CN112377311A - Robust gain scheduling fault-tolerant controller for input-limited aero-engine - Google Patents

Robust gain scheduling fault-tolerant controller for input-limited aero-engine Download PDF

Info

Publication number
CN112377311A
CN112377311A CN202010544469.7A CN202010544469A CN112377311A CN 112377311 A CN112377311 A CN 112377311A CN 202010544469 A CN202010544469 A CN 202010544469A CN 112377311 A CN112377311 A CN 112377311A
Authority
CN
China
Prior art keywords
engine
fault
input
robust
vector
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
Application number
CN202010544469.7A
Other languages
Chinese (zh)
Inventor
缑林峰
蒋宗霆
刘志丹
孙楚佳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN202010544469.7A priority Critical patent/CN112377311A/en
Publication of CN112377311A publication Critical patent/CN112377311A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F02—COMBUSTION ENGINES; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS
    • F02C—GAS-TURBINE PLANTS; AIR INTAKES FOR JET-PROPULSION PLANTS; CONTROLLING FUEL SUPPLY IN AIR-BREATHING JET-PROPULSION PLANTS
    • F02C9/00—Controlling gas-turbine plants; Controlling fuel supply in air- breathing jet-propulsion plants
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F01—MACHINES OR ENGINES IN GENERAL; ENGINE PLANTS IN GENERAL; STEAM ENGINES
    • F01D—NON-POSITIVE DISPLACEMENT MACHINES OR ENGINES, e.g. STEAM TURBINES
    • F01D21/00—Shutting-down of machines or engines, e.g. in emergency; Regulating, controlling, or safety means not otherwise provided for
    • F01D21/003—Arrangements for testing or measuring
    • 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
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Combustion & Propulsion (AREA)
  • Chemical & Material Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention provides an input-limited robust gain scheduling fault-tolerant controller for an aircraft engine. The robust controller group fault-tolerant control module generates a control vector v and outputs the control vector v to the input limiting module, the input limiting module generates a limited control input vector u and outputs the limited control input vector u to the aeroengine body, and the gas path component fault diagnosis module diagnoses the gas path component fault of the engine, calculates the health parameter h of the engine and outputs the health parameter h to the robust controller group fault-tolerant control module; and the fault-tolerant control module of the robust controller group calculates and obtains an adaptive robust controller by utilizing a plurality of robust controllers designed in the robust controller group according to the input health parameter h and the scheduling parameter alpha. The invention can well control the real engine under the condition of failure of an engine gas path component on the premise of ensuring the safe operation of the engine, has stronger robustness, ensures the safe operation of the engine, gives full play to the performance of the engine and improves the safety and the performance of the airplane.

Description

Robust gain scheduling fault-tolerant controller for input-limited aero-engine
Technical Field
The invention relates to the technical field of aero-engine control, in particular to an input-limited aero-engine robust gain scheduling fault-tolerant controller.
Background
An aircraft engine is a complex nonlinear dynamical system, and when the aircraft engine works in a wide flight envelope, the working state of the engine continuously changes along with the change of external conditions and flight conditions. Aiming at strong nonlinearity of an aircraft engine and uncertainty of a model, a robust gain scheduling control method is provided in the prior art, the engine is divided into a series of working points, a robust controller is designed at each working point, and finally a proper robust controller is selected to control the engine by adopting the gain scheduling method.
The robust gain scheduling control method for the aero-engine can control the aero-engine. However, the requirements of modern warplanes on the performance of aircraft engines are continuously increased, the structures of the aircraft engines are more and more complex, and the engine faults account for 1/3 of the total faults of the aircraft due to the severe and variable operating environments of the engines. Wherein, the gas circuit part failure accounts for more than 90% of the total failure of the engine, and the maintenance cost accounts for 60% of the total maintenance cost of the engine. In order to ensure the safe operation of the engine and to make the failed engine provide sufficient performance to ensure the safe flight of the aircraft or have high maneuverability, the performance of the failed engine must be recovered, and the fault-tolerant control of the engine is performed to ensure the normal and stable operation of the control system and good performance. Therefore, the research on the fault tolerance control method of the gas circuit component of the engine is of great significance.
According to the traditional fault-tolerant control method for the gas circuit component, when the gas circuit component of the aeroengine fails, the control rule is corrected, so that the thrust of the engine is always matched with the throttle lever, and the thrust of the engine is effectively guaranteed. However, these design methods do not address the issue of current controller and engine model mismatches that result in degraded or even unstable control system performance. When the engine has a gas path component fault, the linear model of the engine at the same working point is also changed greatly. Therefore, a controller designed according to an engine model in a normal state generally cannot guarantee the performance of the engine when a gas path component fails, or even cannot guarantee the closed loop stability of a control system.
In addition, excessive control inputs can cause engine damage, and therefore we need to consider the design of controllers with limited control inputs.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an input-limited robust gain scheduling fault-tolerant controller for an aircraft engine, which has stronger robustness, can still well control a real engine under the condition of failure of an engine air path component, ensures the safe work of the engine, gives full play to the performance of the engine, and improves the safety and the performance of the aircraft.
The technical scheme of the invention is as follows:
the robust gain scheduling fault-tolerant controller of the aero-engine with limited input is characterized in that: the system comprises a robust controller group fault-tolerant control module, an input limiting module and a gas circuit component fault diagnosis module;
the robust controller group fault-tolerant control module, the input limit module, the gas path component fault diagnosis module, the aircraft engine body and a plurality of sensors on the aircraft engine form a gas path component fault scheduling control loop;
the robust controller group fault-tolerant control module generates a control vector v and outputs the control vector v to the input limiting module, the input limiting module generates a limited control input vector u and outputs the limited control input vector u to the aeroengine body, and the sensor obtains an aeroengine measurement parameter y; the control input vector u and the measurement parameter y are jointly input into the gas circuit component fault diagnosis module, and the gas circuit component fault diagnosis module diagnoses the fault condition of the gas circuit component of the engine to obtain a health parameter h of the aircraft engine and outputs the health parameter h to the PID controller group fault-tolerant control module;
the robust controller group fault-tolerant control module, the input limit module, the aircraft engine body and a plurality of sensors on the aircraft engine also form a scheduling parameter scheduling control loop; outputting a scheduling parameter alpha to a robust controller group fault-tolerant control module by a sensor; the input limiting module limits the amplitude of the control input vector and avoids damage to the engine caused by excessive control input to the engine;
the robust controller group fault-tolerant control module is internally provided with a plurality of robust controllers which are respectively designed by utilizing a plurality of linear uncertainty engine models, and the linear uncertainty engine models are obtained by linearizing nonlinear models of the aero-engine under different set working points and under different gas path component faults and then adding a pickup block;
the robust controller group fault-tolerant control module utilizes a plurality of robust controllers designed in the robust controller group to calculate and obtain an adaptive robust controller according to an input health parameter h and a scheduling parameter alpha, and the robust controller generates a control input vector u according to a difference e between a reference input r and a measurement parameter y.
Further, the process of designing a plurality of robust controllers in the fault-tolerant control module of the robust controller group is as follows: selecting q working points in a full flight envelope according to a scheduling parameter alpha to linearize an engine nonlinear model containing health parameters to obtain q linearized models containing the health parameters, obtaining 11q linearized models at the positions where the engine has no air path component fault and a specific air path component fault respectively by adjusting the values of the health parameters, adding a camera block to obtain 11q linear uncertain engine models, and designing corresponding robust controllers for the 11q linear uncertain engine models respectively to form a robust controller group.
Further, the gas path component fault diagnosis module comprises a nonlinear onboard engine model and a piecewise linearization Kalman filter;
the nonlinear airborne engine model is an engine nonlinear model with health parameters:
Figure BDA0002540070780000031
y＝g(x,u,h)
wherein
Figure BDA0002540070780000032
In order to control the input vector,
Figure BDA0002540070780000033
in the form of a state vector, the state vector,
Figure BDA0002540070780000034
in order to output the vector, the vector is,
Figure BDA0002540070780000035
for the health parameter vector, f (-) is an n-dimensional differentiable nonlinear vector function representing the system dynamics, and g (-) is an m-dimensional differentiable nonlinear vector function producing the system output; the nonlinear onboard engine model is input into a control input vector u and a health parameter h of the previous period, and the output health steady-state reference value (x) of the nonlinear onboard engine modelaug,NOBEM,yNOBEM) The method comprises the steps of taking the current period as an estimated initial value of a piecewise linearization Kalman filter;
the inputs of the piecewise linearization Kalman filter are a measurement parameter y and a healthy steady-state reference value (x) output by a nonlinear airborne engine modelaug,NOBEM,yNOBEM) According to the formula
Figure BDA0002540070780000036
Calculating to obtain a health parameter h of the engine in the current period; wherein
Figure BDA0002540070780000037
K is the gain of Kalman filtering
Figure BDA0002540070780000038
P is the Ricini equation
Figure BDA0002540070780000039
The solution of (1); coefficient AaugAnd CaugAccording to the formula
Figure BDA00025400707800000310
Determining, and A, C, L, M is an augmented linear state variable model reflecting engine performance degradation obtained by regarding the health parameter h as the control input of the engine and linearizing the nonlinear on-board engine model at a healthy steady-state reference point
Figure BDA0002540070780000041
Coefficient (c):
Figure BDA0002540070780000042
Figure BDA0002540070780000043
w is the system noise, v is the measurement noise, and the corresponding covariance matrices are the diagonal matrices Q and R.
Furthermore, the robust controller group fault-tolerant control module is an adaptive robust controller obtained by interpolating according to the input health parameter h and the scheduling parameter alpha.
Further, the robust controller group fault-tolerant control module selects two adjacent set working points alpha according to the current scheduling parameter alpha of the aero-engineiAnd alphai+1And obtaining two set operating points alphaiAnd alphai+1Controller K for engine without component failureiAnd Ki+1Various typical component failures Δ hbase_jController (2)
Figure BDA0002540070780000044
Δhbase_jThe value of the jth element representing the vector Δ h is Δ hbaseThe value of the other element is 0, i.e. Δ hbase_jIndicating 10 different component failures, e.g. Δ hbase_1Indicates that the fan has failed and the amount of change in fan efficiency is Δ hbase. According to the formula
Figure BDA0002540070780000045
Figure BDA0002540070780000046
Calculating to obtain the selected working point alpha of the aeroengineiAnd alphai+1Robust controller K under current component fault degree (health parameter h) of engineiAnd Ki+1(wherein. DELTA.hjIs the jth element of the vector Δ h; only if the | | delta h | | | is less than or equal to | | | delta hmaxFault condition of engine gas path component, when | | | delta h | | non-woven hair>||ΔhmaxThe engine has failed); according to the formula
Figure BDA0002540070780000047
And calculating to obtain the current adaptive fault-tolerant robust controller K (alpha) of the aero-engine. Further, the input limiting module adopts a multidimensional rectangular saturation function, and controls an input vector u to be:
Figure BDA0002540070780000051
Figure BDA0002540070780000052
wherein v is1And vmTo control the elements of a vector v, v1，maxAnd vm,maxIs the clipping value of the corresponding element of the control vector v.
Further, the scheduling parameter α includes a fan rotation speed or a compressor rotation speed of the aircraft engine.
Further, the measurement parameters include the temperature and pressure at the outlet of the air inlet, the outlet of the fan, the outlet of the air compressor, the rear of the high-pressure turbine and the rear of the low-pressure turbine, the rotating speed of the fan and the rotating speed of the air compressor.
Advantageous effects
Compared with the prior art, the robust gain scheduling fault-tolerant controller of the aero-engine with limited input utilizes the inherent modules in the traditional gain scheduling controller, improves the fault-tolerant control module of the robust controller group by additionally arranging a fault diagnosis module of gas path components, and additionally arranges a plurality of groups of robust controllers under the condition of different gas path component faults of the engine. The gas circuit component fault diagnosis module realizes accurate judgment of gas circuit component faults through reliable estimation of health parameters, and further combines the traditional scheduling parameters, on the premise of ensuring safe work of an engine, gain scheduling control of the engine when the gas circuit component faults is realized, the robust performance is high, the engine can still work safely when the gas circuit component faults occur, the control precision of the gain scheduling of the engine when the gas circuit component faults occur is improved to the maximum extent, the transition time of a control system is shortened, and the dynamic deviation and the static deviation of the control system are reduced. The nonlinear controlled system is controlled by the controller, so that the system can obtain ideal dynamic and static control quality in the whole working range.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a block diagram of an input-limited robust gain scheduling fault-tolerant controller for an aircraft engine according to the present invention;
fig. 2 is a schematic structural diagram of a fault diagnosis module of the gas circuit component in the gas circuit component fault scheduling control circuit according to the embodiment;
fig. 3 is a schematic structural diagram of a kalman filter in the fault diagnosis module of the gas path component according to the embodiment;
FIG. 4 is a schematic representation of a non-linear engine model of the present invention.
Detailed Description
The performance of gas circuit components can be degraded due to factors such as natural wear, corrosion, scale deposit, thermal creep and the like in the operation process of the aero-engine, and faults can be caused when the performance is degraded to a certain degree; in addition, the gas path member may also be damaged by foreign matter inhalation, mechanical fatigue fracture, or the like. The former failure occurs slowly, while the latter failure occurs rapidly. When the air path component of the engine fails and does not fail, part of the performance of the engine at the moment can seriously deviate from the rated state. Taking a turbine part as an example, when the turbine part fails, the working efficiency of the turbine part will be reduced, that is, the capability of converting the fuel gas with high temperature and high pressure into mechanical energy will be reduced, and corresponding power can be provided for a fan or a compressor part to enable the turbine part to work in a new balance state. At this time, the engine also deviates greatly from the original state. The failure of the gas circuit component can cause that a nonlinear model established during the design of the engine is seriously mismatched with a real engine during the failure of the gas circuit component, so that a gain scheduling controller designed according to the nonlinear model can not well control the engine with the failed gas circuit component, the performance of the engine is seriously reduced, the stability of a control system can not be even ensured, and the safe operation of the engine can not be ensured. The analytical study procedure of the present invention is given below in view of this problem.
1. Engine gas path component fault diagnosis
The failure of the gas path component can cause the corresponding characteristic parameter of the component to change. The engine gas circuit component faults are finally characterized on the changes of the working efficiency and the flow rate of different rotor components, namely the engine fault position and the fault degree can be revealed from the changes of the efficiency coefficients or the flow rate coefficients of the wind fan, the compressor, the main combustion, the high-pressure turbine and the low-pressure turbine components, and the efficiency coefficients or the flow rate coefficients of the fan, the compressor, the main combustion chamber, the high-pressure turbine and the low-pressure turbine components are called as health parameters.
Establishing engine nonlinear model with health parameters based on component method
Figure BDA0002540070780000071
y＝g(x,u,h)
Wherein
Figure BDA0002540070780000072
In order to control the input vector,
Figure BDA0002540070780000073
in the form of a state vector, the state vector,
Figure BDA0002540070780000074
in order to output the vector, the vector is,
Figure BDA0002540070780000075
for the health parameter vector, f (-) is an n-dimensional differentiable nonlinear vector function representing the system dynamics, and g (-) is an m-dimensional differentiable nonlinear vector function producing the system output.
And (3) regarding the health parameter h as the control input of the engine, and linearizing the nonlinear model of the engine at a healthy steady-state reference point by adopting a small perturbation method or a fitting method.
Figure BDA0002540070780000076
Wherein
A′＝A,B′＝(B L),C′＝C,
D′＝(D M),Δu′＝(Δu Δh)T
w is system noise, v is measurement noise, h is a health parameter, Δ h ═ h-h0(ii) a W and v are uncorrelated white gaussian noise, the mean value is 0, and the covariance matrix is diagonal matrices Q and R, which satisfies the following conditions:
E(w)＝0 E[wwT]＝Q
E(v)＝0 E[vvT]＝R
Δ represents the amount of change of the parameter, h0Representing an engine initial state health parameter.
Further obtains an augmented linear state variable model reflecting the performance degradation of the engine
Figure BDA0002540070780000077
Wherein the coefficient matrix is obtained by:
Figure BDA0002540070780000078
Figure BDA0002540070780000079
these coefficients have different values at different operating states of the engine.
In fact, the health parameters are difficult or even impossible to measure, and the pressure, temperature, speed, etc. of each part of the engine are easy to obtain by measurement, and are generally called "measurement parameters", mainly including the temperature and pressure at the outlet of the air inlet, the outlet of the fan, the outlet of the compressor, the temperature and pressure after the high-pressure turbine and the low-pressure turbine, the speed of the fan and the speed of the compressor. When the working environment of the engine does not change, the change of the health parameter can cause the corresponding change of the measured parameter, and an aerodynamic thermodynamic relation exists between the health parameter and the measured parameter. Thus, an optimal estimation filter can be designed to achieve optimal estimation of the health parameter by measuring the parameter.
For a graded component failure, the corresponding failed component health parameter changes slowly, so over the time period in which a single failure diagnosis is performed, it can be considered that the requirements are met
Figure BDA0002540070780000081
For the mutant component failure, the severity of the component failure is more concerned when the engine works stably again after the failure occurs, and the health parameter change of the failed component is still satisfied after the engine works stably again
Figure BDA0002540070780000082
Further converting the health parameters into state variables to obtain
Figure BDA0002540070780000083
Wherein
Figure BDA0002540070780000084
Figure BDA0002540070780000085
The established gas path component fault diagnosis module mainly comprises two parts, wherein one part is a nonlinear airborne engine model based on health parameters, and the other part is a piecewise linear Kalman filter. The basic working principle is that the output of the nonlinear airborne engine model is used as a steady-state reference value of the piecewise linear Kalman filter, health parameters are expanded, online real-time estimation is carried out through the piecewise linear Kalman filter, and finally the online real-time update is fed back to the nonlinear airborne engine model, so that the real-time tracking of an actual engine is realized.
The kalman estimation equation is:
Figure BDA0002540070780000086
k is the gain of Kalman filtering
Figure BDA0002540070780000087
P is the Ricini equation
Figure BDA0002540070780000088
The solution of (1); healthy steady-state reference value (x) output by using nonlinear airborne modelaug,NOBEM,yNOBEM) As formula
Figure BDA0002540070780000091
The initial value of (a) can be obtained by the following calculation formula:
Figure BDA0002540070780000092
the health parameter h of the engine can be obtained according to the calculation formula, and the fault diagnosis of the gas circuit component of the engine is realized.
2. Robust controller design with uncertain model of health parameters
Uncertainty inevitably exists in any practical system, and can be divided into two categories, disturbance signal and model uncertainty. The disturbing signal includes interference, noise, and the like. The uncertainty of the model represents the difference between the mathematical model and the actual object.
Model uncertainty may have several reasons, some parameters in the linear model are always in error; parameters in the linear model may change due to non-linearity or changes in operating conditions; artificial simplification during modeling; degradation of engine performance due to wear and the like.
The uncertainty may adversely affect the stability and performance of the control system.
The error between the actual engine and the nominal model (which is a conventional non-linear model of the engine without healthy parameters) can be expressed as a shot block Δ. Adding a camera block into a nominal model to establish an uncertain model of an engine
Figure BDA0002540070780000093
Figure BDA0002540070780000094
It can also be represented as
G(s)＝[I+Δ(s)]Gnom(s)
And finally, designing the robust controller by using a traditional robust controller design method according to the uncertain model.
3. Gain scheduling fault tolerant control design
The essence of gain scheduling control is to design a set of linearized controllers, which are then regularly combined to be able to control a non-linear system. The basic principle of the gain scheduling fault-tolerant control is to select a series of working points, obtain engine linearization models under different set working points and different gas circuit component faults, and design corresponding robust controllers respectively to obtain the robust controller group in fig. 1.
Referring to FIG. 4, a set of scheduling parameter values α is selectedi1, 2.. q, representing the dynamic range of the system, and dividing the flight envelope into several subintervals and using these points as operating points. At the operating point, there are these equations
Figure BDA0002540070780000101
Figure BDA0002540070780000102
Wherein
Figure BDA0002540070780000103
For the selected i-th operating point, udiTo be at the moment of time
Figure BDA0002540070780000104
Steady state control input required to maintain equilibrium, hdiIs a time of day
Figure BDA0002540070780000105
The health parameter of (1).
By using a small disturbance method, a linear model of the health parameters of each working condition point can be obtained, and a linear model of the engine in a normal state and a performance degradation h state is obtained.
Referring to fig. 4, the upper and lower solid lines represent non-linear models of no-air path component failure and air path component failure h, respectively, of the engine. A series of small black dots represent different working points of the engine, and linearization is carried out at each working point to obtain a linear model. Aiming at linear models of an engine in a normal state and different gas path component fault states, a series of robust controllers are respectively designed to obtain the robust controller group in the graph 1. The controller gain is then linearly interpolated between the selected operating points so that the closed loop system is stable and has good performance for all fixed parameter values. The parameter α is a scheduling parameter, which may be defined herein as a fan speed or a compressor speed of the aircraft engine, and may be measured in real time. Another scheduling variable of the control system is a health parameter h reflecting the degree of failure of engine gas path components. The working principle is that the robust controller group fault-tolerant control module in fig. 1 performs linear interpolation according to the scheduling parameter and the health parameter to obtain a corresponding robust controller to control the system.
4. Interpolation of controller
This section illustrates the scheduling calculation principle of the robust controller group fault-tolerant control module in fig. 1 that obtains the corresponding robust controller through scheduling parameter and health parameter scheduling linear interpolation.
Respectively in the normal state of the engine and various typical component faults delta hbase_jDesigning a series of linear robust controllers under the state, and selecting each working point alphaiAnd (5) controlling. This will result in the controller in the fault tolerant control module of the robust controller group of FIG. 1
Figure BDA0002540070780000106
And then interpolating the controller according to the scheduling parameter alpha and the health parameter h, and then controlling the system by using the obtained interpolated controller.
Two adjacent peripheral working points alpha are selected according to the current scheduling parameter alpha of the engineiAnd alphai+1And obtaining two set operating points alphaiAnd alphai+1Controller K for engine without component failureiAnd Ki+1Various typical component failures Δ hbase_jController (2)
Figure BDA0002540070780000111
Δhbase_jThe value of the jth element representing the vector Δ h is Δ hbaseThe value of the other element is 0, i.e. Δ hbase_jIndicating 10 different component failures, e.g. Δ hbase_1Indicates that the fan has failed and the amount of change in fan efficiency is Δ hbase. The working point alpha can be obtained by linear interpolationiController for gas circuit component fault h
Figure BDA0002540070780000112
Likewise, the operating point α can be obtainedi+1Controller for gas circuit component fault h
Figure BDA0002540070780000113
We use the piecewise linear interpolation method, from robust controller set K1,K2,...,KqLinear interpolation is performed between each pair of controllers. A linear interpolation controller K (α) at the current degradation degree h of the current scheduling parameter α is obtained, i is 1,2
Figure BDA0002540070780000114
According to the formula, a corresponding controller under the condition that a certain air path component has a fault at a certain working point can be obtained, and the engine is effectively controlled.
5. Input restriction of a system
Referring to fig. 1, the input limit module in fig. 1 uses a multidimensional rectangular saturation function to model the physical limit on the control input of the system. Limiting inputs to aircraft engine controls, particularly for fuel flow inputs. The multidimensional saturation function may also handle other control input constraints including the throat area of the aft nozzle. The function is a multi-dimensional rectangular saturation function defined as
Figure BDA0002540070780000121
Wherein v is1And vmTo control the elements of a vector v, v1，maxAnd vm,maxIs the clipping value of the corresponding element of the control vector v. For all
Figure BDA0002540070780000122
The following formula gives sat (·)
Figure BDA0002540070780000123
Based on the above process, the robust gain scheduling fault-tolerant controller for an aero-engine with limited input proposed in the embodiment is given below, and as shown in fig. 1, the robust gain scheduling fault-tolerant controller mainly includes a robust controller group fault-tolerant control module and a gas path component fault diagnosis module.
The robust controller group fault-tolerant control module, the input limit module, the gas path component fault diagnosis module, the aircraft engine body and a plurality of sensors on the aircraft engine form a gas path component fault scheduling control loop 10.
The robust controller group fault-tolerant control module generates a control vector v and outputs the control vector v to the input limiting module, the input limiting module generates a limited control input vector u and outputs the limited control input vector u to the aeroengine body, and the sensor obtains an aeroengine measurement parameter y; the control input vector u and the measurement parameter y are jointly input into the gas circuit component fault diagnosis module, the gas circuit component fault diagnosis module resolves to obtain a health parameter h of the aircraft engine, and outputs the health parameter h to the robust controller group fault-tolerant control module.
The robust controller group fault-tolerant control module, the input limit module, the aircraft engine body and a plurality of sensors on the aircraft engine also form a scheduling parameter scheduling control loop 20; and outputting the scheduling parameter alpha to the robust controller group fault-tolerant control module by the sensor.
The input limit module limits the magnitude of the control input vector to avoid engine damage caused by excessive control input to the engine.
In a preferred specific implementation manner, the input limiting module adopts a multidimensional rectangular saturation function, and controls an input vector u to be:
Figure BDA0002540070780000131
Figure BDA0002540070780000132
wherein v is1And vmTo control the elements of a vector v, v1，maxAnd vm,maxIs the clipping value of the corresponding element of the control vector v.
The robust controller group fault-tolerant control module is internally designed with a plurality of robust controllers which are respectively designed by utilizing a plurality of linearization models, and the linearization models are obtained by linearizing nonlinear models of the aero-engine under different set working points and different gas path component faults of the aero-engine.
In a preferred embodiment, several robust controllers can be designed by the following process: selecting q working points in the full flight envelope according to the scheduling parameter alpha to linearize the engine nonlinear model containing the health parameters to obtain q linearized models containing the health parameters, adjusting the values of the health parameters to obtain 11q linearized models at the positions where the engine has no air path component fault and a specific air path component fault, and designing corresponding robust controllers for the 11q linearized models respectively to form a robust controller group.
The robust controller group fault-tolerant control module utilizes a plurality of robust controllers designed in the robust controller group to calculate and obtain an adaptive robust controller according to an input health parameter h and a scheduling parameter alpha, and the robust controller generates a control input vector u according to a difference e between a reference input r and a measurement parameter y.
In a preferred embodiment, the adaptive robust controller can be obtained by interpolating according to the input health parameter h and the scheduling parameter α:
firstly, two adjacent set working points alpha are selected according to the current scheduling parameter alpha of the aeroengineiAnd alphai+1And obtaining two set operating points alphaiAnd alphai+1Controller K for engine without component failureiAnd Ki+1Various typical component failures Δ hbase_jController (2)
Figure BDA0002540070780000133
Δhbase_jThe value of the jth element representing the vector Δ h is Δ hbaseThe value of the other element is 0, i.e. Δ hbase_jIndicating 10 different component failures, e.g. Δ hbase_1Indicates that the fan has failed and the amount of change in fan efficiency is Δ hbase. According to the formula
Figure BDA0002540070780000141
Figure BDA0002540070780000142
Calculating to obtain the selected working point alpha of the aeroengineiAnd alphai+1Robust controller K under current component fault degree (health parameter h) of engineiAnd Ki+1(wherein. DELTA.hjIs the jth element of the vector Δ h; only if the | | delta h | | | is less than or equal to | | | delta hmaxFault condition of engine gas path component, when | | | delta h | | non-woven hair>||ΔhmaxThe engine has failed); according to the formula
Figure BDA0002540070780000143
And calculating to obtain the current adaptive fault-tolerant robust controller K (alpha) of the aero-engine.
The gas circuit component fault diagnosis module comprises a nonlinear onboard engine model and a piecewise linearization Kalman filter.
The nonlinear airborne engine model is an engine nonlinear model with health parameters:
Figure BDA0002540070780000144
y＝g(x,u,h)
wherein
Figure BDA0002540070780000145
In order to control the input vector,
Figure BDA0002540070780000146
in the form of a state vector, the state vector,
Figure BDA0002540070780000147
in order to output the vector, the vector is,
Figure BDA0002540070780000148
for the health parameter vector, f (-) is an n-dimensional differentiable nonlinear vector function representing the system dynamics, and g (-) is an m-dimensional differentiable nonlinear vector function producing the system output; the nonlinear onboard engine model is input into a control input vector u and a health parameter h of the previous period, and the output health steady-state reference value (x) of the nonlinear onboard engine modelaug,NOBEM,yNOBEM) As the estimated initial value of the current period of the piecewise linearization Kalman filter.
The inputs of the piecewise linearization Kalman filter are a measurement parameter y and a healthy steady-state reference value (x) output by a nonlinear airborne engine modelaug,NOBEM,yNOBEM) According to the formula
Figure BDA0002540070780000149
And calculating to obtain the health parameter h of the engine in the current period.
Wherein
Figure BDA00025400707800001410
K is the gain of Kalman filtering
Figure BDA00025400707800001411
P is the Ricini equation
Figure BDA0002540070780000151
The solution of (1); coefficient AaugAnd CaugAccording to the formula
Figure BDA0002540070780000152
Determining, and A, C, L, M is an augmented linear state variable model reflecting engine performance degradation obtained by regarding the health parameter h as the control input of the engine and linearizing the nonlinear on-board engine model at a healthy steady-state reference point
Figure BDA0002540070780000153
Coefficient (c):
Figure BDA0002540070780000154
Figure BDA0002540070780000155
w is the system noise, v is the measurement noise, and the corresponding covariance matrices are the diagonal matrices Q and R.
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, and that variations, modifications, substitutions and alterations can be made in the above embodiments by those of ordinary skill in the art without departing from the principle and spirit of the present invention.

Claims (8)

1. An input-limited robust gain scheduling fault-tolerant controller for an aircraft engine, characterized by: the system comprises a robust controller group fault-tolerant control module, an input limiting module and a gas circuit component fault diagnosis module;
the robust controller group fault-tolerant control module, the input limit module, the gas path component fault diagnosis module, the aircraft engine body and a plurality of sensors on the aircraft engine form a gas path component fault scheduling control loop;
the robust controller group fault-tolerant control module generates a control vector v and outputs the control vector v to the input limiting module, the input limiting module generates a limited control input vector u and outputs the limited control input vector u to the aeroengine body, and the sensor obtains an aeroengine measurement parameter y; the control input vector u and the measurement parameter y are jointly input into the gas path component fault diagnosis module, the gas path component fault diagnosis module resolves to obtain a health parameter h of the aircraft engine and outputs the health parameter h to the robust controller group fault-tolerant control module;
the robust controller group fault-tolerant control module, the input limit module, the aircraft engine body and a plurality of sensors on the aircraft engine also form a scheduling parameter scheduling control loop; outputting a scheduling parameter alpha to a robust controller group fault-tolerant control module by a sensor;
the input limiting module limits the amplitude of the control input vector and avoids damage to the engine caused by excessive control input to the engine;
the robust controller group fault-tolerant control module is internally provided with a plurality of robust controllers which are respectively designed by utilizing a plurality of linear uncertainty engine models, and the linear uncertainty engine models are obtained by linearizing nonlinear models of the aero-engine under different set working points and under different gas path component faults and then adding a pickup block;
the robust controller group fault-tolerant control module utilizes a plurality of robust controllers designed in the robust controller group to calculate and obtain an adaptive robust controller according to an input health parameter h and a scheduling parameter alpha, and the robust controller generates a control input vector u according to a difference e between a reference input r and a measurement parameter y.
2. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: the process of designing a plurality of robust controllers in the robust controller group fault-tolerant control module is as follows: selecting q working points in a full flight envelope according to a scheduling parameter alpha to linearize an engine nonlinear model containing health parameters to obtain q linearized models containing the health parameters, obtaining 11q linearized models at the positions where the engine has no air path component fault and a specific air path component fault respectively by adjusting the values of the health parameters, adding a camera block to obtain 11q linear uncertain engine models, and designing corresponding robust controllers for the 11q linear uncertain engine models respectively to form a robust controller group.
3. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: and the robust controller group fault-tolerant control module obtains an adaptive robust controller according to the input health parameter h and the scheduling parameter alpha by interpolation.
4. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: the robust controller group fault-tolerant control module selects two set working points alpha which are adjacent to each other in front and back according to the current scheduling parameter alpha of the aero-engineiAnd alphai+1And obtaining two set operating points alphaiAnd alphai+1Controller K for engine without component failureiAnd Ki+1Various typical component failures Δ hbase_jController (2)
Figure RE-FDA0002899195580000021
Δhbase_jThe value of the jth element representing the vector Δ h is Δ hbaseThe value of the other element is 0, i.e. Δ hbase_jThere are 10 different component failures indicated,for example,. DELTA.hbase_1Indicates that the fan has failed and the amount of change in fan efficiency is Δ hbase. According to the formula
Figure RE-FDA0002899195580000022
Figure RE-FDA0002899195580000023
Calculating to obtain the selected working point alpha of the aeroengineiAnd alphai+1Robust controller K under current component fault degree (health parameter h) of engineiAnd Ki+1(wherein. DELTA.hjIs the jth element of the vector Δ h; only if the | | delta h | | | is less than or equal to | | | delta hmaxFault condition of engine gas path component, when | | | delta h | | non-woven hair>||ΔhmaxThe engine has failed); according to the formula
Figure RE-FDA0002899195580000024
And calculating to obtain the current adaptive fault-tolerant robust controller K (alpha) of the aero-engine.
5. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: the gas circuit component fault diagnosis module comprises a nonlinear onboard engine model and a piecewise linearization Kalman filter;
the nonlinear airborne engine model is an engine nonlinear model with health parameters:
Figure RE-FDA0002899195580000031
y＝g(x,u,h)
wherein
Figure RE-FDA0002899195580000032
In order to control the input vector,
Figure RE-FDA0002899195580000033
in the form of a state vector, the state vector,
Figure RE-FDA0002899195580000034
in order to output the vector, the vector is,
Figure RE-FDA0002899195580000035
for the health parameter vector, f (-) is an n-dimensional differentiable nonlinear vector function representing the system dynamics, and g (-) is an m-dimensional differentiable nonlinear vector function producing the system output; the nonlinear onboard engine model is input into a control input vector u and a health parameter h of the previous period, and the output health steady-state reference value (x) of the nonlinear onboard engine modelaug,NOBEM,yNOBEM) The method comprises the steps of taking the current period as an estimated initial value of a piecewise linearization Kalman filter;
the inputs of the piecewise linearization Kalman filter are a measurement parameter y and a healthy steady-state reference value (x) output by a nonlinear airborne engine modelaug,NOBEM,yNOBEM) According to the formula
Figure RE-FDA0002899195580000036
Calculating to obtain a health parameter h of the engine in the current period; wherein
Figure RE-FDA0002899195580000037
K is the gain of Kalman filtering
Figure RE-FDA0002899195580000038
P is the Ricini equation
Figure RE-FDA0002899195580000039
The solution of (1); coefficient AaugAnd CaugAccording to the formula
Figure RE-FDA00028991955800000310
Determining, and A, C, L, M is an augmented linear state variable model reflecting engine performance degradation obtained by regarding the health parameter h as the control input of the engine and linearizing the nonlinear on-board engine model at a healthy steady-state reference point
Figure RE-FDA00028991955800000311
Coefficient (c):
Figure RE-FDA0002899195580000041
Figure RE-FDA0002899195580000042
w is the system noise, v is the measurement noise, and the corresponding covariance matrices are the diagonal matrices Q and R.
6. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: the input limiting module adopts a multidimensional rectangular saturation function, and controls an input vector u to be:
Figure RE-FDA0002899195580000043
Figure RE-FDA0002899195580000044
wherein v is1And vmTo control the elements of a vector v, v1，maxAnd vm,maxIs the clipping value of the corresponding element of the control vector v.
7. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: the scheduling parameter alpha comprises the fan rotating speed or the compressor rotating speed of the aircraft engine.
8. The input-limited aero-engine robust gain scheduling fault tolerant controller of claim 1 wherein: the measurement parameters comprise the temperature and pressure of an air inlet outlet, a fan outlet, a gas compressor outlet, a high-pressure turbine rear part and a low-pressure turbine rear part, the fan rotating speed and the gas compressor rotating speed.
CN202010544469.7A 2020-06-15 2020-06-15 Robust gain scheduling fault-tolerant controller for input-limited aero-engine Pending CN112377311A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010544469.7A CN112377311A (en) 2020-06-15 2020-06-15 Robust gain scheduling fault-tolerant controller for input-limited aero-engine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010544469.7A CN112377311A (en) 2020-06-15 2020-06-15 Robust gain scheduling fault-tolerant controller for input-limited aero-engine

Publications (1)

Publication Number Publication Date
CN112377311A true CN112377311A (en) 2021-02-19

Family

ID=74586337

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010544469.7A Pending CN112377311A (en) 2020-06-15 2020-06-15 Robust gain scheduling fault-tolerant controller for input-limited aero-engine

Country Status (1)

Country Link
CN (1) CN112377311A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040123600A1 (en) * 2002-11-13 2004-07-01 Brunell Brent Jerome Adaptive model-based control systems and methods for controlling a gas turbine
US20040267424A1 (en) * 2003-06-24 2004-12-30 Visteon Global Technologies, Inc. System and method of robust fault detection for a vehicle steer-by-wire system
EP1538319A1 (en) * 2003-12-05 2005-06-08 General Electric Company Apparatus for model predictive control of aircraft gas turbine engines
CN111271181A (en) * 2020-04-04 2020-06-12 西北工业大学 Two-degree-of-freedom [ mu ] controller for conservative gain reduction scheduling of aero-engine

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040123600A1 (en) * 2002-11-13 2004-07-01 Brunell Brent Jerome Adaptive model-based control systems and methods for controlling a gas turbine
US20040267424A1 (en) * 2003-06-24 2004-12-30 Visteon Global Technologies, Inc. System and method of robust fault detection for a vehicle steer-by-wire system
EP1538319A1 (en) * 2003-12-05 2005-06-08 General Electric Company Apparatus for model predictive control of aircraft gas turbine engines
CN111271181A (en) * 2020-04-04 2020-06-12 西北工业大学 Two-degree-of-freedom [ mu ] controller for conservative gain reduction scheduling of aero-engine

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MEHRDAD PAKMEHR .ETAL: "Gain Scheduled Control of Gas Turbine Engines: Stability and Verification", 《JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER》 *

Similar Documents

Publication Publication Date Title
CN111271181B (en) Two-degree-of-freedom [ mu ] controller for conservative gain reduction scheduling of aero-engine
CN111859555A (en) Robust fault-tolerant controller for maximum thrust state of input-limited aircraft engine
CN111608808A (en) Input-limited aeroengine gain scheduling fault-tolerant controller
CN111273554B (en) Two-degree-of-freedom H-infinity controller for conservative state reduction of maximum thrust of aircraft engine
CN108829928B (en) Turboshaft engine adaptive component-level simulation model construction method
CN111880403A (en) Fault-tolerant two-degree-of-freedom [ mu ] controller for maximum thrust state of aircraft engine
CN111856919A (en) Fault-tolerant controller for gain scheduling of failure of gas path component of aero-engine
US8849542B2 (en) Real time linearization of a component-level gas turbine engine model for model-based control
CN110502840B (en) Online prediction method for gas circuit parameters of aero-engine
CN112284752A (en) Variable cycle engine resolution redundancy estimation method based on improved state tracking filter
CN107357176A (en) A kind of aeroengine test run Data Modeling Method
CN111856929B (en) Two-degree-of-freedom H-infinity controller for fault-tolerant gain scheduling of aero-engine
CN111305954A (en) Input-limited aero-engine conservative robust gain reduction scheduling controller
CN111830827B (en) Two-degree-of-freedom [ mu ] controller for fault-tolerant gain scheduling of aero-engine
CN112346336A (en) Robust gain scheduling fault-tolerant controller for failure of aero-engine gas path component
CN112327602A (en) Variable cycle engine gas path component fault gain scheduling fault-tolerant controller
CN112947064A (en) Aero-engine maximum thrust control optimization method considering gas circuit component faults
CN111852662A (en) Fault-tolerant two-degree-of-freedom H-infinity controller for maximum thrust state of aircraft engine
CN111852663A (en) Conservative robust gain reduction scheduling controller for variable cycle engine
CN112377311A (en) Robust gain scheduling fault-tolerant controller for input-limited aero-engine
CN111456856B (en) Robust controller for reducing conservative maximum thrust state of aero-engine
CN111459028B (en) Conservative two-degree-of-freedom mu controller for reducing maximum thrust state of aero-engine
CN111456857B (en) Two-degree-of-freedom H-infinity controller for conservative gain reduction scheduling of aero-engine
CN110985216A (en) Intelligent multivariable control method for aero-engine with online correction
CN114047692B (en) Turbofan engine robust fault-tolerant anti-interference model reference dynamic output feedback control method

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20210219

WD01 Invention patent application deemed withdrawn after publication