CN117687379A - Aeroengine control system actuating mechanism fault detection method based on unknown input observer - Google Patents
Aeroengine control system actuating mechanism fault detection method based on unknown input observer Download PDFInfo
- Publication number
- CN117687379A CN117687379A CN202311588165.0A CN202311588165A CN117687379A CN 117687379 A CN117687379 A CN 117687379A CN 202311588165 A CN202311588165 A CN 202311588165A CN 117687379 A CN117687379 A CN 117687379A
- Authority
- CN
- China
- Prior art keywords
- control system
- engine control
- aero
- fault
- observer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007246 mechanism Effects 0.000 title claims abstract description 81
- 238000001514 detection method Methods 0.000 title claims abstract description 47
- 238000002955 isolation Methods 0.000 claims abstract description 49
- 238000000034 method Methods 0.000 claims abstract description 38
- 239000011159 matrix material Substances 0.000 claims description 37
- 239000013598 vector Substances 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 9
- 241000965478 Darksidea epsilon Species 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 description 16
- 238000003745 diagnosis Methods 0.000 description 10
- 238000006073 displacement reaction Methods 0.000 description 9
- 238000013178 mathematical model Methods 0.000 description 5
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 241000287196 Asthenes Species 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0275—Fault isolation and identification, e.g. classify fault; estimate cause or root of failure
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides an aeroengine control system actuating mechanism fault detection method based on an unknown input observer, which comprises the following steps: establishing a fault detection model based on an unknown input observer for an aero-engine control system so as to detect whether an executing mechanism in the aero-engine control system has faults; when determining that the execution mechanism in the aero-engine control system has faults, respectively designing unknown input observers for a plurality of execution mechanisms of the aero-engine control system to form a special unknown input observer group so as to determine the execution mechanism with faults in the aero-engine control system, thereby realizing fault isolation. According to the invention, an unknown input observer is introduced, so that unknown external disturbance and change can be better processed, the dynamic stability of a model is improved, and the fault of an actuating mechanism can be accurately detected; the method can also deal with fault detection of nonlinear and time-varying systems, and is suitable for different types of actuating mechanisms and control systems.
Description
Technical Field
The invention belongs to the technical field of fault diagnosis of an execution mechanism of an aero-engine control system, and particularly relates to a fault detection method of the execution mechanism of the aero-engine control system based on an unknown input observer.
Background
The aeroengine is a pneumatic thermodynamic system with complex structure, strong nonlinearity and wide working range, has high requirements on working performance and reliability, and the control system is a key for guaranteeing stable and efficient operation of the aeroengine in the working range. The full-authority digital electronic control system (FADEC) has the advantages of high control precision, rich control modes and the like, and has the defects of poor reliability, and the failure number of the FADEC system accounts for about 29% of the total failure number of the engine according to statistics, so that the FADEC system is continuously developed towards the directions of comprehensive failure diagnosis, prediction and health management. The FADEC system comprises an electronic controller, an actuating mechanism, a sensor and the like, wherein the actuating mechanism is a key link of information transmission between the electronic controller and the engine, and analysis and research work performed on the electronic controller is very necessary and valuable. Because of the limitations of the structure of the executing mechanism and the aero-engine of the controlled object, the digital simulation technology taking the establishment of the mathematical model of the executing mechanism as the core is the most important research means at present, and the modeling and simulation analysis of the executing mechanism is as if a digital test bed is provided for the executing mechanism.
Because the actuating mechanism works under the severe working environment of high temperature, high pressure, strong vibration and strong electromagnetic for a long time, faults are easy to occur, and the control quality is reduced. The actuating mechanism consists of a series of mechanical hydraulic devices, and when a certain part fails, the chain reaction is easy to cause, so that the performance of an engine system is affected, and even the serious consequences of the death of a robot caused by the damage of the engine are caused. Therefore, in order to ensure the safety and reliability of the aeroengine during the flight, it is very urgent and necessary to develop the fault diagnosis research work on the actuator.
The current fault diagnosis method for the execution mechanism of the aero-engine control system comprises the following steps: acquiring the characteristics of an execution mechanism of the aeroengine through a state estimator, and designing an execution mechanism fault diagnosis system by using an online monitoring and offline diagnosis method; establishing a mathematical model of a closed loop of the engine executing mechanism, and realizing the on-line diagnosis of the fault of the executing mechanism according to the deviation between the model output and the sensor measured value; and taking the driving quantity of the execution mechanism as system input, establishing a fault model under white noise, and analyzing and processing the filtering residual error containing fault information to realize the diagnosis of the fault of the execution mechanism. These fault diagnosis methods have the following drawbacks: because of the complex nonlinear and time-varying nature of aircraft engine control systems, mathematical models of aircraft engine actuators are typically built based on assumptions and simplifications, which may not be fully realistic, thus resulting in a failure to build an accurate mathematical model that accurately reflects the state of the system, poor dynamic stability of the system, failure of the actuators to be accurately detected, and poor adaptability to various types of actuators and control systems.
Disclosure of Invention
The invention aims to solve the defects that the existing fault diagnosis method for the execution mechanism of the aeroengine control system cannot establish an accurate mathematical model, so that the dynamic stability of the system is poor, the adaptability is poor, and the method cannot be applied to various types of execution mechanisms and control systems.
In order to achieve the above purpose, the technical solution provided by the present invention is:
the fault detection method for the execution mechanism of the aeroengine control system based on the unknown input observer is characterized by comprising the following steps of:
step 1, a fault detection model based on an unknown input observer is established for an aero-engine control system to detect whether an executing mechanism in the aero-engine control system breaks down, and the method comprises the following substeps:
step 1.1, an initial model is established for an aero-engine control system:
wherein x is R n 、u∈R m And y.epsilon.R p Respectively representing the state quantity, input quantity and output quantity of the aeroengine control system, wherein n, m and p are respectively the number of the state quantity, the number of the input quantity and the number of the output quantity of the model, and n is less than m and less than p and f a ∈R m Representing failure of the actuator, g (x, t) ∈R n A epsilon R is an uncertainty term of a nonlinear model of the engine n×n 、B∈R n×m 、C∈R p×n And D.epsilon.R p×m A constant matrix is obtained based on a certain steady-state point of the engine;
step 1.2, introducing an integral observer for an aero-engine control system:
step 1.3, a fault detection model is established for an aero-engine control system:
wherein x is a ∈R n+p 、y a ∈R p The state quantity and the output quantity after the integral observer is respectively introduced into the aero-engine control system, d epsilon R r For system interference, r is the number of system interference quantities, g a (x a T) is a new uncertainty item obtained by transforming an uncertainty item g (x, t) of a nonlinear model of the aero-engine, A a ∈R (n+p)×(n+p) 、B a ∈R (n+p)×m 、C a ∈R p×(n+p) New coefficient matrix, E, which is a combination of coefficient matrices A, B, C, D a ∈R (n+p)×r A constant coefficient matrix representing an aeroengine control system interference matrix;
step 1.4, designing an unknown input observer for an aero-engine control system:
wherein z is E R n+p For an unknown state quantity of the input observer,estimating variables for the state of an unknown input observer, N ε R (n+p)×(n+p) 、J∈R (n+p)×m 、F∈R (n+p)×p 、M∈R (n+p)×(n+p) And H.epsilon.R (n+p)×p Are all intermediate matrix variables;
step 1.5, calculating a fault detection residual error of an aero-engine control system:
in the formula e a Representing the state estimation error of the state,
step 1.6, judging whether an executing mechanism in an aero-engine control system fails:
wherein ε represents a threshold value of a fault detection residual;
step 2, when determining that an execution mechanism in the aero-engine control system has faults, respectively designing unknown input observers for a plurality of execution mechanisms of the aero-engine control system to form a special unknown input observer group so as to determine the execution mechanism with faults in the aero-engine control system, thereby realizing fault isolation, and comprising the following substeps:
step 2.1, establishing a fault isolation model for an aero-engine control system:
wherein i represents the serial number of the actuator,for intermediate matrix variables, +.>For matrix B a All column vectors except the ith column vector, +.> Is the vector f a All elements except the i-th row element;
step 2.2, respectively designing unknown input observers for a plurality of execution mechanisms of the aeroengine control system:
wherein z is i ∈R n+p For an unknown state quantity of the input observer,estimating variables, moments for states of unknown input observersArray N i ∈R (n+p)×(n+p) 、J i ∈R (n+p)×m 、F i ∈R (n+p)×p 、M i ∈R (n+p)×(n+p) And H i ∈R (n+p)×p All are intermediate variable matrixes;
step 2.3, calculating fault isolation residual errors of an aeroengine control system:
in the method, in the process of the invention,representing the state estimation error of the i-th observer,/->
Step 2.4, determining a faulty actuating mechanism in the aero-engine control system:
wherein ε i Representing the threshold value of the fault isolation residual.
Further, step 1.3 comprises the sub-steps of:
step 1.3.1, calculating a state space equation according to the established initial model of the aeroengine control system and an integral observer:
step 1.3.2, introducing system interference to obtain a new state space equation:
step 1.3.3, when no actuator in the aero-engine control system fails, f a =0, a fault detection model of the aero-engine control system is obtained.
Further, step 2.1 further comprises the sub-steps of:
step 2.0.1, establishing an initial fault isolation model for an aeroengine control system:
in the method, in the process of the invention,is f a Row vector of row i, +.>For matrix B a Is the ith column vector of (2);
step 2.0.2, when the ith actuator in the aircraft engine control system has not failed,a fault isolation model of the aircraft engine control system is obtained.
Further, ε and ε i The values of (2) are all the same.
The invention has the advantages that:
according to the method for detecting the fault of the execution mechanism of the aeroengine control system based on the unknown input observer, the fault detection of the execution mechanism of the aeroengine control system is divided into two steps of judging whether the execution mechanism breaks down and determining the execution mechanism which breaks down, the unknown input observer is introduced into the system in each step, and whether the execution mechanism breaks down or not and isolating the execution mechanism which breaks down is determined by comparing the residual error with a threshold value. Therefore, the method for detecting the fault of the execution mechanism of the aeroengine control system can better process unknown external disturbance and change by introducing the unknown input observer capable of estimating the state and unknown input of the system, can better cope with various operation and environmental conditions possibly occurring in the aeroengine control system, improves the dynamic stability of a model, ensures that the system can better cope with the change of dynamic performance, can accurately detect the fault of the execution mechanism, and reduces the possibility of false alarm or missing report of the fault of the execution mechanism. In addition, the method based on the unknown input observer can cope with fault detection of nonlinear and time-varying systems, has stronger adaptability, and is suitable for different types of actuating mechanisms and control systems.
Drawings
The features and advantages of the present invention will become more readily understood from the following description with reference to the accompanying drawings, which are not drawn to scale, and some features are exaggerated or reduced to show details of particular components, in which:
FIG. 1 is a flowchart of a method for detecting an aircraft engine control system actuator fault based on an unknown input observer according to an exemplary embodiment of the present invention;
FIG. 2 is a Simulink simulation block diagram of a method for detecting faults of an aircraft engine control system actuator based on an unknown input observer according to an exemplary embodiment of the present invention;
FIG. 3 is a simulation result of fault detection and isolation for an aircraft engine control system without faults;
FIG. 4 is a simulation result of fault detection and isolation when the aircraft engine control system is in fault state 1;
FIG. 5 is a simulation result of fault detection and isolation when the aircraft engine control system is in fault state 2;
FIG. 6 is a simulation result of fault detection and isolation when the aircraft engine control system is in fault state 3;
fig. 7 is a simulation result of fault detection and isolation when the aircraft engine control system is in fault state 4.
Detailed Description
The invention will be described in detail below with the aid of exemplary embodiments of the invention with reference to the accompanying drawings. It should be noted that the following detailed description of the present invention is for illustrative purposes only and is not intended to be limiting.
Referring to fig. 1, an aircraft engine control system actuator fault detection method based on an unknown input observer as an exemplary embodiment of the present invention may include the steps of:
step 1, a fault detection model based on an unknown input observer is established for an aero-engine control system to detect whether an executing mechanism in the aero-engine control system has faults;
and 2, when determining that the execution mechanism in the aero-engine control system fails, respectively designing unknown input observers for a plurality of execution mechanisms of the aero-engine control system to form a special unknown input observer group so as to determine the execution mechanism with the failure in the aero-engine control system, thereby realizing failure isolation.
Step 1 may include step 1.1 of establishing an initial model for the aircraft engine control system:
wherein x is R n 、u∈R m And y.epsilon.R p Respectively representing the state quantity, input quantity and output quantity of the aeroengine control system, wherein n, m and p are respectively the number of the state quantity, the number of the input quantity and the number of the output quantity of the model, and n is less than m and less than p and f a ∈R m Representing failure of the actuator, g (x, t) ∈R n A epsilon R is an uncertainty term of a nonlinear model of the engine n×n 、B∈R n×m 、C∈R p×n And D.epsilon.R p×q A constant matrix is obtained based on a certain steady-state point of the engine;
step 1.2, introducing an integral observer for an aero-engine control system:
step 1.3, a fault detection model is established for an aero-engine control system:
wherein x is a ∈R n+p 、y a ∈R p The state quantity and the output quantity after the integral observer is respectively introduced into the aero-engine control system, d epsilon R r For system interference, r is the number of system interference quantities, g a (x a T) is a new uncertainty item obtained by transforming an uncertainty item g (x, t) of a nonlinear model of the aero-engine, A a ∈R (n+p)×(n+p) 、B a ∈R (n+p)×m 、C a ∈R p×(n+p) A new coefficient matrix formed by combining coefficient matrices A, B, C, D, matrix B is formed due to the full rank of matrix B, D columns a Rank of column full, E a ∈R (n+p)×r Constant coefficient matrix representing an aeroengine control system interference matrix, and matrix E a Rank is listed as full rank.
In some embodiments of the invention, step 1.3 may comprise the sub-steps of:
step 1.3.1, calculating a state space equation according to the established initial model of the aeroengine control system and an integral observer:
step 1.3.2, introducing system interference in consideration of modeling deviation, external disturbance and the like, and obtaining a new state space equation:
the system satisfies the following assumptions:
suppose 1: rank (C) a E a )＝rank(E a )＝r
Suppose 2: the complex variables s with non-negative real parts are all satisfied:
suppose 3: the nonlinear term g (x, t) is in accordance with the lipschitz continuous (Lipschitz continuity) condition with respect to state x, namely:
wherein L is g Is a known Lipschitz constant.
Lemma 1: when hypothesis 3 is satisfied, nonlinear term g a (x a T) about state x a Meets the continuous conditions of Lipohsh, namely:
and (3) proving:
then
If hypothesis 3 is satisfied, then:
and finishing the verification by the primer 1.
Step 1.3.3 when there is no aero-engine control systemFailure of the actuator, f a =0, a fault detection model of the aero-engine control system is obtained.
Step 1 may include step 1.4 of designing an unknown input observer for the aircraft engine control system:
the formula satisfies the above assumptions 1 to 3, wherein z εR n+p For an unknown state quantity of the input observer,estimating variables for the state of an unknown input observer, N ε R (n+p)×(n+p) 、J∈R (n+p)×m 、F∈R (n+p)×p 、M∈R (n+p)×(n+p) And H.epsilon.R (n+p)×p Are intermediate matrix variables.
The matrices N, J, F and M satisfy the following relationship:
M＝I n+p +HC a (14)
N＝MA a -KC a (15)
F＝K(I p +C a H)-MA a H (16)
J＝MB a (17)
in the formula, the matrix K epsilon R (n+p)×p Is an intermediate matrix variable.
Defining a fault detection state estimation error of a control system asThen:
e a ＝x a -z+Hy a ＝(I n+p +HC a )x a -z (18)
the derivative of the state estimation error is:
in the method, in the process of the invention,
according to assumption 1, then there is:
ME a ＝(I n+p +HC a )E a ＝0 (20)
the special solution is as follows:
H * ＝-E a [(C a E a ) T (C a E a )] -1 (C a E a ) T (21)
so that:
theorem 1: for the followingThen e.fwdarw.0 when t.fwdarw.infinity, i.e. the system is progressively stable.
Theorem 2: satisfying the system (1) under assumptions 1-3 and the unknown input observer (13), if the existence matrix N, M and the positive definite matrix P satisfy the equation (23), the state estimation error (18) generated by the unknown input observer is for an arbitrary initial value e a (0) Progressive stabilization.
N T P+PN+L g PMM T P+L g I n+p ＜0 (23)
And (3) proving: selecting lyapunov functionThe combination formula (22) is as follows:
considering the quotation 1 yields:
substituting formula (25) into formula (24):
considering (23),from theorem 1, the system is progressively stable.
Theorem 2 proves complete.
Theorem 3 (Schur's complement theorem): for a given symmetry matrixWherein S is 11 Is a x a dimension. The following three conditions are equivalent:
(i)S＜0
(ii)S 11 ＜0，
(iii)S 22 ＜0，
according to theorem 3, formula (23) may be rewritten as:
let n=ma a -KC a Substituted into (27) and letObtaining:
in summary, the design process of the unknown input observer is summarized as follows:
1) Calculation of
2) Solving the LMI (28) to obtain a matrixAnd calculate +.>
3) The matrix N, F, J is calculated according to equations (15) to (17).
Step 1 may include step 1.5, calculating a fault detection residual of the aircraft engine control system:
as is known from theorem 1, when the system fails,when t → infinity, e a 0, r.fwdarw.0. When the system fails, i.e. f a At +.0, the system is no longer stable.
Step 1 may include step 1.6, determining whether an actuator in the aircraft engine control system has failed:
where ε represents a threshold value of the fault detection residual and may be empirically set by a plurality of experiments.
Therefore, when r < epsilon, judging that no executing mechanism in the aero-engine control system fails, continuing to detect, and when r is more than or equal to epsilon, judging that the executing mechanism in the aero-engine control system fails, and continuing to step 2 to isolate the executing mechanism with failure in the aero-engine control system.
Step 2 may include step 2.1 of building a fault isolation model for an aircraft engine control system:
wherein i represents the serial number of the actuator,for intermediate matrix variables, +.>For matrix B a All column vectors except the ith column vector, +.>To augment interference vector +.>Is the vector f a All elements except the i-th row element represent all actuator fault vectors except the i-th actuator fault.
In an exemplary embodiment, step 2.1 may further comprise the following sub-steps before step 2.1:
step 2.0.1, establishing an initial fault isolation model for an aeroengine control system:
in the method, in the process of the invention,is f a Line vector of the ith line of (c) representing the ith executionRunning gear fault vector, ">For matrix B a Is the ith column vector of (2);
step 2.0.2, when the ith actuator in the aircraft engine control system has not failed,obtaining a fault isolation model of an aeroengine control system:
in particular, the method comprises the steps of,when formula (32) can be described as:
defining an augmented interference vector as:
equation (33) can be rewritten as equation (31).
The following assumptions are now presented:
suppose 4:
(1)rank(C a E f )＝rank(E f ) R+m, where E f ＝[E a B a ]∈R (n+p)×(r+m)
(2) For all complex variables s satisfying Re(s). Gtoreq.0:
and (4) lemma 2: when hypothesis 4 is satisfied, then
(1)
(2) For all complex variables s satisfying Re(s). Gtoreq.0:
the solving method is similar to the above, and will not be described again.
Step 2 may include step 2.2 of designing unknown input observers for a plurality of actuators of the aircraft engine control system, respectively:
the formula satisfies the above assumptions 3 to 4, wherein z i ∈R n+p For an unknown state quantity of the input observer,estimating a variable for the state of an unknown input observer, matrix N i ∈R (n+p)×(n+p) 、J i ∈R (n+p)×m 、F i ∈R (n+p)×p 、M i ∈R (n +p)×(n+p) And H i ∈R (n+p)×p Are intermediate variable matrices.
Step 2 may include step 2.3, calculating a fault isolation residual of the aircraft engine control system:
in the method, in the process of the invention,representing the state estimation error of the i-th observer,/->
As can be seen from theorem 1, when the ith actuator is notIn the event of a failure of the device,when t → infinity>r i And 0. When the ith actuating mechanism fails, namely +.>When the system is no longer stable.
Step 2 may include step 2.4 of determining a failed actuator in the aircraft engine control system:
wherein ε i The threshold value representing the fault isolation residual may be empirically set by a number of trials. In particular embodiments, ε and ε i The values of (2) are all the same.
Therefore, when it is determined that an actuator in the aircraft engine control system has failed in step 1, in step 2, an unknown input observer is designed for each actuator in the aircraft engine control system to form a dedicated unknown input observer group, and whether or not the failure has occurred is determined for each actuator, so that it is determined which actuator has failed, and isolation of the actuator failure is achieved.
As described above, the method for detecting the fault of the actuator of the control system of the aeroengine based on the unknown input observer of the present invention divides the fault detection of the actuator of the control system of the aeroengine into two steps of judging whether there is a fault of the actuator and determining the faulty actuator, and introduces the unknown input observer to the system in each step and determines whether there is a fault of the actuator and isolates the faulty actuator by comparing the residual error with the threshold value. Therefore, the method for detecting the fault of the execution mechanism of the aeroengine control system can better process unknown external disturbance and change by introducing the unknown input observer capable of estimating the state and unknown input of the system, can better cope with various operation and environmental conditions possibly occurring in the aeroengine control system, improves the dynamic stability of a model, ensures that the system can better cope with the change of dynamic performance, can accurately detect the fault of the execution mechanism, and reduces the possibility of false alarm or missing report of the fault of the execution mechanism. In addition, the method based on the unknown input observer can cope with fault detection of nonlinear and time-varying systems, has stronger adaptability, and is suitable for different types of actuating mechanisms and control systems.
The method for detecting the fault of the execution mechanism of the aeroengine control system based on the unknown input observer provided by the invention is further described with reference to fig. 2 to 7 in combination with examples.
The working condition of the engine at the height H=8 km, the Mach number Ma=0.8 and the throttle lever angle PLA=28 is selected as an example, the correlation matrix of the unknown input observer is obtained by solving, and a fault detection and isolation simulation system based on the unknown input observer is designed as shown in fig. 2, wherein the system interference is a Gaussian variable with the mean value of 0 and the variance of 0.002, and the measurement noise is a Gaussian variable with the mean value of 0 and the variance of 0.01.
In the experiment, three fault situations of four execution mechanisms of an aeroengine control system are verified respectively:
first, no actuator fails, and the results of fault detection and isolation simulation are shown in FIG. 3;
second, a single actuator fails: in the fault state 1, when t=5 s, the compressor guide vane rotating mechanism has slope fault, the deviation value transmitted to the engine is increased from 0 to 3mm within 6s, and after t=11 s, the deviation value is kept unchanged by 3mm, and the fault detection and isolation simulation result is shown in fig. 4; and a fault state 2, when t=8s, the tail nozzle displacement sensor bursts the iron core fracture fault, so that a constant deviation fault is generated, the value fed back by the displacement sensor is increased by 0.98mm, and the fault detection and isolation simulation results are shown in fig. 5;
third, multiple actuators fail: in the fault state 3, when t=15s, the fan guide vane actuating cylinder suddenly leaks, and the diameter gap of the internal leakage module is 0.2mm; when t=20s, the tail nozzle displacement sensor bursts the iron core fracture fault, so that the generated constant deviation fault occurs, the value fed back by the displacement sensor is increased by 0.98mm, and the fault detection and isolation simulation results are shown in fig. 6; and a fault state 4, when t=5s, the tail nozzle displacement sensor bursts the iron core fracture fault, so that the generated constant deviation fault occurs, and the value fed back by the displacement sensor is increased by 0.98mm; at t=10s, the differential pressure valve spring is fatigued, the elastic coefficient is reduced from 8.711N/mm to 1.711N/mm within 5s, and after t=15s, the elastic coefficient is kept unchanged at 1.711N/mm; and when t=20s, the constant deviation fault occurs in the compressor guide vane rotating mechanism, the number transmitted to the engine is increased by 3mm, and the fault detection and isolation simulation results are shown in fig. 7.
According to engineering experience and numerical simulation, selecting fault detection and isolation thresholds as follows:
ε＝ε 1 ＝ε 2 ＝ε 3 ＝ε 4 ＝0.025
as can be seen from fig. 3: because of the nonlinear system, the interference of the system and the measurement noise, the residual error has larger fluctuation, but the fault detection and isolation residual error of the control system does not exceed a threshold value all the time, which indicates that no actuating mechanism in the aeroengine control system has faults.
As can be seen from fig. 4: when the slope fault occurs to the guide vane rotating mechanism of the air compressor, the fault detection residual error is gradually increased and exceeds 0.025 in 3s, so that the fault is detected, namely, the fault occurs to an executing mechanism in the control system of the aero-engine. Meanwhile, the fault isolation residual errors of the guide vanes of the air compressor are gradually increased, the fault isolation residual errors exceed 0.025 within 3s, and the fault isolation residual errors corresponding to other actuating mechanisms are not more than 0.025 all the time, which indicates that only the actuating device of the guide vanes of the air compressor breaks down, so that fault isolation is realized.
As can be seen from fig. 5: when the tail pipe displacement sensor suddenly fails, the fault detection residual error suddenly increases and exceeds 0.025, which indicates that the fault is detected, namely that an executing mechanism in an aero-engine control system breaks down. The fault isolation residual error of the tail pipe is suddenly increased and exceeds 0.025, and the isolation residual error corresponding to the other non-faulty actuating mechanisms is not more than 0.025 all the time, which indicates that only the tail pipe actuating device is faulty, thereby realizing fault isolation.
As can be seen from fig. 6: at t=15s, the fan vane actuator suddenly fails to leak, the fault detection residual suddenly increases and exceeds 0.025, indicating that a fault is detected, i.e., that an actuator in the aircraft engine control system is faulty. At the same time, the fan vane fault isolation residual suddenly increases and exceeds 0.025. At t=20s, the nozzle displacement sensor experiences constant deviation faults, and the nozzle fault isolation residuals also suddenly increase and exceed 0.025. The isolation residual errors corresponding to the other non-faulty execution mechanisms are not more than 0.025 all the time, so that the fault isolation of the multiple execution mechanisms is realized.
As can be seen from fig. 7: when t=5s, the tail pipe displacement sensor has constant deviation fault, and the fault detection residual error suddenly increases and exceeds a threshold value, so that the fault is detected, namely, the fault occurs to an executing mechanism in the aeroengine control system. At the same time, the nozzle fault isolation residual also suddenly increases and exceeds the threshold. At t=10s, the differential valve spring experiences a ramp fault, and the main fuel fault isolation residual increases gradually, exceeding 0.025 in 3 s. At t=20s, the compressor vane rotating mechanism suffers from constant deviation faults, and the compressor vane fault isolation residual also increases suddenly and exceeds 0.025. The fan guide vane actuating device has no fault, and the corresponding isolation residual is not more than 0.025 all the time, so that the fault isolation of the multiple actuating mechanisms is realized.
As described, the applicability and the effectiveness of the fault detection method for the execution mechanism of the aeroengine control system based on the unknown input observer are verified through simulation experiments.
The features that are mentioned and/or shown in the above description of exemplary embodiments of the invention may be combined in the same or similar way in one or more other embodiments in combination with or instead of the corresponding features of the other embodiments. Such combined or substituted solutions should also be considered to be included within the scope of the invention.
Claims (4)
1. An aeroengine control system actuator fault detection method based on an unknown input observer is characterized by comprising the following steps:
step 1, a fault detection model based on an unknown input observer is established for an aero-engine control system to detect whether an executing mechanism in the aero-engine control system breaks down, and the method comprises the following substeps:
step 1.1, an initial model is established for an aero-engine control system:
wherein x is R n 、u∈R m And y.epsilon.R p Respectively representing the state quantity, input quantity and output quantity of the aeroengine control system, wherein n, m and p are respectively the number of the state quantity, the number of the input quantity and the number of the output quantity of the model, and n is less than m and less than p and f a ∈R m Representing failure of the actuator, g (x, t) ∈R n A epsilon R is an uncertainty term of a nonlinear model of the engine n×n 、B∈R n×m 、C∈R p×n And D.epsilon.R p×m A constant matrix is obtained based on a certain steady-state point of the engine;
step 1.2, introducing an integral observer for an aero-engine control system:
step 1.3, a fault detection model is established for an aero-engine control system:
wherein x is a ∈R n+p 、y a ∈R p The state quantity and the output quantity after the integral observer is respectively introduced into the aero-engine control system, d epsilon R r For system interference, r is the number of system interference quantities, g a (x a T) is a new uncertainty item obtained by transforming an uncertainty item g (x, t) of a nonlinear model of the aero-engine, A a ∈R (n+p)×(n+p) 、B a ∈R (n+p)×m 、C a ∈R p×(n+p) New coefficient matrix, E, which is a combination of coefficient matrices A, B, C, D a ∈R (n+p)×r A constant coefficient matrix representing an aeroengine control system interference matrix;
step 1.4, designing an unknown input observer for an aero-engine control system:
wherein z is E R n+p For an unknown state quantity of the input observer,estimating variables for the state of an unknown input observer, N ε R (n+p)×(n+p) 、J∈R (n+p)×m 、F∈R (n+p)×p 、M∈R (n+p)×(n+p) And H.epsilon.R (n+p)×p Are all intermediate matrix variables;
step 1.5, calculating a fault detection residual error of an aero-engine control system:
in the formula e a Representing the state estimation error of the state,
step 1.6, judging whether an executing mechanism in an aero-engine control system fails:
wherein ε represents a threshold value of a fault detection residual;
step 2, when determining that an execution mechanism in the aero-engine control system has faults, respectively designing unknown input observers for a plurality of execution mechanisms of the aero-engine control system to form a special unknown input observer group so as to determine the execution mechanism with faults in the aero-engine control system, thereby realizing fault isolation, and comprising the following substeps:
step 2.1, establishing a fault isolation model for an aero-engine control system:
wherein i represents the serial number of the actuator,for intermediate matrix variables, +.>For matrix B a All column vectors except the ith column vector, +.> Is the vector f a All elements except the i-th row element;
step 2.2, respectively designing unknown input observers for a plurality of execution mechanisms of the aeroengine control system:
wherein z is i ∈R n+p For an unknown state quantity of the input observer,estimating a variable for the state of an unknown input observer, matrix N i ∈R (n+p)×(n+p) 、J i ∈R (n+p)×m 、F i ∈R (n+p)×p 、M i ∈R (n+p)×(n+p) And H i ∈R (n+p)×p All are intermediate variable matrixes;
step 2.3, calculating fault isolation residual errors of an aeroengine control system:
in the method, in the process of the invention,representing the state estimation error of the i-th observer,/->
Step 2.4, determining a faulty actuating mechanism in the aero-engine control system:
wherein ε i Representing the threshold value of the fault isolation residual.
2. The method for detecting faults in an aircraft engine control system actuator based on an unknown input observer as claimed in claim 1, wherein step 1.3 comprises the sub-steps of:
step 1.3.1, calculating a state space equation according to the established initial model of the aeroengine control system and an integral observer:
step 1.3.2, introducing system interference to obtain a new state space equation:
step 1.3.3, when no actuator in the aero-engine control system fails, f a =0, a fault detection model of the aero-engine control system is obtained.
3. The method for detecting faults in an aircraft engine control system actuator based on an unknown input observer according to claim 1 or 2, step 2.1 further comprising the sub-steps of:
step 2.0.1, establishing an initial fault isolation model for an aeroengine control system:
in the method, in the process of the invention,is f a Row vector of row i, +.>For matrix B a Is the ith column vector of (2);
step 2.0.2, when navigatingThe ith actuator in the air engine control system has not failed,a fault isolation model of the aircraft engine control system is obtained.
4. The method for detecting faults of an aircraft engine control system actuator based on an unknown input observer according to claim 1 or 2, characterized by: epsilon and epsilon i The values of (2) are all the same.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311282637X | 2023-10-07 | ||
| CN202311282637 | 2023-10-07 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117687379A true CN117687379A (en) | 2024-03-12 |
Family
ID=90134284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311588165.0A Pending CN117687379A (en) | 2023-10-07 | 2023-11-27 | Aeroengine control system actuating mechanism fault detection method based on unknown input observer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117687379A (en) |
-
2023
- 2023-11-27 CN CN202311588165.0A patent/CN117687379A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110118128B (en) | Fault diagnosis and fault-tolerant control method for micro gas turbine sensor | |
| CN109857094B (en) | Two-stage Kalman filtering algorithm-based aeroengine fault diagnosis method | |
| US5919267A (en) | Neural network fault diagnostics systems and related method | |
| RU2667691C1 (en) | Method of fault diagnostics and the fault diagnostic system | |
| US7415328B2 (en) | Hybrid model based fault detection and isolation system | |
| CN111474919B (en) | Aeroengine control system sensor fault diagnosis method based on AANN network group | |
| Xin et al. | Fault diagnosis based on measurement reconstruction of HPT exit pressure for turbofan engine | |
| CN106055770A (en) | Diagnostic method for gas path faults of aero-engine based on sliding mode theory | |
| CN111538316B (en) | Performance-based fault diagnosis method and system for actuating mechanism of closed-loop control system | |
| CN110083076B (en) | Gas turbine pneumatic actuator fault semi-physical simulation platform and simulation method | |
| Luo et al. | Intelligent model-based diagnostics for vehicle health management | |
| Wei et al. | Aircraft engine sensor fault diagnostics based on estimation of engine's health degradation | |
| Liu et al. | A divide and conquer approach to anomaly detection, localization and diagnosis | |
| Lamoureux et al. | An approach to the health monitoring of a pumping unit in an aircraft engine fuel system | |
| Zhao et al. | Hardware-in-the-loop simulation platform for fault diagnosis of rocket engines | |
| CN117687379A (en) | Aeroengine control system actuating mechanism fault detection method based on unknown input observer | |
| CN115407662B (en) | Robust fault detection method based on expansion fault detection observer and member collection estimation | |
| CN111191770B (en) | Rocket system health state assessment method based on fuzzy neural network | |
| Zhao et al. | Fault detection and diagnosis for sensor in an aero-engine system | |
| Zarate et al. | Computation and monitoring of the deviations of gas turbine unmeasured parameters | |
| Li et al. | Multiple fault diagnosis of aeroengine control system based on autoassociative neural network | |
| CN117908500A (en) | Aeroengine control system actuating mechanism fault detection method based on sliding mode observer | |
| Bird et al. | Diagnosis of turbine engine transient performance with model-based parameter estimation techniques | |
| KR20210021857A (en) | Intelligent diagnosing system available of facility and equipment operation by dynamic simulator of marine structure | |
| Xue et al. | Application of Kalman filters for the fault diagnoses of aircraft engine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |