Three Degree Of Freedom helicopter model is following:

$x^T=[\mathrm{\&epsiv;},p,\mathrm{\&lambda;},\frac{\&PartialD;}{\&PartialD;t}\mathrm{\&epsiv;},\frac{\&PartialD;}{\&PartialD;t}p,\frac{\&PartialD;}{\&PartialD;t}\mathrm{\&lambda;}]$

Wherein, ε, p, λ, λ are expressed as up-down angle, roll angle, crab angle, up-down angle angular velocity, roll angle angular velocity and crab angle angular velocity respectively.

9. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: the output vector y of said Three Degree Of Freedom helicopter model is following:

y ^T＝[ε,p,λ]

Wherein, ε, p, λ are expressed as up-down angle, roll angle and crab angle respectively.

10. said a kind of to Three Degree Of Freedom helicopter direct adaptive reconstruct control method according to claim 1, it is characterized in that: said reference model is a digital model.

Images