A continuously adjustable aircraft failure model library construction method

Abstract

The application discloses a continuously adjustable airplane fault model library construction method and belongs to the technical field of system simulation verification. The application is aimed at key accessories of an airplane take-off and landing system, establishes a mechanism mapping of fault modes and model variables according to FMEA and physical failure analysis, forms fault models containing output pressure deviation, coil short circuit or open circuit, valve core sticking, signal drift, internal leakage or external leakage and the like, and defines three-layer standardized interface structures, namely, a fault enable control interface, a fault control interface and a model physical coupling interface. Finally, a standardized fault model library is constructed through structured storage. The application overcomes the limitations of discrete preset of fault severity, model invasive replacement and lack of multi-accessory coupling simulation in the prior art, and provides support for high-confidence and full-coverage virtual verification of the airplane take-off and landing system.

Description

Technical Field

[0001] This invention belongs to the field of aircraft takeoff and landing system simulation and verification technology, specifically involving a method for constructing a continuously adjustable aircraft fault model library. Background Technology

[0002] Modern aircraft takeoff and landing systems comprise a braking subsystem, a steering subsystem, and a landing gear retraction and extension subsystem. This integrated electromechanical-hydraulic system consists of components such as the landing gear control unit, brake servo valve assembly, wheel speed sensors, brake pressure sensors, steering control valves, steering feedback sensors, steering actuators, hydraulic solenoid valves (for door / door lock retraction and extension), and hydraulic solenoid valves (for landing gear / landing gear lock switching). Its reliability directly impacts aircraft safety during takeoff and landing.

[0003] Traditional reliability verification relies heavily on physical testing, i.e., performing normal function and fault verification on test benches or real aircraft. However, this method has insurmountable limitations: 1. High cost and high risk: Reproducing internal failures of core components, especially destructive or gradual degradation failures, may cause irreversible damage to expensive hardware or even lead to safety incidents.

[0004] 2. Low coverage of failure scenarios: Due to time, cost and technical feasibility limitations, physical tests cannot exhaust all single failures, common-cause failures and complex scenarios of multiple failure timing combinations, which is incompatible with the completeness requirements of aircraft safety assessment.

[0005] Therefore, this invention proposes a method for constructing a continuously adjustable aircraft fault model library. Summary of the Invention

[0006] The technical problem to be solved: To overcome the shortcomings of existing technologies, this invention provides a method for constructing a continuously adjustable aircraft fault model library. It analyzes the fault mechanisms of key components, establishes a mechanism mapping between component fault modes and model variables, and, according to a unified interface specification, establishes a series of standardized fault models that start from deep physical failure mechanisms, possess high simulation fidelity, continuously adjustable fault severity, and exhaustive combination scenarios. This supports high-confidence integrated simulation verification and safety analysis. This invention solves the problems in existing aircraft takeoff and landing system fault simulation technologies, such as the inability to dynamically inject fault models due to the requirement for discrete preset fault severity and the need for prior replacement of fault models.

[0007] The technical solution of this invention is: a method for constructing a continuously adjustable aircraft fault model library, comprising the following steps: Step 1: Establish a mathematical model that includes faults. For each type of critical accessory of the aircraft take-off and landing system, based on historical fault data, fault mode and effect analysis and physical failure analysis, select fault modes and map the abstract fault phenomena into quantifiable and adjustable variables or structures in the simulation model to establish the mechanism mapping between accessory fault modes and model variables. Step 2: Establish a standardized fault model encapsulation format and interface specification for integrated simulation. Based on the mathematical model in Step 1, configure three types of interfaces for the fault models of each attachment: a fault enable control interface for activating or bypassing the fault kernel, a fault control interface for dynamically defining the fault severity and evolution mode, and a model physical coupling interface that exchanges signals or data with the main system and is completely consistent with the ports of normal components, thus obtaining standardized encapsulation. The fault control interface supports input of scalar, vector, and even time series functions, thereby realizing continuous adjustment of fault severity and exhaustive coverage of multiple fault combination scenarios. Step 3: Store the standardized fault model and its descriptive information encapsulated in Step 2 in a structured manner to obtain a standardized fault model library that can be continuously adjusted and is plug-and-play integrated.

[0008] A further technical solution of the present invention is: in step 1, the key accessories include at least a servo valve, a wheel speed sensor, a brake pressure sensor, a cornering feedback sensor, a cornering control valve, a cornering actuator, and hydraulic solenoid valves for door / door lock retraction and extension and for landing gear / landing gear lock switching.

[0009] A further technical solution of the present invention is: in step 1, the fault model of the servo valve includes at least one of the following fault modes: Output pressure deviation fault is mapped to the introduction of an adjustable bias in the "pressure-current" static characteristic curve of the servo valve. A short circuit fault in the control coil is mapped to the degradation or zeroing of the electromagnetic resistance, electromagnetic inductance, and electromagnetic force constant in the torque motor model. The degree of fault is continuously adjusted by the short circuit proportional parameter. An open circuit fault in the control coil is mapped to a forced zero coil current, zero electromagnetic force, and the valve core returning to the mechanical zero position under the action of the spring. The mechanical jamming fault of the valve core is mapped to the introduction of a nonlinear friction model into the valve core dynamic equation, and the severity of the jamming is continuously adjusted by the parameters of the Stribeck or LuGre model.

[0010] A further technical solution of the present invention is: in step 1, the fault models of the wheel speed sensor, brake pressure sensor, and cornering feedback sensor include at least one of the following fault modes: Signal drift fault is mapped as a time-varying deviation superimposed on the real output signal. The time-varying deviation is a constant value or other time-varying function, which includes a linear drift function or a more complex function. Among them, the more complex functions include a first-order hysteresis function used to simulate the drift gradually approaching a stable value from the initial value and a periodic function to simulate the sensor output fluctuation caused by periodic temperature changes. A complete signal failure is mapped to an output signal that is constant at zero, at full scale, or at a random value. A further technical solution of the present invention is: in step 1, the fault model of the turning control valve includes at least one of the following fault modes: The modeling method for valve core mechanical jamming fault is the same as that for servo valve valve core mechanical jamming fault mode. Internal leakage increases the fault, which is mapped to adding a parameterized leakage flow path between the valve ports, with the flow area as a parameter of the fault severity; A failure of the pressure reducing valve is reflected in a change in the stiffness or pre-compression of the pressure regulating spring, causing the set pressure to drift.

[0011] A further technical solution of the present invention is: in step 1, the fault model of the turning actuator includes at least one of the following fault modes: Piston seal failure leads to internal leakage, which is reflected as internal flow between the two chambers of the actuator. The leakage flow rate is calculated using a nonlinear throttling formula. Piston rod seal failure leads to external leakage, which is reflected as oil leakage from the high-pressure chamber to the external environment. The leakage path dynamically switches according to the direction of movement.

[0012] A further technical solution of the present invention is: in step 1, the fault model of the hydraulic solenoid valve includes at least one of the following fault modes: The modeling method for coil short circuit or open circuit faults is the same as that for servo valve control coil short circuit or open circuit fault modes. The valve core mechanical jamming fault is mapped to a forced overwrite of the valve core displacement state variable to a fixed value and a zero velocity.

[0013] A further technical solution of the present invention is as follows: In step 2, the fault enable control interface accepts a Boolean type trigger signal. When the signal is "0", the fault kernel output is zero and the model behaves as a fault-free state. When the signal is "1", the fault mode of the current attachment is activated. The fault control interface accepts a set of parameters to dynamically define the severity and evolution mode of the fault. The model physical coupling interface inherits or interfaces with the mechanical, hydraulic, electrical and signal ports of the original attachment to realize the plug-and-play or parameter replacement integration of the fault model with normal components. A further technical solution of the present invention is: the three types of interfaces configured in the fault models of the servo valve, wheel speed sensor, brake pressure sensor, turning feedback sensor, turning control valve, turning actuator and hydraulic solenoid valve are three-layer standardized interface structures: each accessory fault model includes a fault enable control interface, a set of fault control interfaces corresponding to each fault mode, and a set of model physical coupling interfaces that are completely consistent with the ports of normal components. Among them, the parameters received by the fault control interface of different accessory models correspond to the adjustable variables in the mathematical models of each fault mode in step 1, including bias current, short circuit ratio, jamming coefficient, wear degree and leakage area. The port definition of the physical coupling interface of the model corresponds to the physical port of the mathematical model in step 1; step 2 encapsulates the mathematical model established in step 1 through three types of interfaces in a standardized manner, so that the mathematical model can obtain plug-and-play integration capability.

[0014] A further technical solution of the present invention is: in step 3, the structured storage method is as follows: the standardized fault model name reflects the unique identifier and the subsystem to which it belongs, and the description document includes the fault mode / normal mode, interface information and version information of the standardized fault model, so as to support the retrieval, management and experimental design of the standardized fault model.

[0015] Beneficial effects The beneficial effects of this invention are as follows: This invention provides a method for constructing a multi-accessory fault model library that supports integrated simulation of aircraft takeoff and landing systems. It can establish a series of standardized fault models for key multi-physical domain accessories in aircraft takeoff and landing systems, based on deep physical failure mechanisms, possessing high simulation fidelity and continuously adjustable fault severity, to support high-confidence integrated simulation verification and safety analysis. This invention accurately maps the fault mechanisms of each accessory to quantifiable variation rules for specific parameters or structures in the simulation model and establishes a unified interface specification, enabling fault models constructed for different types of accessories (hydraulic, mechanical, electrical, sensor) to be standardized, managed, and easily integrated into system simulation.

[0016] Compared with existing technologies, the model established using this method has the following advantages: 1. This invention achieves continuous adjustment of fault severity by introducing scalar, vector, and time series functions as inputs to the fault control interface; the simulation results and experimental results of each accessory model have an error of less than 5%; 2. This invention achieves non-intrusive fault injection by defining a three-layer standardized interface. Each accessory model can be connected in series in the original system. The fault enable interface can directly switch the fault mode or normal mode of the model without modifying the internal logic of the upstream and downstream models or replacing the fault model before simulation. The independent fault control interface allows for dynamic changes to the fault injection timing, fault state, fault type, fault severity, and different fault combinations during simulation. 3. All fault effects in this invention are controlled by parameters, which facilitates sensitivity analysis and batch simulation testing; 4. This invention achieves exhaustive coverage of fault test scenarios. On the one hand, all fault models adopt continuous parameter design (such as the degree of jamming and leakage coefficient, which are continuously adjustable in the range of 0 to 1), and intermediate states can be arbitrarily set from "no fault" to "complete failure", without any coverage blind spots caused by discrete grading; on the other hand, through automated scripts, the combination space of multiple fault modes and multiple severity levels can be systematically traversed. With m fault modes and n severity levels, n^m test cases can be generated and executed in batches, achieving theoretically exhaustive combination coverage and providing complete test input for the safety verification of aircraft take-off and landing systems. Attached Figure Description

[0017] Figure 1 This is a diagram illustrating the composition of an aircraft takeoff and landing system in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the establishment of a continuously adjustable aircraft fault model library construction method in an embodiment of the present invention. Detailed Implementation

[0018] The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.

[0019] Simulation-based fault injection techniques are gradually becoming an alternative. Several related methods already exist in the existing technology: Patent CN104536303A proposes a fault injection method for auxiliary power systems based on dSPACE, which simulates faults through Simulink modeling and fault trees. However, its fault model is tightly coupled with the main system, lacks a standardized encapsulation interface, and the severity of the fault can only be adjusted to a limited extent through preset discrete parameters (such as 10% or 20%), making it impossible to simulate the continuous evolution process of the fault.

[0020] Patents CN105260555A and CN106250608A, based on the Modelica language, inject faults by annotating fault parameters. However, these two technologies only support static parameter replacement (e.g., replacing normal parameters with fixed fault values), and the fault severity is also discretely set and cannot be dynamically adjusted during simulation. Furthermore, their fault injection requires modifying or reloading the model before simulation, lacking the flexibility of "plug and play."

[0021] Patent CN106682298A constructs a fault simulation model library for aviation hydraulic servo systems, allowing changes to fault parameters and injection timing. However, it still requires model replacement (replacing the normal model with the fault model) before simulation, and cannot achieve dynamic activation / bypass of faults during runtime. Furthermore, its fault severity is limited to finite discrete values ​​such as 10%, 20%, 50%, and 100%, failing to accurately simulate actual physical degradation processes.

[0022] Patent CN115659444A defines fault propagation logic and interfaces based on Modelica, but its focus is on modeling the relationship between fault trees and fault propagation. It does not involve the continuous adjustment of the severity of faults at the appendix level, nor does it provide a unified three-layer encapsulation interface that supports dynamic injection.

[0023] To address the aforementioned problems, this invention proposes a method for constructing a continuously adjustable aircraft fault model library, comprising the following steps: Step 1: Establish a mathematical model that includes faults. For each type of critical accessory of the aircraft take-off and landing system, based on historical fault data, fault mode and effect analysis and physical failure analysis, select fault modes and map the abstract fault phenomena into quantifiable and adjustable variables or structures in the simulation model to establish the mechanism mapping between accessory fault modes and model variables. Step 2: Establish a standardized fault model encapsulation format and interface specification for integrated simulation. Based on the mathematical model in Step 1, configure three types of interfaces for each auxiliary model: a fault enable control interface for activating or bypassing the fault kernel, a fault control interface for dynamically defining fault severity and evolution mode, and a model physical coupling interface that exchanges signals or data with the main system and is completely consistent with the ports of normal components, thus obtaining standardized encapsulation. The fault control interface supports input of scalar, vector, and even time series functions, thereby realizing continuous adjustment of fault severity and exhaustive coverage of multiple fault combination scenarios. Step 3: Store the standardized fault model and its descriptive information encapsulated in Step 2 in a structured manner to obtain a standardized fault model library that can be continuously adjusted and is plug-and-play integrated.

[0024] The above technical solution will be further explained below with reference to examples and accompanying drawings: In one embodiment, see Figure 1 As shown, a critical accessory fault model library for an aircraft take-off and landing system is constructed, including a braking subsystem, a steering subsystem, and a landing gear retraction and extension subsystem. It is an electromechanical-hydraulic integrated system composed of accessories such as landing gear control unit, brake servo valve assembly, wheel speed sensor, brake pressure sensor, steering control valve, steering feedback sensor, steering actuator, hydraulic solenoid valve (for cabin door or cabin door lock retraction and extension), and hydraulic solenoid valve (for landing gear or landing gear lock switching).

[0025] For aircraft takeoff and landing systems, the key accessories include servo valves, wheel speed sensors, brake pressure sensors, turn feedback sensors, turn control valves, turn actuators, and hydraulic solenoid valves (used for door / door lock retraction and extension, and landing gear / landing gear lock switching).

[0026] Reference Figure 2 As shown, the method for establishing a critical accessory fault model library for the aircraft takeoff and landing system includes the following steps: Step 1. Establish a mathematical model that includes the fault: To establish a mathematical model that includes faults, a mechanistic mapping between accessory fault modes and model variables should be established. For each type of accessory in an aircraft takeoff and landing system, the construction of its fault model should follow the following mapping principles: First, based on historical failure data, FMEA analysis, and physical failure analysis, representative failure modes are selected. Then, the abstract failure phenomena are transformed into specific operable variables or structures in the simulation model.

[0027] (1) Servo valve fault model: Typical failure modes of servo valves include, but are not limited to: output pressure deviation failure, control coil short circuit failure, control coil open circuit failure, and valve core mechanical jamming failure.

[0028] Servo valve fault mode 1: Output pressure deviation fault; The output pressure deviation fault is mapped to the static characteristic curve of the servo valve by introducing a bias I_bias. With an adjustable bias I_bias superimposed on the current input signal interface, the simplified linear model expression for the servo valve output pressure deviation fault is as follows: P_out = K_p×(I_in - I_0+ I_bias) Where P_out is the servo valve output pressure, I_in is the input control current, K_p is the pressure gain, I_0 is the zero-point current, and I_bias is the bias.

[0029] Servo valve fault mode 2: Control coil short circuit fault; A short-circuit fault in the control coil is mapped to the degradation or zeroing of electromagnetic parameters (electromagnetic resistance R0, electromagnetic inductance L0, electromagnetic force constant K_t0) in the torque motor model. The fault severity is the short-circuit ratio S_short (0~1). The parameters after the fault can be modeled as follows: R_fault = R0 × (1 - S_short) L_fault = L0 × (1 - S_short)^2 (approximate) K_t_fault =K_t0×(1 - S_short) Where R_fault represents the coil resistance (Ω) after the fault, and L_fault represents the inductance (H). , K_t_fault represents the electromagnetic force constant (N / A); R0 represents the coil resistance under normal conditions, L0 represents the coil inductance under normal conditions, and K_t0 represents the electromagnetic force constant under normal conditions.

[0030] The electrical equation changes from u = R0×i + L0×di / dt under normal conditions to u = R_fault × i + L_fault × di / dt under fault conditions; The electromagnetic force changes from F_m = K_t0×i under normal conditions to F_m = K_t_fault×i under fault conditions. Where u is the coil driving voltage; i is the coil current; and F_m is the electromagnetic force.

[0031] During modeling, in the torque motor electrical sub-model of the servo valve, R0 is... 、 L0 、 The three parameters K_t0 are changed from constants to variables controlled by the input port. A fault configuration function is created to calculate and update these three parameter values ​​in real time based on the externally input S_short value.

[0032] Servo valve fault mode 3: Control coil open circuit fault; An open-circuit fault in the control coil is mapped to a coil current i=0, an electromagnetic force F_m=0, and the valve core returns to and remains in the mechanical zero position under the action of the reset spring.

[0033] Electrical equation: The loop impedance is infinite, i≡0.

[0034] Mechanical equation: The electromagnetic force term disappears in the valve core dynamics equation. Under the action of the return spring, the valve core moves towards the mechanical zero position. The equation of motion is:

[0035] Where m is the valve core mass, B is the viscous damping coefficient, K is the spring stiffness, and t is the current simulation time. The final valve core displacement x tends towards the spring equilibrium position x_0. The servo valve output pressure is 0 or stabilizes at a fixed value, no longer responding to control commands. During modeling, an ideal switch controlled by a fault signal is connected in series in the coil circuit. When a fault is triggered, i.e., when the control coil is open and the control switch signal S_open is 0, the switch opens, the current is forced to zero, and the valve core displacement state variable is locked to the mechanical zero position.

[0036] Servo valve failure mode 4: Valve core mechanical jamming failure; The mechanical jamming fault of the valve core is mapped to the introduction of a nonlinear friction model into the valve core dynamic equation, and the introduction / increase of nonlinear friction parameters (F_s, F_c) based on the Stribeck or LuGre model.

[0037] The classic Stribeck friction model describes this phenomenon, and the relationship between the frictional force F_f and the velocity v is as follows: F_f(v) = [F_c + (F_s - F_c) × e^{-(|v| / v_s)^δ}] × sgn(v) + B_v × v Where: F_s is the maximum static friction force, which increases significantly during a fault, simulating starting difficulties caused by particle jamming; F_c is the Coulomb friction force, which increases during a fault, simulating constant resistance during motion; v_s is the Stribeck velocity, whose value may change during a fault, affecting the frictional transition characteristics; B_v is the viscous friction coefficient, which usually does not change much. δ is the Stribeck exponent, a shape factor that controls the steepness of the friction curve's descent, typically taking a value between 0.5 and 2.

[0038] The severity of the valve core mechanical jamming fault, S_stiction1 [0~1], can be mapped to a function of these parameters, for example: F_s=F_s0×(1+α×S_stiction1), where F_s0 is the normal value and α is the amplification factor. During modeling, in the force balance equation of the servo valve core, the friction term is replaced from the simple linear viscous damping B×v to the aforementioned nonlinear Stribeck friction model. During modeling, in simulation software (such as SimulationX, AMESim), a "nonlinear friction" function module is called or defined, with the valve core movement speed v as input and the fault friction force F_f as output. Key parameters such as F_s and F_c are used as external input interfaces for the model and controlled through S_stiction1.

[0039] (2) Fault models for wheel speed sensor, brake pressure sensor, and steering feedback sensor: These types of sensors have similar failure modes, and are modeled using the same method.

[0040] Sensor fault mode 1: Signal drift fault; Sensor output signal drift faults are mapped as a slowly varying deviation, Bias(t), superimposed on the true output signal. The model is established according to the following relationship: ω_meas(t)=ω_real(t)+S_stiction2×Bias(t) Where ω_meas(t) is the actual measured output value of the sensor; ω_real(t) is the actual measured physical quantity value; Bias(t) is the bias value; and S_stiction2 is the severity of the signal drift fault. The start of time t is the moment of fault injection.

[0041] Bias(t) can be a constant value or other time-varying functions, including the linear drift function Bias(t) = a × t (where a is the drift velocity, defined as a fault parameter) or more complex functions, representing a time-varying bias signal generator.

[0042] The more complex function includes a first-order hysteresis function to simulate the drift gradually approaching a stable value from the initial value: Bias(t) = Bias final (1-e -t / τ ); Among them, Bias final The maximum drift (V) is defined as a fault parameter; τ is a time constant that controls the drift speed.

[0043] More complex functions include periodic functions to simulate fluctuations in sensor output caused by periodic temperature changes: Bias(t) = A sin(2πft + Ø); Where A is the maximum drift amplitude, defined as a fault parameter; f is the drift frequency, such as the temperature change frequency; and Ø is the initial phase.

[0044] Sensor failure mode 2: Complete signal failure; A complete sensor signal failure is mapped to an output signal that is constant at zero, at its maximum value, or a random value. A multiplexer is used to output the true signal under normal conditions and switch to a fixed fault value (0, maximum value, or noise) during a fault. The model is established based on the following relationships: ω_meas(t) = 0 (output zero), ω_meas(t) = Full_Scale (output full scale), or ω_meas(t) = Random (output random value). Where ω_meas(t) is the actual measured output value of the sensor; Full_Scale is the output full scale; and Random is the output random value.

[0045] (3) Fault model of turning control valve: Fault mode 1: Valve core mechanical jamming fault; This fault modeling is exactly the same as the mechanical jamming fault mode of the brake servo valve core.

[0046] Fault Mode 2: Increased internal leakage; Internal leakage in a turning control valve refers to the increased clearance between the valve core and valve sleeve due to wear, causing hydraulic oil to leak directly from the pressure port P to the return port T, without participating in the actuator drive.

[0047] Leakage flow rate is described using the orifice flow rate formula: ×

[0048] in, It is leakage flow; It is the leakage coefficient; It is the pressure difference across the valve port; Leakage coefficient The relationship with the radial clearance h is as follows:

[0049] Where: b is the circumference of the gap ring; h is the clearance between the valve sleeve and the valve body; μ is the viscosity of the hydraulic oil; and l is the width of the clearance. The severity of the fault is reflected by the clearance h between the valve sleeve and the valve body:

[0050] in, This is a normal gap (usually between 1-5 μm). This is the maximum gap (typically between 30-100 μm). This is a parameter indicating the severity of the fault (between 0 and 1).

[0051] Fault mode 3: Pressure reducing valve malfunction (set pressure drift); The failure of the pressure reducing valve in the turning control valve is mapped to a change in stiffness k of the pressure regulating spring inside the valve due to fatigue, or a change in pre-compression x0 due to wear of the pilot spring seat, which causes the set pressure to deviate from the original value.

[0052] Ideally set pressure: ; Actual set pressure after the malfunction: / S Where: k0 is the nominal stiffness of the pressure regulating spring, in Newtons per meter (N / m); Δkmax is the maximum change in spring stiffness; severity_k is the degree of change in spring stiffness, [0,1]; x0 is the spring pre-compression, in meters (m); Δxmax is the maximum change in pre-compression (reduced due to wear); severity_x is the degree of change in pre-compression, [0,1]; S is the pressure-bearing area of ​​the valve core, in square meters (m²). 2 ).

[0053] During modeling, in the pressure-reducing valve sub-model of the combined valve, the spring stiffness k and pre-compression x0 are set to be externally configurable. Static drift faults at the set pressure are simulated by changing Δk and Δx.

[0054] (4) Turning actuator; Failure mode 1: Piston seal failure leading to internal leakage; Internal leakage in a turning actuator, caused by piston seal failure, manifests as wear, aging, or scratches on the piston dynamic seal (such as a Glyd ring), leading to undesirable internal oil flow between the two chambers (chamber A and chamber B) of the actuator. Internal leakage flow rate (m³) 3 / s) The calculation formula is as follows:

[0055] in, This is the normal internal leakage coefficient, which is directly proportional to the degree of seal damage. P_A is the pressure in chamber A of the actuator, and P_B is the pressure in chamber B of the actuator (Pa).

[0056]

[0057] in, It is the normal internal leakage coefficient (usually a very small fixed value, such as 1e-10). It is the maximum internal leakage coefficient (usually 1e-6). It indicates the severity of the internal leakage fault (between 0 and 1).

[0058] During modeling, a fault leakage branch is directly connected between the hydraulic ports of chambers A and B of the actuator. This branch is the nonlinear resistance described by the above formula. In the SimulationX hydraulic library, this can be achieved using the Hydraulic Leakage component or a custom Hydraulic Resistor.

[0059] Failure mode 2: Piston rod seal failure leading to external leakage; Piston seal failure in a turning actuator leads to external leakage, which in turn affects the piston rod seal (such as a step seal). This causes oil in the high-pressure chamber (usually the pressure-bearing side) to leak directly into the external environment, resulting in decreased pressure holding capacity and oil loss. The formula for calculating the external leakage flow rate is as follows:

[0060] in: P_active represents the external leakage flow rate (m3 / s); P_active represents the pressure of the current pressure chamber of the actuator cylinder, which needs to be determined based on the direction of movement to determine whether it is chamber A or chamber B. The external leakage coefficient is... ; This is the maximum external leakage coefficient (e.g., 1e-5). The severity of the external leakage fault (between 0 and 1).

[0061] During modeling, an additional leakage branch is led from the hydraulic port of the pressure chamber of the actuator cylinder to the oil tank (ambient pressure). Logic is needed to determine which chamber is currently pressure-bearing and dynamically switch the leakage path. This can be implemented using a controlled switch in the simulation.

[0062] (5) Hydraulic solenoid valve (used for hatch / hatch door lock retraction and extension, landing gear / landing gear lock switch). Fault mode 1: Coil short circuit / open circuit fault; Principle and Modeling: Similar to the electrical fault modes of a brake servo valve. Resistance and electromagnetic force decrease during a short circuit; electromagnetic force is zero during an open circuit.

[0063] Fault mode 2: Valve core mechanical jamming fault; The mechanical jamming fault of the hydraulic solenoid valve spool is mapped by forcibly overriding the spool displacement state variable to a fixed value and setting the velocity to zero. Modeling principle formula: x(t) = X_jammed (when severity3 > 0), v(t) = 0, a(t) = 0 Where x(t) is the valve core displacement; severity3 is the severity of the fault. Since the solenoid valve jamming is a binary state of "jammed" and "not jammed", there is no continuously adjustable fault severity. When severity3=0, it is the normal movement state of the valve core, and when severity3=1, it is the jammed state of the valve core; X_jammed is the jammed position, which is a constant; v(t) is the valve core velocity; a(t) is the valve core acceleration.

[0064] Step 2. Establish a standardized fault model encapsulation format and interface specification for integrated simulation: The standardized encapsulation format of the fault model specifies that each accessory model exposes three types of interfaces: (1) Fault Enable Control Interface: Receives a Boolean trigger signal. When the received signal is “0”, the fault kernel output is zero, and the model behaves as a fault-free state. When the received signal is “1”, the current accessory fault mode is activated and affects the system.

[0065] (2) Fault control interface: Input of severity of different fault modes, used to receive a set of parameters for dynamically defining the severity and evolution mode of the fault, and supports input of constant value and time-varying function.

[0066] (3) Model physical coupling interface: This is the port for exchanging signals or data with the main system. Its type is completely consistent with the port of the normal component being replaced. Inherit or interface with the physical ports (mechanical, hydraulic, electrical, signal) of the original accessory model to achieve "plug and play" or "parameter replacement" integration.

[0067] The interfaces of the fault models of each attachment established in step 2 are shown in the table below.

[0068]

[0069] The interface of the servo valve fault model includes a fault enable control interface for switching between fault and normal modes of the model; a set of independent fault control interfaces for receiving fault type and severity, where I_bias, S_short, S_open, and S_stiction1 correspond to fault modes 1, 2, 3, and 4 of the servo valve fault model in step 1, respectively; and a set of model physical coupling interfaces, which are completely consistent with the normal component ports.

[0070] The interfaces for the fault models of the wheel speed sensor, brake pressure sensor, and cornering feedback sensor include a fault enable control interface for switching between fault and normal modes of the model; a set of independent fault control interfaces for receiving fault type and severity, where Bias and S_stiction2 correspond to fault mode 1 of the fault models of the wheel speed sensor, brake pressure sensor, and cornering feedback sensor in step 1; and meas corresponds to fault mode 2 of the fault models of the wheel speed sensor, brake pressure sensor, and cornering feedback sensor in step 1; and a set of model physical coupling interfaces, which are completely consistent with the ports of normal components.

[0071] The interface of the turning control valve fault model includes a fault enable control interface for switching between fault and normal modes of the model; it includes a set of independent fault control interfaces for receiving fault type and severity, where S_stiction3 corresponds to fault mode 1 of the turning control valve fault model in step 1; severity corresponds to fault mode 2 of the turning control valve fault model in step 1; severity_k and severity_x correspond to fault mode 3 of the turning control valve fault model in step 1; and it includes a set of model physical coupling interfaces, which are completely consistent with the normal component ports.

[0072] The interface of the turning actuator fault model includes a fault enable control interface for switching between fault and normal modes of the model; it includes a set of independent fault control interfaces for receiving fault type and severity, with Severity_k1 and severity_k2 corresponding to fault modes 1 and 2 of the turning actuator fault model in step 1, respectively; and it includes a set of model physical coupling interfaces, which are completely consistent with the normal component ports.

[0073] The interface of the hydraulic solenoid valve fault model includes a fault enable control interface for switching between fault and normal modes of the model; a set of independent fault control interfaces for receiving fault type and severity, where S_short1 and S_open1 correspond to fault mode 1 of the hydraulic solenoid valve fault model in step 1; severity3 and S_JamPos correspond to fault mode 2 of the hydraulic solenoid valve fault model in step 1; and a set of model physical coupling interfaces, which are completely consistent with the normal component ports.

[0074] Step 3. Store the fault model and its descriptive information in a structured manner: When naming a model, it must reflect its unique identifier and the subsystem it belongs to. The model's documentation should include its fault / normal modes, interface information, author and version, and modeling date. This information is not used in real-time simulation but serves to support model retrieval, management, and experimental design. The table below shows examples of the corresponding names and descriptions for stored models.

[0075]

[0076] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A method for constructing a continuously adjustable aircraft fault model library, characterized in that, Includes the following steps: Step 1: Establish a mathematical model that includes faults. For each type of critical accessory of the aircraft take-off and landing system, based on historical fault data, fault mode and effect analysis and physical failure analysis, select fault modes and map the abstract fault phenomena into quantifiable and adjustable variables or structures in the simulation model to establish the mechanism mapping between accessory fault modes and model variables. Step 2: Establish a standardized fault model encapsulation format and interface specification for integrated simulation. Based on the mathematical model in Step 1, configure three types of interfaces for the fault models of each attachment: a fault enable control interface for activating or bypassing the fault kernel, a fault control interface for dynamically defining the fault severity and evolution mode, and a model physical coupling interface that exchanges signals or data with the main system and is completely consistent with the port of normal components, thereby obtaining standardized encapsulation. The fault control interface supports inputs of scalar, vector, and even time series functions, thereby enabling continuous adjustment of fault severity and exhaustive coverage of multiple fault combination scenarios. Step 3: Store the standardized fault model and its descriptive information encapsulated in Step 2 in a structured manner to obtain a standardized fault model library that can be continuously adjusted and is plug-and-play integrated.

2. The method for constructing a continuously adjustable aircraft fault model library according to claim 1, characterized in that: In step 1, the key accessories include at least a servo valve, a wheel speed sensor, a brake pressure sensor, a cornering feedback sensor, a cornering control valve, a cornering actuator, and hydraulic solenoid valves for door / door lock retraction and extension and for landing gear / landing gear lock switching.

3. The method for constructing a continuously adjustable aircraft fault model library according to claim 2, characterized in that: In step 1, the fault model of the servo valve includes at least one of the following fault modes: Output pressure deviation fault is mapped to the introduction of an adjustable bias in the "pressure-current" static characteristic curve of the servo valve. A short circuit fault in the control coil is mapped to the degradation or zeroing of the electromagnetic resistance, electromagnetic inductance, and electromagnetic force constant in the torque motor model. The degree of fault is continuously adjusted by the short circuit proportional parameter. An open circuit fault in the control coil is mapped to a forced zero coil current, zero electromagnetic force, and the valve core returning to the mechanical zero position under the action of the spring. The mechanical jamming fault of the valve core is mapped to the introduction of a nonlinear friction model into the valve core dynamic equation, and the severity of the jamming is continuously adjusted by the parameters of the Stribeck or LuGre model.

4. The method for constructing a continuously adjustable aircraft fault model library according to claim 3, characterized in that: In step 1, the fault models for the wheel speed sensor, brake pressure sensor, and cornering feedback sensor include at least one of the following fault modes: Signal drift fault is mapped as a time-varying deviation superimposed on the real output signal. The time-varying deviation is a constant value or other time-varying function, which includes a linear drift function or a more complex function. Among them, the more complex functions include a first-order hysteresis function used to simulate the drift gradually approaching a stable value from the initial value and a periodic function to simulate the sensor output fluctuation caused by periodic temperature changes. A complete signal failure is mapped to an output signal that is constant at zero, at full scale, or at a random value.

5. The method for constructing a continuously adjustable aircraft fault model library according to claim 4, characterized in that: In step 1, the failure model of the turning control valve includes at least one of the following failure modes: The modeling method for valve core mechanical jamming fault is the same as that for servo valve valve core mechanical jamming fault mode. Internal leakage increases the fault, which is mapped to adding a parameterized leakage flow path between the valve ports, with the flow area as a parameter of the fault severity; A failure of the pressure reducing valve is reflected in a change in the stiffness or pre-compression of the pressure regulating spring, causing the set pressure to drift.

6. The method for constructing a continuously adjustable aircraft fault model library according to claim 5, characterized in that: In step 1, the failure model of the turning actuator includes at least one of the following failure modes: Piston seal failure leads to internal leakage, which is reflected as internal flow between the two chambers of the actuator. The leakage flow rate is calculated using a nonlinear throttling formula. Piston rod seal failure leads to external leakage, which is reflected as oil leakage from the high-pressure chamber to the external environment. The leakage path dynamically switches according to the direction of movement.

7. The method for constructing a continuously adjustable aircraft fault model library according to claim 6, characterized in that: In step 1, the failure model of the hydraulic solenoid valve includes at least one of the following failure modes: The modeling method for coil short circuit or open circuit faults is the same as that for servo valve control coil short circuit or open circuit fault modes. The valve core mechanical jamming fault is mapped to a forced overwrite of the valve core displacement state variable to a fixed value and a zero velocity.

8. The method for constructing a continuously adjustable aircraft fault model library according to claim 7, characterized in that: In step 2, the fault enable control interface receives a Boolean trigger signal. When the signal is "0", the fault kernel output is zero and the model behaves as a fault-free state. When the signal is "1", the fault mode of the current attachment is activated. The fault control interface accepts a set of parameters to dynamically define the severity and evolution mode of the fault. The model physical coupling interface inherits or interfaces with the mechanical, hydraulic, electrical and signal ports of the original attachment to realize the plug-and-play or parameter replacement integration of the fault model with normal components.

9. The method for constructing a continuously adjustable aircraft fault model library according to claim 2, characterized in that: The fault models of the servo valve, wheel speed sensor, brake pressure sensor, cornering feedback sensor, cornering control valve, cornering actuator and hydraulic solenoid valve are configured with three types of interfaces in a three-layer standardized interface structure: each accessory fault model includes a fault enable control interface, a set of fault control interfaces corresponding to each fault mode, and a set of model physical coupling interfaces that are completely consistent with the ports of normal components. Among them, the parameters received by the fault control interface of different accessory models correspond to the adjustable variables in the mathematical models of each fault mode in step 1, including bias current, short circuit ratio, jamming coefficient, wear degree and leakage area. The port definition of the physical coupling interface of the model corresponds to the physical port of the mathematical model in step 1; step 2 encapsulates the mathematical model established in step 1 through three types of interfaces in a standardized manner, so that the mathematical model can obtain plug-and-play integration capability.

10. A method for constructing a continuously adjustable aircraft fault model library according to any one of claims 1-9, characterized in that: In step 3, the structured storage method includes: reflecting the unique identifier and the subsystem to which the standardized fault model belongs in the naming of the standardized fault model, and including the fault mode and normal mode, interface information and version information of the standardized fault model in the specification document, so as to support the retrieval, management and experimental design of the standardized fault model.

Images