Flight control servo actuation system fault identification method and system

Abstract

The invention provides a flight control servo actuation system fault identification method and system, and relates to the technical field of aviation system detection, and the method comprises the steps: obtaining a historical data set; training the fault recognition model based on the historical data set to obtain a trained fault recognition model; constructing a performance model of the to-be-identified system based on the physical architecture and the design parameter model of the to-be-identified system; performing sensitivity analysis on the performance model to obtain an identification parameter set, and performing screening to obtain an observable parameter set and an unobservable parameter set; inputting the data of the observable parameter set and the data of the system output parameter set into the trained fault identification model to obtain an unobservable parameter prediction data set; and judging the unobservable parameter prediction data set in combination with a set threshold range to obtain a fault recognition result. According to the method, the unobservable parameter prediction data set and the set threshold range are compared, so that the unit with the fault is accurately positioned.

Description

Technical Field

[0001] The present invention relates to the technical field of aviation system detection, and in particular to a method and system for identifying faults of a flight control servo actuation system. Background Art

[0002] During the use of the flight control servo actuation system, since the system consists of many units and the parameter cross-linking relationship between the units is complex, unit failure will cause parameter jumps and cause system output abnormalities. Therefore, the design parameters need to be monitored during the system operation. When a fault occurs, safety design measures such as redundancy switching and compensation can be implemented to ensure the normal operation of the system.

[0003] The fault-tolerant implementation of the system's redundancy switching is to analyze the fault behavior based on parameter monitoring, identify the corresponding faulty unit and switch it. At present, the unit fault detection is mainly carried out in the form of multi-signal flow graphs. When the fault occurs, it is reported and the unit switching is realized through the control center. This method provides great support for the safe operation of the system. However, for the flight control servo actuation system, due to the electromechanical and hydraulic coupling phenomenon, some faults cannot be directly measured on the one hand, such as the wear of the electro-hydraulic valve core, the blockage between the electro-hydraulic valve and the mode conversion valve, etc., which cause the pressure supply capacity to be reduced and cannot be directly located, resulting in the need to directly cut off the oil channel after the fault occurs; on the other hand, there are many types of faults in the flight control servo actuation system, such as coupling failure and faults that gradually change over time, and such faults cannot be accurately characterized by models, resulting in the inability to fully identify and analyze the faults. Combined with the above two points, the traditional fault identification and location methods cannot effectively support the accurate location of the faults of the flight control servo actuation system, and potential faults and coupling faults cannot be fully identified, resulting in safety hazards in the flight control servo actuation system during the execution of the mission. Summary of the invention

[0004] The object of the present invention is to provide a method and system for identifying faults in a flight control servo actuation system, which achieves accurate positioning of a faulty unit by comparing an unobservable parameter prediction data set and a set threshold range.

[0005] A method for identifying a fault in a flight control servo actuator system, comprising:

[0006] Acquire a historical data set; the historical data set includes a historical observable parameter data set, a historical unobservable parameter set, and a historical system output parameter data set;

[0007] Constructing a fault identification model, taking the historical observable parameter data set and the historical system output parameter data set as inputs of the fault identification model, taking the historical unobservable parameter set as a label, training the fault identification model, and obtaining the trained fault identification model;

[0008] Construct a performance model of the system to be identified based on the physical architecture and design parameter model of the system to be identified;

[0009] Performing sensitivity analysis on the performance model to obtain an identification parameter set; the identification parameter set includes input instructions, oil supply pressure, actuator output speed, actuator output displacement, servo valve spool displacement, actuator zero point coordinate value and sensor gain;

[0010] The identification parameter set is screened to obtain an observable parameter set and an unobservable parameter set; the observable parameter set includes input instructions, oil supply pressure, actuator output speed and actuator output displacement; the unobservable parameter set includes servo valve spool displacement, actuator zero point coordinates and sensor gain;

[0011] Based on the identification parameter set, input instruction data, oil supply pressure data, actuator output speed data, actuator output displacement data and system output parameter data set of the system to be identified are obtained; the system output parameter data set includes system output speed and system output displacement;

[0012] Input the input command data, oil supply pressure data, actuator output speed data, actuator output displacement data, the system output speed and the system output displacement into the trained fault identification model to obtain a servo valve spool displacement prediction value, an actuator 0 point coordinate value prediction value and a sensor gain prediction value;

[0013] The predicted value of the 0-point coordinate value of the actuator is judged. If the predicted value of the 0-point coordinate value of the actuator is within the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is normal. If the predicted value of the 0-point coordinate value of the actuator exceeds the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is faulty.

[0014] The sensor gain prediction value is judged. If the sensor gain prediction value is within the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is normal. If the sensor gain prediction value exceeds the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is faulty.

[0015] The predicted value of the servo valve spool displacement is judged. If the predicted value of the servo valve spool displacement is within the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is normal. If the predicted value of the servo valve spool displacement exceeds the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is faulty.

[0016] Optionally, the performance model includes an input instruction set, a system output parameter set, and a system parameter set;

[0017] The input instruction set expression is:

[0018] u={u1,u2,…,u i};

[0019] The system output parameter set expression is:

[0020] y={y1,y2};

[0021] The system parameter set expression is:

[0022] x′={x′1,x′2,…,x′ k};

[0023] Where: u represents the input instruction set, y represents the system output parameter set, x' represents the system parameter set, i represents the number of dimensions of the input instruction set, k represents the number of dimensions of the system parameter set, and u i represents the i-th input instruction in the input instruction set, y1 represents the system output speed, y2 represents the system output displacement, x' k Represents the kth system parameter in the system parameter set.

[0024] Optionally, the identification parameter set expression is:

[0025] x={x1,x2,x3,x4,x5,x6,x7};

[0026] The observable parameter set expression is:

[0027] x o ={x1,x2,x3,x4};

[0028] The unobservable parameter set expression is:

[0029] x uo ={x5,x6,x7};

[0030] Where: x represents the identification parameter set, x o represents the observable parameter set, x uo represents the unobservable parameter set, x1 represents the input command, x2 represents the oil supply pressure, x3 represents the actuator output speed, x4 represents the actuator output displacement, x5 represents the actuator 0 point coordinate value, x6 represents the sensor gain, and x7 represents the servo valve core displacement.

[0031] Optionally, the observable parameter data set, the unobservable parameter data set and the system output parameter data set are all Real data.

[0032] The present invention also provides a flight control servo actuation system fault identification system, which comprises:

[0033] A first data acquisition module is used to acquire a historical data set; the historical data set includes a historical observable parameter data set, a historical unobservable parameter set and a historical system output parameter data set;

[0034] A model training module is used to construct a fault identification model, take the historical observable parameter data set and the historical system output parameter data set as inputs of the fault identification model, take the historical unobservable parameter set as a label, train the fault identification model, and obtain the trained fault identification model;

[0035] A performance model module is used to construct a performance model of the system to be identified based on the physical architecture and design parameter model of the system to be identified;

[0036] A sensitivity analysis module, used for performing sensitivity analysis on the performance model to obtain an identification parameter set; the identification parameter set includes input instructions, oil supply pressure, actuator output speed, actuator output displacement, servo valve spool displacement, actuator zero point coordinate value and sensor gain;

[0037] A parameter screening module, used for screening the identified parameter set to obtain an observable parameter set and an unobservable parameter set; the observable parameter set includes input instructions, oil supply pressure, actuator output speed and actuator output displacement; the unobservable parameter set includes servo valve spool displacement, actuator zero point coordinates and sensor gain;

[0038] A second data acquisition module acquires input instruction data, oil supply pressure data, actuator output speed data, actuator output displacement data and system output parameter data set of the system to be identified based on the identification parameter set; the system output parameter data set includes system output speed and system output displacement;

[0039] A data prediction module, inputting the input command data, the oil supply pressure data, the actuator output speed data, the actuator output displacement data, the system output speed and the system output displacement into the trained fault identification model, and obtaining a servo valve spool displacement prediction value, an actuator 0 point coordinate value prediction value and a sensor gain prediction value;

[0040] The first judgment module judges the predicted value of the 0-point coordinate value of the actuator. If the predicted value of the 0-point coordinate value of the actuator is within the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is normal; if the predicted value of the 0-point coordinate value of the actuator exceeds the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is faulty;

[0041] The second judgment module judges the sensor gain prediction value. If the sensor gain prediction value is within the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is normal; if the sensor gain prediction value exceeds the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is faulty.

[0042] The third judgment module is used to judge the predicted value of the servo valve spool displacement. If the predicted value of the servo valve spool displacement is within the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is normal. If the predicted value of the servo valve spool displacement exceeds the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is faulty.

[0043] Optionally, the performance model includes an input instruction set, a system output parameter set, and a system parameter set;

[0044] The input instruction set expression is:

[0045] u={u1,u2,…,u i};

[0046] The system output parameter set expression is:

[0047] y={y1,y2};

[0048] The system parameter set expression is:

[0049] x′={x′1,x′2,…,x′ k};

[0050] Where: u represents the input instruction set, y represents the system output parameter set, x' represents the system parameter set, i represents the number of dimensions of the input instruction set, k represents the number of dimensions of the system parameter set, and u i represents the i-th input instruction in the input instruction set, y1 represents the system output speed, y2 represents the system output displacement, x' k Represents the kth system parameter in the system parameter set.

[0051] Optionally, the identification parameter set expression is:

[0052] x={x1,x2,x3,x4,x5,x6,x7};

[0053] The observable parameter set expression is:

[0054] x o ={x1,x2,x3,x4};

[0055] The unobservable parameter set expression is:

[0056] x uo ={x5,x6,x7};

[0057] Where: x represents the identification parameter set, x o represents the observable parameter set, x uo represents the unobservable parameter set, x1 represents the input command, x2 represents the oil supply pressure, x3 represents the actuator output speed, x4 represents the actuator output displacement, x5 represents the actuator 0 point coordinate value, x6 represents the sensor gain, and x7 represents the servo valve core displacement.

[0058] Optionally, the observable parameter data set, the unobservable parameter data set and the system output parameter data set are all Real data.

[0059] The effects of the present invention are as follows:

[0060] The method for identifying faults in a flight control servo actuator system of the present invention first obtains key parameters that affect the system output based on sensitivity analysis, so that the system output is more targeted and accurate, and performs screening to obtain observable parameters and unobservable parameters, and further combines a trained fault identification model to predict the unobservable parameter set of the system to be identified, so as to obtain an unobservable parameter prediction data set, obtain a set threshold range based on the unobservable parameter data set, and judge the unobservable parameter prediction data set based on the set threshold range to obtain an identification result, accurately locate the unit where the fault occurs, and avoid problems such as failure coupling and unclear parameter association. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of a method for identifying a fault of a flight control servo actuator system of the present invention;

[0062] Figure 2 is a schematic diagram of a performance model of a flight control servo actuation system of the present invention;

[0063] Figure 3 is a schematic diagram of the system output when the sensor gain of the present invention is abnormal;

[0064] Figure 4 It is a schematic diagram of system output when the sensor gain of the present invention is normal. DETAILED DESCRIPTION

[0065] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0066] Figure 1 FIG. 1 is a flow chart of a method for identifying a fault in a flight control servo actuator system of the present invention. Figure 1 The present invention provides a method for identifying a fault in a flight control servo actuator system, which comprises:

[0067] S1, obtain historical data sets. The historical data sets include historical observable parameter data sets, historical unobservable parameter data sets, and historical system output parameter data sets.

[0068] S2, build a fault recognition model, use the historical observable parameter data set and the historical system output parameter data set as the input of the fault recognition model, use the historical unobservable parameter set as a label, train the fault recognition model, and obtain a trained fault recognition model.

[0069] S3, constructing a performance model of the system to be identified based on the physical architecture and design parameter model of the system to be identified.

[0070] S4, sensitivity analysis is performed on the performance model to obtain an identification parameter set, which includes input command, oil supply pressure, actuator output speed, actuator output displacement, servo valve spool displacement, actuator zero point coordinate value and sensor gain.

[0071] S5, screening the identified parameter set to obtain an observable parameter set and an unobservable parameter set. The observable parameter set includes input instructions, oil supply pressure, actuator output speed and actuator output displacement; the unobservable parameter set includes servo valve spool displacement, actuator zero point coordinates and sensor gain.

[0072] S6, based on the identification parameter set, the input instruction data, the oil supply pressure data, the actuator output speed data, the actuator output displacement data and the system output parameter data set of the system to be identified are obtained. The system output parameter data set includes the system output speed and the system output displacement. Preferably, the observable parameter data set, the unobservable parameter data set and the system output parameter data set are all Real data.

[0073] S7, input the input command data, oil supply pressure data, actuator output speed data, actuator output displacement data, system output speed and system output displacement into the trained fault identification model to obtain the servo valve spool displacement prediction value, actuator 0 point coordinate value prediction value and sensor gain prediction value.

[0074] S8, judge the predicted value of the actuator 0-point coordinate value. If the predicted value of the actuator 0-point coordinate value is within the actuator set threshold range, the unit in the system to be identified corresponding to the actuator 0-point coordinate value is normal. If the predicted value of the actuator 0-point coordinate value exceeds the actuator set threshold range, the unit in the system to be identified corresponding to the actuator 0-point coordinate value is faulty.

[0075] S9, judging the sensor gain prediction value. If the sensor gain prediction value is within the sensor set threshold range, the unit in the system to be identified corresponding to the sensor gain is normal. If the sensor gain prediction value exceeds the sensor set threshold range, the unit in the system to be identified corresponding to the sensor gain is faulty.

[0076] S10, judging the predicted value of the servo valve core displacement. If the predicted value of the servo valve core displacement is within the servo valve set threshold range, the unit in the system to be identified corresponding to the servo valve core is normal. If the predicted value of the servo valve core displacement exceeds the servo valve set threshold range, the unit in the system to be identified corresponding to the servo valve core is faulty.

[0077] Preferably, the actuator setting threshold range and the sensor setting threshold range are determined based on experimental data or based on expert knowledge.

[0078] Specifically, the performance model includes an input instruction set, a system output parameter set, and a system parameter set.

[0079] The input instruction set expression is:

[0080] u={u1,u2,…,u i}.

[0081] The system output parameter set expression is:

[0082] y={y1,y2}.

[0083] The system parameter set expression is:

[0084] x′={x′1,x′2,…,x′ k};

[0085] Where: u represents the input instruction set, y represents the system output parameter set, x' represents the system parameter set, i represents the number of dimensions of the input instruction set, k represents the number of dimensions of the system parameter set, and u i represents the i-th input instruction in the input instruction set, y1 represents the system output speed, y2 represents the system output displacement, x' k Represents the kth system parameter in the system parameter set.

[0086] Preferably, the system parameter set includes input instructions, oil supply pressure, actuator output speed, actuator output displacement, servo valve core displacement, actuator 0 point coordinate value, sensor gain, internal leakage between the actuator drive chamber and the oil return chamber, external leakage between the actuator drive chamber and the oil return chamber, solenoid coil current input and solenoid coil voltage input.

[0087] Furthermore, the identification parameter set expression is:

[0088] x={x1,x2,x3,x4,x5,x6,x7}.

[0089] The observable parameter set expression is:

[0090] x o ={x1,x2,x3,x4}.

[0091] The expression of the unobservable parameter set is:

[0092] x uo ={x5,x6,x7};

[0093] Where: x represents the identification parameter set, x o represents the observable parameter set, x uo represents the unobservable parameter set, x1 represents the input command, x2 represents the oil supply pressure, x3 represents the actuator output speed, x4 represents the actuator output displacement, x5 represents the actuator 0 point coordinate value, x6 represents the sensor gain, and x7 represents the servo valve core displacement.

[0094] The present invention also provides a flight control servo actuation system fault identification system, which comprises:

[0095] The first data acquisition module is used to acquire a historical data set; the historical data set includes a historical observable parameter data set, a historical unobservable parameter set and a historical system output parameter data set.

[0096] The model training module is used to build a fault identification model. The historical observable parameter data set and the historical system output parameter data set are used as the input of the fault identification model, and the historical unobservable parameter set is used as a label to train the fault identification model to obtain a trained fault identification model.

[0097] The performance model module is used to build a performance model of the system to be identified based on the physical architecture and design parameter model of the system to be identified.

[0098] The sensitivity analysis module is used to perform sensitivity analysis on the performance model to obtain an identification parameter set, which includes input command, oil supply pressure, actuator output speed, actuator output displacement, servo valve spool displacement, actuator zero point coordinate value and sensor gain.

[0099] The parameter screening module is used to screen the identification parameter set to obtain the observable parameter set and the unobservable parameter set. The observable parameter set includes the input command, the oil supply pressure, the actuator output speed and the actuator output displacement. The unobservable parameter set includes the servo valve spool displacement, the actuator zero point coordinate and the sensor gain.

[0100] The second data acquisition module acquires input instruction data, oil supply pressure data, actuator output speed data, actuator output displacement data and system output parameter data set of the system to be identified based on the identification parameter set. The system output parameter data set includes system output speed and system output displacement.

[0101] The data prediction module inputs the input command data, oil supply pressure data, actuator output speed data, actuator output displacement data, system output speed and system output displacement into the trained fault identification model to obtain the servo valve spool displacement prediction value, actuator 0 point coordinate value prediction value and sensor gain prediction value.

[0102] The first judgment module judges the predicted value of the actuator 0-point coordinate value. If the predicted value of the actuator 0-point coordinate value is within the threshold range set by the actuator, the unit in the system to be identified corresponding to the actuator 0-point coordinate value is normal. If the predicted value of the actuator 0-point coordinate value exceeds the threshold range set by the actuator, the unit in the system to be identified corresponding to the actuator 0-point coordinate value is faulty.

[0103] The second judgment module judges the sensor gain prediction value. If the sensor gain prediction value is within the sensor set threshold range, the unit in the system to be identified corresponding to the sensor gain is normal. If the sensor gain prediction value exceeds the sensor set threshold range, the unit in the system to be identified corresponding to the sensor gain is faulty.

[0104] The third judgment module is used to judge the predicted value of the servo valve spool displacement. If the predicted value of the servo valve spool displacement is within the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is normal. If the predicted value of the servo valve spool displacement exceeds the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is faulty.

[0105] Optionally, the performance model includes an input instruction set, a system output parameter set, and a system parameter set.

[0106] The input instruction set expression is:

[0107] u={u1,u2,…,u i}.

[0108] The system output parameter set expression is:

[0109] y={y1,y2}.

[0110] The system parameter set expression is:

[0111] x′={x′1,x′2,…,x′ k}.

[0112] Where: u represents the input instruction set, y represents the system output parameter set, x' represents the system parameter set, i represents the number of dimensions of the input instruction set, k represents the number of dimensions of the system parameter set, and u i represents the i-th input instruction in the input instruction set, y1 represents the system output speed, y2 represents the system output displacement, x' k Represents the kth system parameter in the system parameter set.

[0113] Optionally, the identification parameter set expression is:

[0114] x={x1,x2,x3,x4,x5,x6,x7}.

[0115] The observable parameter set expression is:

[0116] x o ={x1,x2,x3,x4}.

[0117] The expression of the unobservable parameter set is:

[0118] x uo ={x5,x6,x7}.

[0119] Where: x represents the identification parameter set, x o represents the observable parameter set, x uo represents the unobservable parameter set, x1 represents the input command, x2 represents the oil supply pressure, x3 represents the actuator output speed, x4 represents the actuator output displacement, x5 represents the actuator 0 point coordinate value, x6 represents the sensor gain, and x7 represents the servo valve core displacement.

[0120] Optionally, the observable parameter data set, the unobservable parameter data set and the system output parameter data set are all Real data.

[0121] Specifically, taking a certain flight control servo actuation system as an example, the flight control servo actuation system receives instructions from the flight control computer, and controls the driving direction and flow rate of the hydraulic circuit by controlling the solenoid valve to achieve the extension and retraction of the actuator, thereby achieving control of the aircraft's control surfaces and attitude.

[0122] Build a performance model of the flight control servo actuation system. The performance model of the flight control servo actuation system is a description of the system's physical architecture design and parameters. By combining the Modelica language, after the system receives instructions, it controls the displacement of the spool of the reversing valve through the control loop, and then controls the switching of the hydraulic oil circuit to achieve the displacement control of the actuator. The performance model is as follows Figure 2 shown.

[0123] According to the performance model established, the input command is given, and the oil supply pressure, the actuator 0 point coordinate value, and the sensor gain parameters are set. The system output response under different input and parameter setting conditions is observed. The output under different parameters is analyzed, and the speed at 0.5s and the displacement value at 1.0s are selected, as shown in Table 1.

[0124] Table 1 Velocity at 0.5s and displacement at 1.0s

[0125]

[0126]

[0127]

[0128] Based on the data in Table 1, combined with the system design and usage characteristics, the observable parameter set includes input command, oil supply pressure, actuator output speed and actuator output displacement; the unobservable parameter set includes the actuator 0 point coordinates and sensor gain, and the system output parameter set includes the system output speed and system output displacement. The system output parameter data when the input command is 1.9 is as follows: Figure 3 shown.

[0129] The input command, oil supply pressure, actuator output speed, actuator output displacement, system output speed and system output displacement are input into the trained fault identification model to obtain the actuator zero point coordinate and sensor gain.

[0130] The coordinate of the actuator 0 point at time 1.0 is 3.55015, and the sensor gain is 1.25. The sensor gain exceeds the corresponding set threshold range, and the corresponding sensor is faulty. The sensor is corrected by adding a proportional link. The proportional value of the correction link is 1.599688. After correction, the system output displacement is 19.5cm. Figure 4 As shown, the system responds normally.

[0131] The embodiments described above are only descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for identifying a flight control servo actuation system fault, characterized in that: It includes: Acquire a historical data set; the historical data set includes a historical observable parameter data set, a historical unobservable parameter set, and a historical system output parameter data set; Constructing a fault identification model, taking the historical observable parameter data set and the historical system output parameter data set as inputs of the fault identification model, taking the historical unobservable parameter set as a label, training the fault identification model, and obtaining the trained fault identification model; Construct a performance model of the system to be identified based on the physical architecture and design parameter model of the system to be identified; Performing sensitivity analysis on the performance model to obtain an identification parameter set; the identification parameter set includes input instructions, oil supply pressure, actuator output speed, actuator output displacement, servo valve spool displacement, actuator zero point coordinate value and sensor gain; The identification parameter set is screened to obtain an observable parameter set and an unobservable parameter set; the observable parameter set includes an input instruction, an oil supply pressure, an actuator output speed, and an actuator output displacement; The unobservable parameter set includes the servo valve core displacement, the actuator zero point coordinate and the sensor gain; Based on the identification parameter set, acquiring input instruction data, oil supply pressure data, actuator output speed data, actuator output displacement data and system output parameter data set of the system to be identified; The system output parameter data set includes system output speed and system output displacement; Input the input command data, oil supply pressure data, actuator output speed data, actuator output displacement data, the system output speed and the system output displacement into the trained fault identification model to obtain a servo valve spool displacement prediction value, an actuator 0 point coordinate value prediction value and a sensor gain prediction value; The predicted value of the 0-point coordinate value of the actuator is judged. If the predicted value of the 0-point coordinate value of the actuator is within the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is normal. If the predicted value of the 0-point coordinate value of the actuator exceeds the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is faulty. The sensor gain prediction value is judged. If the sensor gain prediction value is within the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is normal. If the sensor gain prediction value exceeds the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is faulty. The predicted value of the servo valve spool displacement is judged. If the predicted value of the servo valve spool displacement is within the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is normal. If the predicted value of the servo valve spool displacement exceeds the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is faulty.

2. The flight control servo actuation system fault identification method according to claim 1, characterized in that: The performance model includes an input instruction set, a system output parameter set and a system parameter set; The input instruction set expression is: u={u1,u2,…,u i }; The system output parameter set expression is: y = {y1, y2}; The system parameter set expression is: x′={x′1,x′2,…,x′ k }; Where: u represents the input instruction set, y represents the system output parameter set, x′ represents the system parameter set, i represents the number of dimensions of the input instruction set, k represents the number of dimensions of the system parameter set, and u i represents the i-th input instruction in the input instruction set, y1 represents the system output speed, y2 represents the system output displacement, x′ k Represents the kth system parameter in the system parameter set.

3. The flight control servo actuation system fault identification method according to claim 1, characterized in that: The identification parameter set expression is: x={x1, x2, x3, x4, x5, x6, x7}; The observable parameter set expression is: x o ={x1,x2,x3,x4}; The unobservable parameter set expression is: x uo ={x5,x6,x7}; Where: x represents the identification parameter set, x o represents the observable parameter set, x uo represents the unobservable parameter set, x1 represents the input command, x2 represents the oil supply pressure, x3 represents the actuator output speed, x4 represents the actuator output displacement, x5 represents the actuator 0 point coordinate value, x6 represents the sensor gain, and x7 represents the servo valve core displacement.

4. The flight control servo actuation system fault identification method according to claim 1, characterized in that: The observable parameter data set, the unobservable parameter data set and the system output parameter data set are all Real data.

5. A flight control servo actuation system fault identification system, characterized in that: It includes: A first data acquisition module is used to acquire a historical data set; the historical data set includes a historical observable parameter data set, a historical unobservable parameter set and a historical system output parameter data set; A model training module is used to construct a fault identification model, take the historical observable parameter data set and the historical system output parameter data set as inputs of the fault identification model, take the historical unobservable parameter set as a label, train the fault identification model, and obtain the trained fault identification model; A performance model module is used to construct a performance model of the system to be identified based on the physical architecture and design parameter model of the system to be identified; A sensitivity analysis module, used for performing sensitivity analysis on the performance model to obtain an identification parameter set; the identification parameter set includes input instructions, oil supply pressure, actuator output speed, actuator output displacement, servo valve spool displacement, actuator zero point coordinate value and sensor gain; A parameter screening module, used for screening the identified parameter set to obtain an observable parameter set and an unobservable parameter set; the observable parameter set includes an input instruction, an oil supply pressure, an actuator output speed, and an actuator output displacement; The unobservable parameter set includes the servo valve core displacement, the actuator zero point coordinate and the sensor gain; A second data acquisition module acquires input instruction data, oil supply pressure data, actuator output speed data, actuator output displacement data and system output parameter data set of the system to be identified based on the identification parameter set; The system output parameter data set includes system output speed and system output displacement; A data prediction module, inputting the input command data, the oil supply pressure data, the actuator output speed data, the actuator output displacement data, the system output speed and the system output displacement into the trained fault identification model, and obtaining a servo valve spool displacement prediction value, an actuator 0 point coordinate value prediction value and a sensor gain prediction value; The first judgment module judges the predicted value of the 0-point coordinate value of the actuator. If the predicted value of the 0-point coordinate value of the actuator is within the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is normal; if the predicted value of the 0-point coordinate value of the actuator exceeds the threshold range set by the actuator, the unit in the to-be-identified system corresponding to the 0-point coordinate value of the actuator is faulty; The second judgment module judges the sensor gain prediction value. If the sensor gain prediction value is within the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is normal; if the sensor gain prediction value exceeds the sensor set threshold range, the unit in the to-be-identified system corresponding to the sensor gain is faulty. The third judgment module is used to judge the predicted value of the servo valve spool displacement. If the predicted value of the servo valve spool displacement is within the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is normal. If the predicted value of the servo valve spool displacement exceeds the set threshold range of the servo valve, the unit in the system to be identified corresponding to the servo valve spool is faulty.

6. The flight control servo actuation system fault identification system according to claim 5, characterized in that: The performance model includes an input instruction set, a system output parameter set and a system parameter set; The input instruction set expression is: u={u1,u2,…,u i }; The system output parameter set expression is: y = {y1, y2}; The system parameter set expression is: x′={x′1,x′2,…,x′ k }; Where: u represents the input instruction set, y represents the system output parameter set, x′ represents the system parameter set, i represents the number of dimensions of the input instruction set, k represents the number of dimensions of the system parameter set, and u i represents the i-th input instruction in the input instruction set, y1 represents the system output speed, y2 represents the system output displacement, x′ k Represents the kth system parameter in the system parameter set.

7. The flight control servo actuation system fault identification system according to claim 5, characterized in that: The identification parameter set expression is: x={x1, x2, x3, x4, x5, x6, x7}; The observable parameter set expression is: x o ={x1,x2,x3,x4}; The unobservable parameter set expression is: x uo ={x5,x6,x7}; Where: x represents the identification parameter set, x o represents the observable parameter set, x uo represents the unobservable parameter set, x1 represents the input command, x2 represents the oil supply pressure, x3 represents the actuator output speed, x4 represents the actuator output displacement, x5 represents the actuator 0 point coordinate value, x6 represents the sensor gain, and x7 represents the servo valve core displacement.

8. The flight control servo actuation system fault identification system according to claim 5, characterized in that: The observable parameter data set, the unobservable parameter data set and the system output parameter data set are all Real data.

Images