II-type fuzzy logic fault detection method for electromechanical flight actuator

Abstract

The invention discloses an II-type fuzzy logic fault detection method for an electromechanical flight actuator, and the method comprises the steps: building an EMFA dynamic model: analyzing the working principle of the EMFA dynamic model, comprehensively considering a nonlinear term and external disturbance, building the EMFA dynamic model, and completing the model conversion; an II-type fuzzy logic system is established, and an II-type fuzzy disturbance approximator is established based on the II-type fuzzy logic system; based on an II-type fuzzy disturbance approximator, fault detection is carried out, whether faults exist in the EMFA or not is judged, and stability analysis is carried out on the EMFA. According to the fault detection method, the influence of strong nonlinearity and unknown disturbance on the EMFA is comprehensively considered, the fault detection method based on the II-type fuzzy logic system not only can provide effective estimation for nonlinear factors and unknown disturbance, but also can judge whether the EMFA has faults or not, and then the effectiveness and accuracy of the fault detection method are improved.

Description

Technical Field

[0001] The invention relates to a type II fuzzy logic fault detection method for an electromechanical flight actuator, belonging to the technical field of electromechanical flight actuators. Background Art

[0002] At present, most control components of large aircraft are driven by hydraulic servo actuators, which are powered by a centrally organized hydraulic power system. Today, aircraft electrification has become one of the strategic tasks to support the innovation of the aerospace industry, which can effectively reduce weight and significantly improve maintainability. In fact, in the development and design of future generations of aircraft, electric actuators will become a key factor in the construction of electric aircraft. Compared with traditional hydraulic actuators, the application of electric actuators in the aerospace field has received widespread attention. Among them, electromechanical flight actuators (EMFA), as a typical type of electric actuator, have been used in a variety of aerospace aircraft such as drones, civil airliners, and F-35 joint strike fighters. Although EMFA has been used to drive components such as spoilers, aviation fuel doors, and weapon bay doors, for most EMFAs, due to their insufficient continuous use time, they cannot meet reliable failure statistics and cannot achieve the safety requirements of operation in aerospace applications. It can be seen that the high reliability requirements of fault tolerance and small space envelopes are still a challenging task.

[0003] Although the applicability of EMFA in terms of load and speed performance in the aerospace field has been well demonstrated, there are still problems with safety and reliability. EMFA is mostly in high-risk environments. These problems make it particularly important to effectively avoid catastrophic consequences through the design of fault detection methods. Therefore, fault diagnosis and detection are crucial to the safety of aircraft. Therefore, developing effective fault detection methods and promoting their application in aircraft are important measures to ensure the high reliability of EMFA, which also helps aircraft recover from serious faults. In addition, since nonlinearity and load disturbances can have a significant impact on the actuation effect of the actuator, it is necessary to deal with the impact of this negative factor on the system.

[0004] Typically, data-driven methods are used to obtain effective fault information. Specifically, the response of EMFA is measured by manually injecting the main fault into the system. In addition, intelligent control technologies have also been explored in the fault detection design of EMFA, such as deep learning and fuzzy control. In fact, EMFA may encounter several key challenges when constructing fault detection methods, including the difficulty in obtaining accurate dynamic information, the influence of noise signals, and unknown disturbances to fault information. In the fault detection process, the system has nonlinear uncertainties and adverse effects of external interference, and the system fault detection accuracy is low and the false alarm rate is high. Summary of the invention

[0005] Purpose of the invention: In view of the deficiencies in the prior art, the present invention provides a Type II fuzzy logic fault detection method for an electromechanical flight actuator. The present invention establishes a Type II fuzzy disturbance approximator based on a Type II fuzzy logic system to perform fault detection, thereby improving the effectiveness and accuracy of the fault detection method.

[0006] Technical solution: A type II fuzzy logic fault detection method for an electromechanical flight actuator comprises the following steps:

[0007] S1: Establish EMFA dynamic model: Analyze its working principle and comprehensively consider nonlinear terms and external disturbances, construct EMFA dynamic model and complete model conversion;

[0008] S2: By establishing a type II fuzzy logic system, and based on it, a type II fuzzy disturbance approximator is established;

[0009] S3: Based on the type II fuzzy disturbance approximator established in S2, fault detection is performed to determine whether there is a fault in the EMFA and to perform stability analysis on it.

[0010] Preferably, the S1 is specifically:

[0011] The EMFA voltage equation is:

[0012]

[0013] Where L is the self-inductance of the coupling coil, M is the mutual inductance of the coupling coil, and u A ,u B ,u C is the phase voltage, i A ,i B ,i C is the phase current, R is the phase resistance, e A ,e B ,e C is the back electromotive force, ω m is the rotor angular velocity, K e is the motor parameter, f A(μ),f B (μ),f C (μ) is the system waveform function, then:

[0014] e A =K e ω m f A (μ)

[0015] e B =K e ω m f B (μ)

[0016] e C =K e ω m f C (μ)

[0017] The EMFA kinetic equation is:

[0018]

[0019] Among them, T L is the load torque, T f is the friction torque, J is the moment of inertia, B d is the viscous damping parameter, ω m is the rotor angular velocity, T m is the output torque and is related to back EMF, rotor speed, and phase current:

[0020] According to the voltage equation (1) and the dynamic equation (2), the dynamic model of EMFA is converted into a standard control model for fault detection, which is expressed as follows:

[0021]

[0022] The state vector of the actuator is x = [i A ,i B ,i C ,ω m ,x L ] T ,x L is the output displacement of the roller screw, and the control input is u=[u A ,u B ,u C ,T L ] T , the output y is the angular velocity ω output by the actuator motor m , P L is the roller screw lead, ΔT fis the disturbance caused by friction nonlinearity and unknown aerodynamic effects. The overall disturbance of EMFA is Δδ = [0, 0, 0, 0, ΔT f ] T ; From formula (3), we can get:

[0023]

[0024] Where A is the system matrix about the state, B is the system matrix about the input, C is the system matrix of the output channel, u is the system input vector, and d is the external disturbance of the system.

[0025]

[0026] C=[0,0,0,1,0], d=ZΔδ, Z is the matrix of disturbance. Optimization, since EMFA will be affected by nonlinearity and unknown disturbances, for the nonlinear factors of the actuator itself, considering the influence of friction characteristics, the friction adopts the following LuGre friction model:

[0027]

[0028] Among them, z is the LuGre internal variable, ω s is the Stribeck angular velocity, σ0 is the average stiffness coefficient of the bristles between the contact surfaces, σ1 is the damping coefficient of the bristle deformation, σ2 is the viscous damping coefficient, T c is the Coulomb friction torque, T s is the static friction torque, from which we can get:

[0029] T f =T f * +ΔT f (6)

[0030] in,

[0031] Preferably, the controller based on type II fuzzy logic in S2 is specifically:

[0032] Using the mean of the outputs at both ends of the fuzzy logic system, the output of the type II fuzzy logic system is:

[0033]

[0034] y l =Ξ l T ξ l , y r =Ξ r T ξ r (8)

[0035] In the formula, y l and r are the fuzzy logic systems on the left and right, respectively. l ,Ξ r ,ξ l ,ξ r Denote the left and right adjustable parameter vectors and their associated regression matrices respectively, let Then we have:

[0036] y t2f =Ξ T ξ (9)

[0037] According to the universal approximation theorem, we can get:

[0038] Set the objective function G(s):R n →R in the compact set Ω∈R n , where Ω is a compact set, R n is a set of n-dimensional real numbers, and for any ε>0, ε is the approximation error, there exists a type II fuzzy logic system such that Then the function G(x) is expressed as:

[0039]

[0040] Among them, G * (x) = Ξ *T ξ(x) is represented by the optimal interval type II fuzzy logic system, * is the optimal parameter vector of Ξ and is obtained by calculation:

[0041]

[0042] Preferably, the type II fuzzy disturbance approximator established in S2 is specifically:

[0043] Given that the dynamic model of EMFA is expressed by formula (3) and (4), the type II fuzzy logic approximator is specifically:

[0044]

[0045] in, And the i-th adaptive law of the weight vector can be constructed as:

[0046]

[0047] in, Ψ is the internal state vector of the approximator, α = diag[α i ] 5×5 (α i >0), represents the estimated value of d(x,u), Θ=x-Ψ is the internal observation error vector, λ l1i ,λ r1i ,λ r2i ,λ r2i is a pre-designed positive parameter; thus, the disturbance approximation error can achieve consistent stability, that is, the approximation error Θ i Able to converge to a small neighborhood around zero

[0048] Preferably, in S3, a type II fuzzy logic system interference type II fuzzy disturbance approximator is used to obtain estimated values ​​of system nonlinear terms and external interference, and the obtained estimated values ​​are compared with actual values ​​for use in the fault detection process, and are judged against error thresholds under sensor faults and power driver faults to determine whether a fault has occurred.

[0049] Preferably, the fault detection in S3 is specifically as follows:

[0050] The approximation error is defined as E = xx d (E=[e1,...,e5] T ), where x d Define the derivative of the expected instruction as e1, e2, ..., e5 are the errors of the corresponding channels of EMFA respectively. If the approximation error is not greater than the preset value of the approximation error, the EMFA system is considered to be in a fault-free mode. If the approximation error is greater than the preset value of the approximation error, the system operates in a faulty state:

[0051]

[0052] Where Y represents the prefix value threshold associated with different system failure modes;

[0053] Preferably, the stability analysis in S3 is specifically as follows:

[0054] According to formula (10) and formula (11), we get: also, Among them * is the optimal parameter vector, ε=[ε1,...,ε5] T is the approximate error; similarly, we get:

[0055]

[0056] According to formula (10) and formula (11), assuming that the estimation error ε i and the optimal parameter vector are all bounded, so we believe that there is a constant Make From formula (15), we can know that the calculation formula of the disturbance approximation error is:

[0057]

[0058] in, The candidate Lyapunov function is written as:

[0059] W=W1+…+W5 (17)

[0060] The i-th (i=1,2,…,5) function is given by:

[0061]

[0062] Taking the time derivative of formula (18) along (16) yields:

[0063]

[0064] Among them, W i is the i-th Lyapunov function, μ i is the positive constant term of the i-th adaptive law, Θ i is the i-th internal observation error vector, is the i-th adaptive law error vector, ξ i is the i-th type II fuzzy basis function vector; ε i is the i-th estimation error, λ 1i The i-th adaptive law is a positive constant term.

[0065] Substituting formula (13) into (19) we obtain:

[0066]

[0067] Combining equations (19) and (20), we get:

[0068]

[0069] in, From formula (21), we get:

[0070]

[0071] in,

[0072] So far, we can get that the designed type II fuzzy logic approximator can obtain consistent stability, and solve the inequality to get The approximator parameters are chosen according to this inequality so that Θ i Approaching zero.

[0073] Beneficial effects: The present invention comprehensively considers the influence of strong nonlinearity and unknown disturbances on EMFA. The fault detection method based on type II fuzzy logic system can not only provide an effective estimation of nonlinear factors and unknown disturbances, but also determine whether EMFA has a fault, thereby improving the effectiveness and accuracy of the fault detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0075] Figure 1 is a flow chart of the method of the present invention;

[0076] Figure 2 This is the schematic diagram of EMFA;

[0077] Figure 3 It is the EMFA deflection angle output under normal conditions;

[0078] Figure 4 is the torque output of EMFA under sensor failure;

[0079] Figure 5 It is the EMFA deflection angle output under sensor failure;

[0080] Figure 6 is the torque output of EMFA under power driver fault;

[0081] Figure 7 It is the EMFA deflection angle output when the power driver fails. DETAILED DESCRIPTION

[0082] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] In the description of the present invention, it is necessary to understand that the terms "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0084] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0085] like Figure 1 As shown, a type II fuzzy logic fault detection method for an electromechanical flight actuator comprises the following steps:

[0086] S1: Establish the dynamic model of electromechanical flight actuator: analyze its working principle and comprehensively consider nonlinear terms and external disturbances, build the EMFA dynamic model and complete the model conversion;

[0087] like Figure 2 As shown, the EMFA consists of four parts: a brushless DC motor, a mechanical drive device, an electronic controller and a sensor. The actuator completes the conversion from electrical drive to mechanical drive based on the control instructions, and finally completes the output of the rudder surface movement.

[0088] The EMFA voltage equation is:

[0089]

[0090] Where L is the self-inductance of the coupling coil, M is the mutual inductance of the coupling coil, and u A ,u B ,u C is the phase voltage, i A ,i B ,i C is the phase current, R is the phase resistance, e A ,e B ,e C is the back electromotive force, ω m is the rotor angular velocity, K e is the motor parameter, fA (μ),f B (μ),f C (μ) is the system waveform function, then:

[0091] e A =K e ω m f A (μ)

[0092] e B =K e ω m f B (μ)

[0093] e C =K e ω m f C (μ)

[0094] The EMFA kinetic equation is:

[0095]

[0096] Among them, T L is the load torque, T f is the friction torque, J is the moment of inertia, B d is the viscous damping parameter, ω m is the rotor angular velocity, T m is the output torque and is related to back EMF, rotor speed, and phase current:

[0097] According to the voltage equation (1) and the dynamic equation (2), the dynamic model of EMFA is converted into a standard control model for fault detection, which is expressed as follows:

[0098]

[0099] The state vector of the actuator is x = [i A ,i B ,i C ,ω m ,x L ] T ,x L is the output displacement of the roller screw, and the control input is u=[u A ,u B ,u C ,T L ] T , the output y is the angular velocity ω output by the actuator motor m , P L is the roller screw lead, ΔT fis the disturbance caused by friction nonlinearity and unknown aerodynamic effects. The overall disturbance of EMFA is Δδ = [0, 0, 0, 0, ΔT f ] T ; From formula (3), we can get:

[0100]

[0101] Where A is the system matrix about the state, B is the system matrix about the input, C is the system matrix of the output channel, u is the system input vector, and d is the external disturbance of the system.

[0102]

[0103] Z is the matrix of perturbations.

[0104] Since EMFA is affected by nonlinear and unknown disturbances, such as friction, aerodynamic load changes, aircraft attitude and airflow, the nonlinear factors of the actuator itself are considered, and the influence of friction characteristics is considered. The friction adopts the following LuGre friction model:

[0105]

[0106] Among them, z is the LuGre internal variable, ω s is the Stribeck angular velocity, σ0 is the average stiffness coefficient of the bristles between the contact surfaces, σ1 is the damping coefficient of the bristle deformation, σ2 is the viscous damping coefficient, T c is the Coulomb friction torque, T s is the static friction torque, from which we can get:

[0107] T f =T f * +ΔT f (6)

[0108] Among them, T f * =σ1+σ2ω m ;

[0109] S2: By establishing a type II fuzzy logic system, and based on it, a type II fuzzy disturbance approximator is established;

[0110] A type II fuzzy logic system is introduced, and a type II fuzzy disturbance approximator is constructed based on it, which can accurately estimate the nonlinear terms of the actuator and external disturbances. It provides a high level of design flexibility and has a great performance improvement in dealing with uncertainty. Using the mean of the outputs at both ends of the fuzzy logic system, the output of the type II fuzzy logic system is obtained as:

[0111]

[0112]

[0113] In the formula, y l and r are the fuzzy logic systems on the left and right, respectively. l ,Ξ r ,ξ l ,ξ r Denote the left and right adjustable parameter vectors and their associated regression matrices respectively, let Then we have:

[0114] y t2f =Ξ T ξ (9)

[0115] According to the universal approximation theorem, we can get:

[0116] Set the objective function G(s):R n →R in the compact set Ω∈R n , where Ω is a compact set, R n is a set of n-dimensional real numbers, and for any ε>0, ε is the approximation error, there exists a type II fuzzy logic system such that Then the function G(x) is expressed as:

[0117]

[0118] Among them, G * (x) = Ξ *T ξ(x) is represented by the optimal interval type II fuzzy logic system, * is the optimal parameter vector of Ξ and is obtained by calculation:

[0119]

[0120] Given that the dynamic model of EMFA is expressed by formula (3) and (4), the type II fuzzy logic approximator is specifically:

[0121]

[0122] in, And the i-th adaptive law of the weight vector can be constructed as:

[0123]

[0124] in, Ψ is the internal state vector of the approximator, α = diag[α i ] 5×5 (α i >0), represents the estimated value of d(x,u), Θ=x-Ψ is the internal observation error vector, λ l1i ,λ r1i ,λ r2i ,λ r2i is a pre-designed positive parameter; thus, the disturbance approximation error can achieve consistent stability, that is, the approximation error Θ i Able to converge to a small neighborhood around zero

[0125] S3: Based on the type II fuzzy disturbance approximator established in S2, the fuzzy disturbance approximator is deployed in the fault detection method in order to provide an effective estimation of the system nonlinearity and disturbances, and perform fault detection to determine whether there is a fault in the EMFA and perform stability analysis on it.

[0126] In S3, a type II fuzzy logic system interference type II fuzzy disturbance approximator is used to obtain estimated values ​​of system nonlinear terms and external interference, and the obtained estimated values ​​are compared with actual values ​​for use in the fault detection process, and are judged against error thresholds under sensor faults and power driver faults to determine whether a fault occurs.

[0127] The approximation error is defined as E = xx d (E=[e1,...,e5] T ), where x d Define the derivative of the expected instruction as e1, e2, ..., e5 are the errors of the corresponding channels of EMFA respectively. If the approximation error is not greater than the preset value of the approximation error, the EMFA system is considered to be in a fault-free mode. If the approximation error is greater than the preset value of the approximation error, the system operates in a faulty state:

[0128]

[0129] Where Y represents the prefix value threshold associated with different system failure modes;

[0130] According to formula (10) and formula (11), we get: also, Among them * is the optimal parameter vector, ε=[ε1,...,ε5] T is the approximate error; similarly, we get:

[0131]

[0132] According to formula (10) and formula (11), assuming that the estimation error ε i and the optimal parameter vector are all bounded, so we believe that there is a constant Make From formula (15), we can know that the calculation formula of the disturbance approximation error is:

[0133]

[0134] in, The candidate Lyapunov function is written as:

[0135] W=W1+...+W5 (17)

[0136] The i-th (i=1,2,...,5) function is given by:

[0137]

[0138] Taking the time derivative of formula (18) along (16) yields:

[0139]

[0140] Among them, W i is the i-th Lyapunov function, μ i is the positive constant term of the i-th adaptive law, Θ i is the i-th internal observation error vector, is the i-th adaptive law error vector, ξ i is the i-th type II fuzzy basis function vector; ε i is the i-th estimation error, λ 1i The i-th adaptive law is a positive constant term.

[0141] Substituting formula (13) into (19), we obtain:

[0142]

[0143] Combining equations (19) and (20), we get:

[0144]

[0145] Among them, N i =min{μ i ,λ 2i}, From formula (21), we get:

[0146]

[0147] Among them, 0<N≤min{N1,N2,...,N5},

[0148] So far, we can get that the designed type II fuzzy logic approximator can obtain consistent stability, and solve the inequality to get The approximator parameters are chosen according to this inequality so that Θ i Approaching zero.

[0149] Implementation and verification of fault detection method for type II fuzzy logic system:

[0150] Some examples of simulation experiments of EMFA systems affected by nonlinear terms and unknown disturbances are provided below, aiming to demonstrate the effectiveness and accurate performance of the proposed fault detection method design based on type II fuzzy logic system. The simulation experiments consider three cases for testing and comparison respectively.

[0151] 1. No-failure mode

[0152] The equivalent aerodynamic load on the EMFA control surface is used as the load input, and a square wave command signal with an amplitude of 30° and a period of 1 second is received. After about 0.18 seconds, the EMFA enters the position servo drive. At 1 second, an external load disturbance of 10 N·m is added, such as Figure 3 As shown in the figure, although the convergence tracking time of the type II fuzzy logic system is slightly delayed by about 0.01 seconds relative to the transmission control system, it can better track the system reference command, especially after a sudden load disturbance, and can track the system command more sensitively and accurately.

[0153] 2. Sensor failure

[0154] Sensors are the most prone to failure in EMFA, mainly due to mutual coupling of electronic components, electromagnetic field environment interference, abnormal sensitive components, and mutual coupling between sensors, electronic control devices and other electronic devices, which limits the performance of fault diagnosis. Therefore, when the system runs normally for 1 second, a sensor noise fault is injected to analyze the performance of the detection system in the fault signal. Figure 4 and Figure 5 As shown in Figure 1, nonlinearity and fault coupling can interfere with the electronic fault detection of the sensor itself. Compared with the traditional tracking detection system, the fault detection method of the type II fuzzy logic system is more sensitive to the tracking changes of the system electrical signal. Therefore, the designed fault detection strategy weakens the coupling effect between interference and fault, thereby reducing the false alarm rate of system faults and effectively preventing the deterioration of fault effects caused by interference.

[0155] 3. Power driver failure

[0156] The power driver obtains the command signal and adjusts the action of the EMFA. The power driver is susceptible to external interference such as overcurrent, high temperature, and impact vibration, which can cause abnormal output signals. This will interfere with the identification of fault types and the location of fault sources, and weaken the online fault detection capability. When the EMFA operates normally for 1 second, the injected power drive component has an abnormal fault. The role of the analysis detection system in fault signal analysis. Figure 6and Figure 7 As shown in the figure, nonlinearity and fault coupling interfere with the electronic fault detection of the abnormal component itself. Compared with the traditional tracking and monitoring system, the fault detection method of type II fuzzy logic system is more sensitive to the tracking changes of the system electrical signal. Therefore, the proposed fault detection method can weaken the coupling effect between interference and fault, thereby reducing the false alarm rate of system faults and effectively preventing the deterioration of fault effects due to interference.

[0157] Therefore, the present invention designs a fault detection method based on type II fuzzy logic system for EMFAs affected by strong nonlinearity and unknown disturbances. Specifically, we use type II fuzzy logic system to develop a fuzzy disturbance approximator, which can provide effective estimation of nonlinear factors and unknown disturbances; in addition, the approximation error of this method is used in the fault detection process to determine whether a fault has occurred in the EMFA. The stability proof shows that the approximation error of this method can converge to near zero, that is, This can improve the effectiveness of the fault detection method.

[0158] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0159] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A type II fuzzy logic fault detection method for an electromechanical flight actuator, characterized in that: The following steps are involved: S1: Establish the dynamic model of electromechanical flight actuator: analyze its working principle and comprehensively consider nonlinear terms and external disturbances, build the EMFA dynamic model and complete the model conversion; S2: By establishing its controller based on type II fuzzy logic system, and based on this, a type II fuzzy disturbance approximator is established; S3: Based on the type II fuzzy disturbance approximator established in S2, fault detection is performed to determine whether there is a fault in the EMFA and to perform stability analysis on it.

2. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 1, characterized in that: The S1 is specifically: The EMFA voltage equation is: Where L is the self-inductance of the coupling coil, M is the mutual inductance of the coupling coil, and u A ,u B ,u C is the phase voltage, i A ,i B ,i C is the phase current, R is the phase resistance, e A ,e B ,e C is the back electromotive force, ω m is the rotor angular velocity, K e is the motor parameter, f A (μ),f B (μ),f C (μ) is the system waveform function, then: e A =K e oh m f A (m) e B =K e oh m f B (m) e C =K e oh m f C (m) The EMFA kinetic equation is: Among them, T L is the load torque, T f is the friction torque, J is the moment of inertia, B d is the viscous damping parameter, ω m is the rotor angular velocity, T m is the output torque and is related to back EMF, rotor speed, and phase current: According to the voltage equation (1) and the dynamic equation (2), the dynamic model of EMFA is converted into a standard control model for fault detection, which is expressed as follows: The state vector of the actuator is x = [i A ,i B ,i C ,ω m ,x L ] T ,x L is the output displacement of the roller screw, and the control input is u=[u A ,u B ,u C ,T L ] T , the output y is the angular velocity ω output by the actuator motor m , P L is the lead of the roller screw, ΔT f is the disturbance caused by friction nonlinearity and unknown aerodynamic effects. The overall disturbance of EMFA is Δδ = [0, 0, 0, 0, ΔT f ] T ; From formula (3), we can get: Where: A is the system matrix about the state, B is the system matrix about the input, C is the system matrix of the output channel, u is the system input vector, and d is the external disturbance of the system; d=ZΔδ, where Z is the matrix related to the disturbance.

3. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 2, characterized in that: Since EMFA is affected by nonlinearity and unknown disturbances, the nonlinear factors of the actuator itself are considered, and the influence of friction characteristics is considered. The friction adopts the following LuGre friction model: Among them, z is the LuGre internal variable, ω s is the Stribeck angular velocity, σ0 is the average stiffness coefficient of the bristles between the contact surfaces, σ1 is the damping coefficient of the bristle deformation, σ2 is the viscous damping coefficient, T c is the Coulomb friction torque, T s is the static friction torque, from which we can get: T f =T f * +ΔT f (6) Among them, T f * =σ1+σ2ω m ; 。 4. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 3, characterized in that: The controller based on type II fuzzy logic system in S2 is specifically: Using the mean of the outputs at both ends of the fuzzy logic system, the output of the type II fuzzy logic system is: y l =Ξ l T x l ,y r =Ξ r T x r (8) In the formula, y l and r are the fuzzy logic systems on the left and right, respectively. l ,Ξ r ,ξ l ,ξ r Denote the left and right adjustable parameter vectors and their associated regression matrices respectively, let Then we have: y t2f =Ξ T ξ (9) According to the universal approximation theorem, we can get: Set the objective function G(s):R n →R in the compact set Ω∈R n , where Ω is a compact set, R n is a set of n-dimensional real numbers, and for any ε>0, ε is the approximation error, there exists a type II fuzzy logic system such that Then the function G(x) is expressed as: Among them, G * (x) = Ξ *T ξ(x) is represented by the optimal interval type II fuzzy logic system, * is the optimal parameter vector of Ξ and is obtained by calculation:

5. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 4, characterized in that: The type II fuzzy perturbation approximator established in S2 is specifically: Given that the dynamic model of EMFA is expressed by formula (3) and (4), the type II fuzzy logic approximator is specifically: in, And the i-th adaptive law of the weight vector can be constructed as: in, Ψ is the internal state vector of the approximator, α = diag[α i ] 5×5 (α i >0), represents the estimated value of d(x,u), Θ=x-Ψ is the internal observation error vector, λ l1i ,λ r1i ,λ r2i ,λ r2i is a pre-designed positive parameter; thus, the disturbance approximation error can achieve consistent stability, that is, the approximation error Θ i Able to converge to a small neighborhood around zero 6. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 5, characterized in that: In S3, a type II fuzzy logic system interference type II fuzzy disturbance approximator is used to obtain estimated values ​​of system nonlinear terms and external interference, and the obtained estimated values ​​are compared with actual values ​​for use in the fault detection process, and are judged against error thresholds under sensor faults and power driver faults to determine whether a fault occurs.

7. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 6, characterized in that: The fault detection in S3 is specifically as follows: The approximation error is defined as E = xx d (E=[e1,...,e5] T ), where x d Define the expected instructions Number is e1, e2,...,e5 are the errors of the corresponding channels of EMFA respectively. If the approximate error is not greater than the approximate error If the approximation error is greater than the preset value of the approximation error, the EMFA system is considered to be in a fault-free mode. If the approximation error is greater than the preset value of the approximation error, the system operates in a faulty state: Where Y represents the prefix value threshold associated with different system failure modes.

8. The type II fuzzy logic fault detection method for electromechanical flight actuators according to claim 7, characterized in that: The stability analysis in S3 is specifically as follows: According to formula (10) and formula (11), we get: also, Among them * is the optimal parameter vector, ε=[ε1,...,ε5] T is the approximate error; similarly, we get: According to formula (10) and formula (11), assuming that the estimation error ε i and the optimal parameter vector are all bounded, so we believe that there is a constant Make From formula (15), we can know that the calculation formula of the disturbance approximation error is: in, The candidate Lyapunov function is written as: W=W1+…+W5 (17) The i-th (i=1,2,…,5) function is given by: Taking the time derivative of formula (18) along (16) yields: Among them, W i is the i-th Lyapunov function, μ i is the positive constant term of the i-th adaptive law, Θ i is the i-th internal observation error vector, is the i-th adaptive law error vector, ξ i is the i-th type II fuzzy basis function vector; ε i is the i-th estimation error, λ 1i The i-th adaptive law is a positive constant term; Substituting formula (13) into (19), we obtain: Combining equations (19) and (20), we get: Among them, N i =min{μ i ,λ 2i }, From formula (21), we get: Among them, 0<N≤min{N1,N2,...,N5}, So far, we can get that the designed type II fuzzy logic approximator can obtain consistent stability, and solve the inequality to get The approximator parameters are chosen according to this inequality so that Θ i Approaching zero.

Images