CN103149929A - Fault diagnosing and tolerance control method for aircraft longitudinal movement - Google Patents

Abstract

The invention provides a fault diagnosing and tolerance control method for aircraft longitudinal movement. The method comprises the following steps of according to a model and a state combination index discriminant when an aircraft has no fault, carrying out the fault diagnosing; and when the aircraft has a fault, firstly, classifying the fault, analyzing the influence of the whole-aircraft parameters and the flight dynamic properties according to the determined fault type, and remodeling an abrupt-change system after the fault occurs, so as to obtain a new model description when the aircraft has the fault. The design method of a tolerance controller is characterized in that the output of a fault-free model can meet the requirements of military standards by the designed control amount; and when the aircraft has the fault, the new model output is stable by the control input which is formed by the control amount and the non-artificial additional control amount caused by the aircraft fault.

Description

Flight Altitude Moving fault diagnosis and fault tolerant control method

Technical field

The present invention relates to aircraft fault diagnosis and Fault-tolerant Control Design method, particularly relate to a kind of Flight Altitude Moving fault diagnosis and fault tolerant control method.

Background technology

Military spacecraft is a kind of important operational weapon, civilian aircraft is a kind of important vehicles, playing the part of more and more important role in national defence and daily life and work, the fault diagnosis research of aerocraft system more and more is being subject to people's attention; Fault diagnosis is of crucial importance for security requirement high system, detects in time the system failure, makes the timely reconstruct control law of self-repairing system, material damage and the casualties that can avoid system crash and cause thus.

At first grow up in the U.S. from early 1970s to resolve fault detection and diagnosis technology that redundancy takes as the leading factor.1971, at first the Beard of masschusetts, u.s.a Polytechnics has proposed to replace hardware redundancy with resolving redundancy in his PhD dissertation, and organize in vain by system, make system's closed-loop stabilization, obtain the new thought of system failure information by the output of comparative observation device, indicate the birth based on the fault diagnosis technology of resolving redundancy.The research for fault diagnosis abroad mainly concentrates on industrial system and aerospace, at present take the fault diagnosis of Study of Generalized system and nonlinear system as main, the method that adopts is very various, the sliding mode observer method is specifically arranged, neural net method, expert system method, small wave converting method, Unknown Input Observer method etc., and how several method is combined use.For the method for diagnosing faults of linear system comparative maturity, a cover has been arranged than more complete diagnostic method, and some actual application systems have been arranged.USAF has adopted fault diagnosis technology on fighter experimental.Domestic paper to aircraft fault diagnosis and fault-tolerant control problem research is also a lot, the method that adopts be also intelligent or with the self-adapted tolerance control method of state space (1, Zhang Junfeng, Hu Shousong be based on the fighter plane fault diagnosis of multi-kernel learning support vector machine, Southeast China University's journal (natural science edition), 2007, the 37th volume supplementary issue (I), pp1-5; Moral dragon Dong Chaoyang after the first Wang Qing of yellow happiness, based on the fault-tolerant control of the hypersonic aircraft of model reference adaptive, Nanjing Aero-Space University's journal,, the 43rd volume supplementary issue in 2011).

When aircraft breaks down, complete machine aerodynamic force, moment, parameter and input quantity all may change, penetrate an aperture as the aircraft wing and will cause all aerodynamic force of full machine, moment etc. all can suddenly change, and different maneuvers can increase an equivalence input; Yet, model before and after present research method breaks down aircraft is described with identical structure and parameter, only the parameters such as input coefficient are adjusted, the fault-tolerant control program that do not consider air dynamic behaviour in design, causes providing loses contact with reality aircraft and is difficult to use.

Summary of the invention

For the fault-tolerant control program that solves existing Fault Tolerance Control Technology and do not consider air dynamic behaviour in only the aircraft input coefficient being adjusted, designed, cause providing is difficult to use in actual technical matters, the present invention has provided a kind of Flight Altitude Moving fault diagnosis and fault tolerant control method, and model and the combinations of states index discriminant of the method during according to the aircraft non-fault carried out fault diagnosis; At first to failure modes, then analyze the impact of the full machine parameter of aircraft and flight dynamics characteristic for the fault category of determining when aircraft breaks down, sudden change system modelling after then again fault being occured obtains the aircraft stylish model description that breaks down; The method for designing of fault-tolerant controller is that the controlled quentity controlled variable designed makes the output of non-fault model reach army's mark requirement, and the control inputs of the common aircraft that forms of non-artificial additional controlled quentity controlled variable that causes of this controlled quentity controlled variable and aircraft fault when breaking down New model stable output when making aircraft break down.

The technical solution adopted for the present invention to solve the technical problems is: a kind of Flight Altitude Moving fault diagnosis and fault tolerant control method are characterized in comprising the following steps:

1, the nonlinear model of lengthwise movement is during the aircraft non-fault:

$\left\{\begin{array}{c}\stackrel{\·}{h}=V_T\sin (\mathrm{\θ}-\mathrm{\α})\\ {\stackrel{\·}{V}}_T=\frac{\mathrm{QS}}{m}[C_x(\mathrm{\α},\mathrm{\δ})\cos \mathrm{\α}+C_z(\mathrm{\α},\mathrm{\δ})\sin \mathrm{\α}]-g\sin (\mathrm{\θ}-\mathrm{\α})\\ \stackrel{\·}{\mathrm{\α}}=q+\frac{\mathrm{QS}}{m{V}_T}[C_z(\mathrm{\α},\mathrm{\δ})\cos \mathrm{\α}-C_x(\mathrm{\α},\mathrm{\δ})\sin \mathrm{\α}]+g\cos (\mathrm{\θ}-\mathrm{\α})/V_T\\ \stackrel{\·}{\mathrm{\θ}}=q\\ \stackrel{\·}{q}=\frac{{\mathrm{QSc}}_l}{I_y}{C}_m(\mathrm{\α},\mathrm{\δ})\end{array}\right.---\left(1\right)$

In formula: state variable h, V _T, α, θ, q represent respectively flying height, flying speed, the air-flow angle of attack, the angle of pitch, rate of pitch; G is acceleration of gravity; Q, S, m, c _l, I _yRepresent respectively dynamic pressure, wing area, vehicle mass, mean aerodynamic chord, be the moment of inertia of Y-axis around axis; C _x(α, δ), C _z(α, δ), C _m(α, δ) represents respectively longitudinal force, normal force and pitching moment; δ is controlled quentity controlled variable;

2, the fault method for detecting is: when combinations of states index J 〉=1, judge that aircraft has fault;

Wherein:

$J=\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|h_{\mathrm{pre}}\left(i\right)-h_{\mathrm{out}}\left(i\right)|-{\mathrm{\Δh}}_t]+\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|V_{\mathrm{Tpre}}\left(i\right)-V_{\mathrm{Tout}}\left(i\right)|-{\mathrm{\ΔV}}_{\mathrm{Tt}}]+$

$\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|{\mathrm{\α}}_{\mathrm{pre}}\left(i\right)-{\mathrm{\α}}_{\mathrm{out}}\left(i\right)|-\mathrm{\Δ}{\mathrm{\α}}_t]+\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|{\mathrm{\θ}}_{\mathrm{pre}}\left(i\right)-{\mathrm{\θ}}_{\mathrm{out}}\left(i\right)|-\mathrm{\Δ}{\mathrm{\θ}}_t]+---\left(2\right)$

$\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|q_{\mathrm{pre}}\left(i\right)-q_{\mathrm{out}}\left(i\right)|-{\mathrm{\Δq}}_t]$

h _pre(i), V _Tpre(i), α _pre(i), θ _pre(i), q _pre(i) be respectively according to (1) formula time t=iT (i=k, k+1 ..., the h that N) calculates, V _T, α, θ, q value; h _out(i), V _Tout(i), α _out(i), θ _out(i), q _out(i) be time t=iT (i=k, k+1 ..., the N) h that obtains of actual measurement, V _T, α, θ, q value, Δ h _t, Δ V _Tt, Δ α _t, Δ θ _t, Δ q _tBe the detection threshold of setting; $\mathrm{sign}\left(x\right)=\left\{\begin{array}{cc}1& x>0\\ 0& x\≤0\end{array}\right.$ Hook sign function; K represents k sampled point, and the corresponding sampling time is t=kT; N+1 is specification error accumulative total number, and T is the sampling period;

When 3, aircraft breaks down at first to failure modes, analyze again the impact of the full machine parameter of aircraft and flight dynamics characteristic for the fault category of determining, then sudden change system modelling after again fault being occured, when obtaining aircraft and breaking down, the new nonlinear model of lengthwise movement is described as:

$\left\{\begin{array}{c}\stackrel{\·}{h}=V_T\sin (\mathrm{\θ}-\mathrm{\α})\\ {\stackrel{\·}{V}}_T=\frac{Q{S}_f}{m_f}[C_{\mathrm{xf}}(\mathrm{\α},{\mathrm{\δ}}_f)\cos \mathrm{\α}+C_{\mathrm{zf}}(\mathrm{\α},{\mathrm{\δ}}_f)\sin \mathrm{\α}]-g\sin (\mathrm{\θ}-\mathrm{\α})\\ \stackrel{\·}{\mathrm{\α}}=q+\frac{Q{S}_f}{m_f{V}_T}[C_{\mathrm{zf}}(\mathrm{\α},{\mathrm{\δ}}_f)\cos \mathrm{\α}-C_{\mathrm{xf}}(\mathrm{\α},{\mathrm{\δ}}_f)\sin \mathrm{\α}]+g\cos (\mathrm{\θ}-\mathrm{\α})/V_T\\ \stackrel{\·}{\mathrm{\θ}}=q\\ \stackrel{\·}{q}=\frac{{Q{S}_{f}c}_{\mathrm{lf}}}{I_{\mathrm{yf}}}{C}_{\mathrm{mf}}(\mathrm{\α},{\mathrm{\δ}}_f)\end{array}\right.---\left(3\right)$

In formula: S _f, m _f, c _lf, I _yfRepresent respectively when aircraft breaks down wing area, vehicle mass, mean aerodynamic chord, be the moment of inertia of Y-axis around axis; C _xf(α, δ _f), C _zf(α, δ _f), C _mf, (α, δ _f) represent respectively longitudinal force, normal force and pitching moment when aircraft breaks down; ${\mathrm{\δ}}_f=\left[\begin{array}{c}\mathrm{\δ}\\ {\mathrm{\δ}}_{\mathrm{apf}}\end{array}\right]$ Control inputs when aircraft breaks down, the non-artificial additional controlled quentity controlled variable δ that the aircraft fault causes _apf

4, the method for designing of fault-tolerant controller is: the controlled quentity controlled variable δ that designs makes nonlinear model (1) formula stability margin reach the requirement of army mark, and the non-artificial additional controlled quentity controlled variable δ that causes of this controlled quentity controlled variable δ and aircraft fault _apfControl inputs when the common aircraft that forms breaks down ${\mathrm{\δ}}_f=\left[\begin{array}{c}\mathrm{\δ}\\ {\mathrm{\δ}}_{\mathrm{apf}}\end{array}\right]$ Make nonlinear model (3) formula stable output.

the invention has the beneficial effects as follows: the model during by the aircraft non-fault and combinations of states index discriminant are carried out fault diagnosis, can effectively detect fault and failure message is classified, when breaking down, aircraft considers full machine parameter and the modeling again of flight dynamics characteristic, the design fault-tolerant controller makes the output of non-fault model reach army's mark requirement on two model bases, and New model stable output when the control inputs when the common aircraft that forms of the non-artificial additional controlled quentity controlled variable that this controlled quentity controlled variable and aircraft fault cause breaks down makes aircraft break down, having solved existing Fault Tolerance Control Technology only adjusts the aircraft input coefficient, do not consider air dynamic behaviour in design, the fault-tolerant control program that causes providing is difficult to use in actual technical matters.

Below in conjunction with embodiment, the present invention is elaborated.

Embodiment

1, the nonlinear model of lengthwise movement is during the aircraft non-fault:

$\left\{\begin{array}{c}\stackrel{\·}{h}=V_T\sin (\mathrm{\θ}-\mathrm{\α})\\ {\stackrel{\·}{V}}_T=\frac{\mathrm{QS}}{m}[C_x(\mathrm{\α},\mathrm{\δ})\cos \mathrm{\α}+C_z(\mathrm{\α},\mathrm{\δ})\sin \mathrm{\α}]-g\sin (\mathrm{\θ}-\mathrm{\α})\\ \stackrel{\·}{\mathrm{\α}}=q+\frac{\mathrm{QS}}{m{V}_T}[C_z(\mathrm{\α},\mathrm{\δ})\cos \mathrm{\α}-C_x(\mathrm{\α},\mathrm{\δ})\sin \mathrm{\α}]+g\cos (\mathrm{\θ}-\mathrm{\α})/V_T\\ \stackrel{\·}{\mathrm{\θ}}=q\\ \stackrel{\·}{q}=\frac{{\mathrm{QSc}}_l}{I_y}{C}_m(\mathrm{\α},\mathrm{\δ})\end{array}\right.---\left(4\right)$

In formula: state variable h, V _T, α, δ, q represent respectively flying height, flying speed, the air-flow angle of attack, the angle of pitch, rate of pitch; G is acceleration of gravity; Q, S, m, c _l, I _yRepresent respectively dynamic pressure, wing area, vehicle mass, mean aerodynamic chord, be the moment of inertia of Y-axis around axis; C _x(α, δ), C _z(α, δ), C _m(α, δ) represents respectively longitudinal force, normal force and pitching moment; $\mathrm{\δ}=\left[\begin{array}{c}{\mathrm{\δ}}_T\\ {\mathrm{\δ}}_{e1}\\ {\mathrm{\δ}}_{e2}\end{array}\right]$ Be controlled quentity controlled variable, δ _TBe engine throttle opening, δ _e1Be left elevator, δ _e2Be left elevator;

2, the fault method for detecting is: when combinations of states index J 〉=1, judge that aircraft has fault;

Wherein:

$J=\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|h_{\mathrm{pre}}\left(i\right)-h_{\mathrm{out}}\left(i\right)|-{\mathrm{\Δh}}_t]+\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|V_{\mathrm{Tpre}}\left(i\right)-V_{\mathrm{Tout}}\left(i\right)|-{\mathrm{\ΔV}}_{\mathrm{Tt}}]+$

$\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|{\mathrm{\α}}_{\mathrm{pre}}\left(i\right)-{\mathrm{\α}}_{\mathrm{out}}\left(i\right)|-\mathrm{\Δ}{\mathrm{\α}}_t]+\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|{\mathrm{\θ}}_{\mathrm{pre}}\left(i\right)-{\mathrm{\θ}}_{\mathrm{out}}\left(i\right)|-\mathrm{\Δ}{\mathrm{\θ}}_t]+---\left(5\right)$

$\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|q_{\mathrm{pre}}\left(i\right)-q_{\mathrm{out}}\left(i\right)|-{\mathrm{\Δq}}_t]$

h _pre(i), V _Tpre(i), α _pre(i), θ _pre(i), q _pre(i) be respectively according to (4) formula time t=iT (i=k, k+1 ..., the h that N) calculates, V _T, α, θ, q value; h _out(i), V _Tout(i), α _out(i), θ _out(i), q _out(i) be time t=iT (i=k, k+1 ..., the N) h that obtains of actual measurement, V _T, α, θ, q value, Δ h _t, Δ V _Tt, Δ α _t, Δ θ _t, Δ q _tBe the detection threshold of setting; $\mathrm{sign}\left(x\right)=\left\{\begin{array}{cc}1& x>0\\ 0& x\≤0\end{array}\right.$ Be sign function; K represents k sampled point, and the corresponding sampling time is t=kT; N+1 is specification error accumulative total number, and T is the sampling period;

At first to failure modes, then analyze the impact of the full machine parameter of aircraft and flight dynamics characteristic for the fault category of determining when 3, aircraft breaks down; As monolateral in elevating rudder stuck 3 °, wing Local Symmetric when damage, to full machine modeling again, when obtaining aircraft and breaking down, the new nonlinear model of lengthwise movement is described as:

$\left\{\begin{array}{c}\stackrel{\·}{h}=V_T\sin (\mathrm{\θ}-\mathrm{\α})\\ {\stackrel{\·}{V}}_T=\frac{Q{S}_f}{m_f}[C_{\mathrm{xf}}(\mathrm{\α},{\mathrm{\δ}}_f)\cos \mathrm{\α}+C_{\mathrm{zf}}(\mathrm{\α},{\mathrm{\δ}}_f)\sin \mathrm{\α}]-g\sin (\mathrm{\θ}-\mathrm{\α})\\ \stackrel{\·}{\mathrm{\α}}=q+\frac{Q{S}_f}{m_f{V}_T}[C_{\mathrm{zf}}(\mathrm{\α},{\mathrm{\δ}}_f)\cos \mathrm{\α}-C_{\mathrm{xf}}(\mathrm{\α},{\mathrm{\δ}}_f)\sin \mathrm{\α}]+g\cos (\mathrm{\θ}-\mathrm{\α})/V_T\\ \stackrel{\·}{\mathrm{\θ}}=q\\ \stackrel{\·}{q}=\frac{{Q{S}_{f}c}_{\mathrm{lf}}}{I_{\mathrm{yf}}}{C}_{\mathrm{mf}}(\mathrm{\α},{\mathrm{\δ}}_f)\end{array}\right.---\left(6\right)$

In formula: S _f, m _f, c _lf, I _yfRepresent respectively when aircraft breaks down wing area, vehicle mass, mean aerodynamic chord, be the moment of inertia of Y-axis around axis; C _xf(α, δ _f), C _zf(α, δ _f), C _mf(α, δ _f) represent respectively longitudinal force, normal force and pitching moment when aircraft breaks down;

Control inputs when aircraft breaks down, the non-artificial additional controlled quentity controlled variable δ that the aircraft fault causes _apf

4, the method for designing of fault-tolerant controller is: the controlled quentity controlled variable δ that designs makes nonlinear model (4) formula stability margin reach military standard of China, and the non-artificial additional controlled quentity controlled variable δ that causes of this controlled quentity controlled variable δ and aircraft fault _apfControl inputs when the common aircraft that forms breaks down ${\mathrm{\δ}}_f=\left[\begin{array}{c}\mathrm{\δ}\\ {\mathrm{\δ}}_{\mathrm{apf}}\end{array}\right]$ Make nonlinear model (6) formula stable output.

Claims (1)

1. a Flight Altitude Moving fault diagnosis and fault tolerant control method are characterized in comprising the following steps: (a) nonlinear model of lengthwise movement is during the aircraft non-fault:

$\left\{\begin{array}{c}\stackrel{\·}{h}=V_T\sin (\mathrm{\θ}-\mathrm{\α})\\ {\stackrel{\·}{V}}_T=\frac{\mathrm{QS}}{m}[C_x(\mathrm{\α},\mathrm{\δ})\cos \mathrm{\α}+C_z(\mathrm{\α},\mathrm{\δ})\sin \mathrm{\α}]-g\sin (\mathrm{\θ}-\mathrm{\α})\\ \stackrel{\·}{\mathrm{\α}}=q+\frac{\mathrm{QS}}{m{V}_T}[C_z(\mathrm{\α},\mathrm{\δ})\cos \mathrm{\α}-C_x(\mathrm{\α},\mathrm{\δ})\sin \mathrm{\α}]+g\cos (\mathrm{\θ}-\mathrm{\α})/V_T\\ \stackrel{\·}{\mathrm{\θ}}=q\\ \stackrel{\·}{q}=\frac{{\mathrm{QSc}}_l}{I_y}{C}_m(\mathrm{\α},\mathrm{\δ})\end{array}\right.---\left(1\right)$ In formula: state variable h, V _T, α, θ, q represent respectively flying height, flying speed, the air-flow angle of attack, the angle of pitch, rate of pitch; G is acceleration of gravity; Q, S, m, c _l, I _yRepresent respectively dynamic pressure, wing area, vehicle mass, mean aerodynamic chord, be the moment of inertia of Y-axis around axis; C _x(α, δ), C _z(α, δ), C _m(α, δ) represents respectively longitudinal force, normal force and pitching moment; δ is controlled quentity controlled variable;

(b) the fault method for detecting is: when combinations of states index J 〉=1, judge that aircraft has fault; Wherein:

$J=\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|h_{\mathrm{pre}}\left(i\right)-h_{\mathrm{out}}\left(i\right)|-{\mathrm{\Δh}}_t]+\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|V_{\mathrm{Tpre}}\left(i\right)-V_{\mathrm{Tout}}\left(i\right)|-{\mathrm{\ΔV}}_{\mathrm{Tt}}]+$

$\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|{\mathrm{\α}}_{\mathrm{pre}}\left(i\right)-{\mathrm{\α}}_{\mathrm{out}}\left(i\right)|-\mathrm{\Δ}{\mathrm{\α}}_t]+\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|{\mathrm{\θ}}_{\mathrm{pre}}\left(i\right)-{\mathrm{\θ}}_{\mathrm{out}}\left(i\right)|-\mathrm{\Δ}{\mathrm{\θ}}_t]+---\left(2\right)$

$\mathrm{sign}\left[\underset{i=k}{\overset{k+N}{\mathrm{\Σ}}}\right|q_{\mathrm{pre}}\left(i\right)-q_{\mathrm{out}}\left(i\right)|-{\mathrm{\Δq}}_t]$ h _pre(i), V _Tpre(i), α _pre(j), θ _pre(j), q _pre(i) be respectively according to (1) formula time t=iT (i=k, k+1 ..., the h that N) calculates, V _T, α, θ, q value; h _out(j), V _Tout(i), α _out, (i), θ _out(i), q _out(i) be time t=iT (i=k, k+1 ..., the N) h that obtains of actual measurement, V _T, α, θ, q value, Δ h _t, Δ V _Tt, Δ α _t, Δ θ _t, Δ q _tBe the detection threshold of setting; $\mathrm{sign}\left(x\right)=\left\{\begin{array}{cc}1& x>0\\ 0& x\≤0\end{array}\right.$ Be sign function; K represents k sampled point, and the corresponding sampling time is t=kT; N+1 is specification error accumulative total number, and T is the sampling period;

When (c) aircraft breaks down at first to failure modes, analyze again the impact of the full machine parameter of aircraft and flight dynamics characteristic for the fault category of determining, then sudden change system modelling after again fault being occured, when obtaining aircraft and breaking down, the new nonlinear model of lengthwise movement is described as:

$\left\{\begin{array}{c}\stackrel{\·}{h}=V_T\sin (\mathrm{\θ}-\mathrm{\α})\\ {\stackrel{\·}{V}}_T=\frac{Q{S}_f}{m_f}[C_{\mathrm{xf}}(\mathrm{\α},{\mathrm{\δ}}_f)\cos \mathrm{\α}+C_{\mathrm{zf}}(\mathrm{\α},{\mathrm{\δ}}_f)\sin \mathrm{\α}]-g\sin (\mathrm{\θ}-\mathrm{\α})\\ \stackrel{\·}{\mathrm{\α}}=q+\frac{Q{S}_f}{m_f{V}_T}[C_{\mathrm{zf}}(\mathrm{\α},{\mathrm{\δ}}_f)\cos \mathrm{\α}-C_{\mathrm{xf}}(\mathrm{\α},{\mathrm{\δ}}_f)\sin \mathrm{\α}]+g\cos (\mathrm{\θ}-\mathrm{\α})/V_T\\ \stackrel{\·}{\mathrm{\θ}}=q\\ \stackrel{\·}{q}=\frac{{Q{S}_{f}c}_{\mathrm{lf}}}{I_{\mathrm{yf}}}{C}_{\mathrm{mf}}(\mathrm{\α},{\mathrm{\δ}}_f)\end{array}\right.---\left(3\right)$ In formula: S _f, m _f, c _lf, I _yfRepresent respectively when aircraft breaks down wing area, vehicle mass, mean aerodynamic chord, be the moment of inertia of Y-axis around axis; C _xf(α, δ f), C _zf(α, δ _f), C _mf(α, δ _f) represent respectively longitudinal force, normal force and pitching moment when aircraft breaks down; ${\mathrm{\δ}}_f=\left[\begin{array}{c}\mathrm{\δ}\\ {\mathrm{\δ}}_{\mathrm{apf}}\end{array}\right]$ Control inputs when aircraft breaks down, the non-artificial additional controlled quentity controlled variable δ that the aircraft fault causes _apf

(d) method for designing of fault-tolerant controller is: the controlled quentity controlled variable δ that designs makes nonlinear model (1) formula stability margin reach the requirement of army mark, and the non-artificial additional controlled quentity controlled variable δ that causes of this controlled quentity controlled variable δ and aircraft fault _apfControl inputs when the common aircraft that forms breaks down ${\mathrm{\δ}}_f=\left[\begin{array}{c}\mathrm{\δ}\\ {\mathrm{\δ}}_{\mathrm{apf}}\end{array}\right]$ The official gets nonlinear model (3) formula stable output.