CN114142782A - Fault estimation and compensation method for asynchronous motor actuator - Google Patents

Fault estimation and compensation method for asynchronous motor actuator Download PDF

Info

Publication number
CN114142782A
CN114142782A CN202111176407.6A CN202111176407A CN114142782A CN 114142782 A CN114142782 A CN 114142782A CN 202111176407 A CN202111176407 A CN 202111176407A CN 114142782 A CN114142782 A CN 114142782A
Authority
CN
China
Prior art keywords
fault
asynchronous
actuator
mathematical model
motor
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.)
Granted
Application number
CN202111176407.6A
Other languages
Chinese (zh)
Other versions
CN114142782B (en
Inventor
朱延正
童显芳
许诺
王祚
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.)
Huaqiao University
Original Assignee
Huaqiao 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 Huaqiao University filed Critical Huaqiao University
Priority to CN202111176407.6A priority Critical patent/CN114142782B/en
Publication of CN114142782A publication Critical patent/CN114142782A/en
Application granted granted Critical
Publication of CN114142782B publication Critical patent/CN114142782B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/06—Rotor flux based control involving the use of rotor position or rotor speed sensors
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/13—Observer control, e.g. using Luenberger observers or Kalman filters
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/14—Estimation or adaptation of machine parameters, e.g. flux, current or voltage
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P25/00—Arrangements or methods for the control of AC motors characterised by the kind of AC motor or by structural details
    • H02P25/02—Arrangements or methods for the control of AC motors characterised by the kind of AC motor or by structural details characterised by the kind of motor

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

The invention provides an estimation and compensation method for faults of an actuator of an asynchronous motor, which comprises the following steps: building a five-order nonlinear mathematical model of the asynchronous induction motor based on the assumption of a linear magnetic circuit, improving the built mathematical model of the asynchronous induction motor according to the switching characteristic of the asynchronous motor to obtain an improved nonlinear mathematical model of the asynchronous induction motor, and discretizing the improved nonlinear mathematical model of the asynchronous induction motor; establishing a fault detection filter, generating a residual signal through the fault detection filter, and judging whether the motor has an actuator fault according to the value of the residual signal; designing a fault estimator for estimating the fault size of the actuator; compensating the actuator fault through fault reconstruction; the fault estimation and compensation scheme of the asynchronous motor actuator provided by the invention can well estimate the fault when the fault of the actuator occurs, and offset the negative effect caused by the fault, thereby realizing the compensation of the fault.

Description

Fault estimation and compensation method for asynchronous motor actuator
Technical Field
The invention relates to the field of asynchronous motor actuators, in particular to a fault estimation and compensation method for an asynchronous motor actuator.
Background
Asynchronous motors have a very important position for the industrial production and daily life of the current society, and have been applied to various fields of national economy, so that the reliability of realizing the operation of the asynchronous motors in severe working environments is a basic and main requirement of the asynchronous motors in many applications. Essentially, an asynchronous machine is a complex system that is nonlinear, time-varying in parameters, and strongly coupled. In order to achieve better control of the asynchronous motor, the three-phase inverter is widely applied to control of the asynchronous motor due to good control performance of the three-phase inverter.
In addition, the efficient, stable and safe operation of the motor is a very much concerned problem, the motor can be out of order in frequent starting, braking and artificial illegal operation of the motor, if the fault can be found in time before the fault occurs, the loss can be greatly reduced and the safety of operators can be ensured by compensating the fault and stopping the machine for maintenance in time, and the inverter in the asynchronous motor system is one of the parts with higher fault rate, and because the inverter is very important to control the asynchronous motor, an effective strategy for estimating and compensating the fault of the three-phase inverter of the asynchronous motor is urgently needed, so that the estimation and effective compensation of the fault of the inverter of the asynchronous motor can be carried out highly corrected, thereby the influence of the fault on the asynchronous motor is counteracted and the loss is reduced.
Since the inverter corresponds to an actuator in a control system of an asynchronous motor, we refer to the three-phase inverter as the actuator. Currently, many studies have been made on fault estimation and compensation methods for asynchronous motor actuators, such as fault estimation based on signal processing, fault estimation based on statistical analysis, fault estimation based on experts, fault estimation based on mathematical models, and so on. However, certain disadvantages and drawbacks exist, and most do not provide an effective means for fault compensation. On the other hand, both the existing fault estimation and compensation methods are obtained without considering the handover method. At present, based on a switching system method, the problem of comprehensive design aiming at an actuator of fault detection, estimation and compensation of a linear parameter change system with residence time constraint is adopted, and an effective solution is not provided.
Disclosure of Invention
The invention mainly aims to overcome the defects in the prior art, and provides a fault estimation and compensation method for an asynchronous motor actuator, which can accurately estimate and compensate the fault of the asynchronous motor actuator.
The invention adopts the following technical scheme:
a method for estimating and compensating for faults of an asynchronous motor actuator is characterized by comprising the following steps:
building a five-order nonlinear mathematical model of the asynchronous induction motor based on the assumption of a linear magnetic circuit, improving the built mathematical model of the asynchronous induction motor according to the switching characteristics of the asynchronous motor to obtain the improved nonlinear mathematical model of the asynchronous induction motor, and discretizing the improved nonlinear mathematical model of the asynchronous induction motor;
establishing a fault detection filter, generating a residual signal through the fault detection filter, and judging whether the motor has an actuator fault according to the value of the residual signal;
designing a fault estimator for estimating the fault size of the actuator;
and compensating the actuator fault through fault reconstruction.
Specifically, a five-order nonlinear mathematical model of the asynchronous induction motor is built on the assumption of a linear magnetic circuit, and the built mathematical model of the asynchronous induction motor is improved according to the switching characteristic of the asynchronous motor, so that the improved nonlinear mathematical model of the asynchronous induction motor is obtained; the method specifically comprises the following steps:
a five-order nonlinear mathematical model of the asynchronous induction motor is constructed on the assumption of a linear magnetic circuit, and the mathematical model comprises the following steps:
Figure BDA0003295255180000021
the nonlinear mathematical model of the improved asynchronous induction motor is as follows:
Figure BDA0003295255180000022
where ω is the rotor speed, the letters are marked ". cndot." and
Figure BDA0003295255180000023
respectively representing the differential of the corresponding letter physical quantity and defining the left side as the right side; tau isLRepresenting a load moment disturbance;
Figure BDA0003295255180000024
the state vector is represented by a vector of states,
Figure BDA0003295255180000025
which represents the magnetic flux of the rotor,
Figure BDA0003295255180000026
which is representative of the stator current,
Figure BDA0003295255180000027
representing the stator voltage and as a control system input vector;
Figure BDA0003295255180000028
and the rotor speed range is omega epsilon [ omega ]min,ωmax]＝[-110,110]rad/s；a1＝npLsr/(DmLr)，a2＝-Rm/Dm， a3＝-1/Dm，a4,i＝-1/Tr,i，a5,i＝Lsr/Tr,i，a6,i＝Lsr/(Tr,iσLsLr)，a7＝npLsr/(σLsLr)， a8＝1/(σLs)，
Figure BDA0003295255180000029
wherein LsIs a stator inductance, LrIs the rotor inductance, LsrIs mutual inductance, σ is leakage factor, Rs,iIs stator resistance, Rr,iIs rotor resistance, DmIs the moment of inertia, RmIs a viscous damping constant, npIs the number of magnetic pole pairs;
specifically, discretizing the improved nonlinear mathematical model of the asynchronous induction motor specifically comprises:
Figure BDA0003295255180000031
where x (k) represents system state, y (k) represents system output, u (k) represents control input, d (k) represents unknown input, f (k) represents fault input, and matrix Aσk(ρk)，Bσk(ρk),Cσk(ρk) Is a known function of p of a measurable parameter variable, BdIs a given constant matrix of appropriate dimensions, BfIs a matrix representing the impact of each fault on the system; switching signal sigmakThe residence time constraint is satisfied,
Figure BDA0003295255180000032
n represents the number of subsystems in the system; let σ for conveniencekI.e. i
Figure BDA0003295255180000033
Specifically, a fault detection filter is established, and a residual signal is generated by the fault detection filter, specifically:
Figure BDA0003295255180000034
Figure BDA0003295255180000035
is an estimate of x (k) and,
Figure BDA0003295255180000036
is a residual estimate signal, AF,i(ρk)，BF,i(ρk)，CF,i(ρk)，DF,i(ρk) Is the parameter matrix of the fault detection filter.
Specifically, whether the motor has an actuator fault is judged according to the magnitude of the residual signal value, specifically:
the residual estimation equation and its critical values are:
Figure BDA0003295255180000037
Figure BDA0003295255180000038
sL is the time window of the residual estimate, k0Representing an initial estimate; f and d represent fault and external disturbance, respectively; l2Represents the euclidean 2 norm with values of 0, ∞; sup denotes the supremum, i.e. the minimum upper bound.
Specifically, a fault estimator is designed to estimate the fault size of the actuator, specifically:
Figure BDA0003295255180000041
wherein ,
Figure BDA0003295255180000042
in order to be able to estimate the state,
Figure BDA0003295255180000043
in order to be able to estimate the fault,
Figure BDA0003295255180000044
and
Figure BDA0003295255180000045
is the observer gain matrix.
Specifically, the actuator fault is compensated through fault reconstruction, specifically:
Figure BDA0003295255180000046
ur(k) is a reference input, Ki(ρk) Is the LPV feedback gain matrix to be determined,
Figure BDA0003295255180000047
is used to compensate for the effect of the fault on the system.
As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following advantages:
(1) the fault estimation and compensation scheme of the asynchronous motor actuator provided by the invention can well estimate the fault when the fault of the actuator occurs, and offset the negative effect caused by the fault, thereby realizing the compensation of the fault.
Drawings
FIG. 1 is a diagram of steps implemented in a method for estimating and compensating for faults in an asynchronous motor actuator;
FIG. 2 is a schematic diagram of estimation and compensation of asynchronous motor actuator faults;
FIG. 3 is a schematic diagram of a fault detection structure of an asynchronous motor actuator;
FIG. 4 is a schematic diagram of a fault estimation structure of an asynchronous motor actuator;
fig. 5 is a schematic diagram of a fault reconstruction structure of an asynchronous motor actuator.
Detailed Description
The invention is further described below by means of specific embodiments.
Referring to fig. 1, the method for estimating and compensating the fault of the actuator of the asynchronous motor is implemented by the invention, firstly, a dynamic model of the asynchronous motor is built, then, whether the fault exists in a motor system is judged through a residual error, then, the residual error of the system is estimated, finally, the fault is reconstructed according to a residual error estimated value, and finally, the fault compensation is realized, and a schematic diagram of the method is shown in fig. 2.
(1) Modeling
The invention discloses a fault estimation and compensation method for an actuator of an asynchronous motor. Based on the assumption of a linear magnetic circuit, a mathematical model of five-order nonlinear dynamics of an induction motor is as follows:
Figure BDA0003295255180000051
where ω is the rotor speed, as used herein
Figure BDA0003295255180000052
The state vector is represented by a vector of states,
Figure BDA0003295255180000053
which represents the magnetic flux of the rotor,
Figure BDA0003295255180000054
which is representative of the stator current,
Figure BDA0003295255180000055
representing stator voltage and as control system input vector
Figure BDA0003295255180000056
Representing the measurement output. We specify that rotor speed ω is a parameter, such that
Figure BDA0003295255180000057
In order to be a new state vector,
Figure BDA0003295255180000058
is the new output vector. It is found from the analysis that the induction motor does have a switching characteristic, and the stator resistance R is caused by a medium change in the motor environment, particularly a temperature changesVariation of (2) causing rotor resistance R when speed is regulated by switching different selectorsrChange, so that the mathematical model of the system is modified to switch the control system pair as follows when taking into account the switching phenomena
Figure BDA0003295255180000059
Figure BDA00032952551800000510
wherein
Figure BDA00032952551800000511
And the rotor speed range is omega epsilon [ omega ]min,ωmax]＝[-110,110]rad/s。a1＝npLsr/(DmLr)，a2＝-Rm/Dm，a3＝-1/Dm， a4,i＝-1/Tr,i，a5,i＝Lsr/Tr,i，a6,i＝Lsr/(Tr,iσLsLr)，a7＝npLsr/(σLsLr)，a8＝1/(σLs)，
Figure BDA00032952551800000512
wherein LsIs a stator inductance, LrIs the rotor inductance, LsrFor mutual inductance, σ is the leakage factor, Rs,iIs stator resistance, Rr,iIs rotor resistance, DmIs the moment of inertia, RmIs a viscous damping constant, npIs the number of pole pairs. Parameters of the induction machine are shown in the table below, for
Figure BDA00032952551800000513
For the system, the sampling period is TsDiscretization was performed 0.002s and p was introduced1(ω)＝(ω-ωmin)/(ωmax-ω)，p2(ω)＝1-p1(ω) results in a switched Linear Parameter Variation (LPV) system in the discrete time domain, hence M-2 is directly obtained. The system mathematical model is as follows:
Figure BDA0003295255180000061
where x (k) represents system state, y (k) represents system output, u (k) represents control input, d (k) represents unknown input, f (k) represents fault input,
Figure BDA0003295255180000062
fl(k)(l＝1,2,...,m2) Corresponding to a particular fault. Matrix Aσk(ρk)，Bσk(ρk),Cσk(ρk) Is a known function of p, the measurable parameter variable, bounded at a given tight set Θ. B isdIs a given constant matrix of appropriate dimensions, BfIs a matrix representing the impact of each fault on the system. Among all the classes of possible faults, actuator faults are one of the most difficult faults to handle, since actuators can cause great disturbances to the final control result. From the time-dependent behavior, it is assumed that the switching signal σkThe residence time constraint is satisfied and,
Figure BDA0003295255180000063
n represents the number of subsystems in the system (3), when sigma iskWhen the system (3) is represented by i, respectively
Figure BDA0003295255180000064
Figure BDA0003295255180000065
Assuming that the parametric variables are in the form of polyhedra, Ai(ρk) Can be rewritten as:
Figure BDA0003295255180000066
where p is[m]：
Figure BDA0003295255180000067
Is a determined continuous function, Ai，mIs a matrix of known constants
Figure BDA0003295255180000068
M is in the range of { 1. Furthermore, we assume that the mapping p:
Figure BDA0003295255180000069
satisfy the requirement of
Figure BDA00032952551800000610
Figure BDA00032952551800000611
Therefore, it is considered that Ai(ρk) Located in convex hull Co { Ai,1,...,Ai,MOn, for convenience, we use
Figure BDA00032952551800000612
In place of p[m](ρk)。
In the switching sequence to the system (3), the switching time is
Figure RE-GDA00034964392100000614
Residence time of τdAnd satisfy the pair
Figure RE-GDA00034964392100000615
Are all provided with
Figure RE-GDA00034964392100000616
If present
Figure RE-GDA00034964392100000617
Class function xi and
Figure RE-GDA00034964392100000618
class function
Figure RE-GDA00034964392100000622
All initial states of the system
Figure RE-GDA00034964392100000619
And all unknown inputs
Figure RE-GDA00034964392100000620
All satisfy
Figure RE-GDA00034964392100000623
The nonlinear system x (k +1) ═ f (x (k), d (k)) is called input-state stable (ISS). For the purpose of studying (ISS) of the system (3), it is assumed here that (A)i(ρk)， Ci(ρk) Is consistently observable, (A)i(ρk)，Bi(ρk) Are consistently calm, are
Figure RE-GDA00034964392100000621
All holds true, below we perform fault detection, fault estimation and fault compensation on the system.
The method comprises the following steps: the fault detection in the present invention consists of a residual generator and a residual estimator, as shown in the fault tolerant control schematic of fig. 3. The basic idea is to develop a Fault Detection Filter (FDF) that generates a residual signal that is both fault indicator or fault-accentuated, which is sensitive to faults but robust to unknown inputs. When the system has no fault, the residual error is usually zero or close to zero, but when the system has fault, the residual error is obviously different from zero, namely, the value of the residual error is greatly different between fault and non-fault, and the residual error can be well detected. Let σ generate the residual signalkFull-order FDF of i-system (3) such asThe following are shown:
Figure BDA0003295255180000071
herein, the
Figure BDA0003295255180000072
Is an estimate of x (k) and,
Figure BDA0003295255180000073
is a residual signal, AF,i(ρk)，BF,i(ρk)，CF,i(ρk)，DF,i(ρk) Is the parameter matrix of the designed filter, all in the form of LPV. For faster detection of the fault f (k) the following weighted faults can be constructed,
Figure BDA0003295255180000074
is a given stable weighting matrix.
Figure BDA0003295255180000075
Without loss of generality, the minimum implementation of (7) is assumed to be:
Figure BDA0003295255180000076
Figure BDA0003295255180000077
is a weighted fault condition that is a condition of a fault,
Figure BDA0003295255180000078
is a weighted fault, AW，BW，CW，DWIs a known constant matrix of appropriate dimensions.
Is represented by e (k)
Figure BDA0003295255180000079
The model of the augmentation system (3) generates the following augmentation matrix in states including (6) and (8):
Figure BDA00032952551800000710
here, the
Figure BDA00032952551800000711
Figure BDA00032952551800000712
Figure BDA00032952551800000713
While
Figure BDA00032952551800000714
Figure BDA00032952551800000715
Through the analysis, the fault detection problem of the system (3) can be converted into a class
Figure BDA00032952551800000716
And (4) a filtering problem. The main objective here is to design a suitable FDF gain matrix A for the system (3)F,i(ρk)，BF,i(ρk)，CF,i(ρk)，DF,i(ρk) Thereby ensuring that the extended system (9) is progressively stabilized when ω (k) is 0 and that the infimum bound of γ becomes small in the zero initial state, while satisfying:
Figure BDA0003295255180000081
the residual estimation equation and its critical values are:
Figure BDA0003295255180000082
where sL is the time window of the residual estimate, k0Representing the initial estimate, it is noted that in practical estimation the length of the time window is limited, so that estimation of the residual is not realistic over time, and it is always desirable to detect actuator failure as early as possible. Referring to the system (3), whether the system is malfunctioning is determined based on the logic:
Figure RE-GDA0003496439210000083
secondly, the step of: the above work completes the detection of system faults, and the following estimation of system states and faults is realized based on the LPV observer, as shown in fig. 4, and when a fault alarm is triggered in the fault detection stage, the following synchronous state and fault estimator is introduced:
Figure BDA0003295255180000084
here, the
Figure BDA0003295255180000085
In order to be able to estimate the state,
Figure BDA0003295255180000086
in order to be able to estimate the fault,
Figure BDA0003295255180000087
and
Figure BDA0003295255180000088
is the observer gain matrix. The dynamic estimation error is:
Figure BDA0003295255180000089
the state estimation error and fault estimation error are thus derived based on systems (3) and (13) as:
Figure BDA00032952551800000810
wherein
Figure BDA00032952551800000811
Further to the simplification of (15), an extended estimation error system is established as follows:
Figure BDA0003295255180000091
here, the
Figure BDA0003295255180000092
Figure BDA0003295255180000093
By (16) obtaining a gain matrix of a state and fault estimation observer of an augmented system
Figure BDA0003295255180000094
③: finally, the actuator fault is compensated through fault reconstruction, a schematic diagram of a reconstruction structure is shown in fig. 5, and it is assumed that:
rank(Bi(ρk),Bf)＝rank(Bi(ρk)) (17)
in case of activation of an alarm, a Fault Tolerant Control (FTC) is established containing fault compensation based on the estimation of the state and fault as follows:
Figure BDA0003295255180000095
u herer(k) Is a reference input, Ki(ρk) Is the LPV feedback gain matrix to be determined,
Figure BDA0003295255180000096
is used for compensatingThe impact of the fault on the system. Assume reference input ur(k) Is equal to 0 and has
Figure BDA0003295255180000097
Figure BDA0003295255180000098
Figure BDA0003295255180000099
The system (3) thus switches the LPV system at discrete times including fault compensation as:
Figure BDA00032952551800000910
in order to overcome the defects of the multiple Lyapunov equation, the invention adopts an ISS Lyapunov equation with LPV dependent residence time, thereby obtaining a satisfactory ISS gain performance index as follows:
Figure BDA00032952551800000911
wherein
Figure RE-GDA00034964392100000912
And P isi,l(ρk)＞0，
Figure BDA00032952551800000913
l＝0,1,...,τd。
(2) Solving for
The method comprises the following steps: when tau isdAnd gamma is a normal number, and gamma is,
Figure BDA00032952551800000914
in the case of a positive timing matrix, the FDF parameter matrix in (4) can be made to be:
Figure BDA0003295255180000101
secondly, the step of: assume (22) that the following conditions are satisfied:
Figure BDA0003295255180000102
Figure BDA0003295255180000103
Figure BDA0003295255180000104
Figure BDA0003295255180000105
wherein
Figure BDA0003295255180000106
τd＞0，
Figure BDA0003295255180000107
Based on (23) - (26) we can calculate the ISS gain performance index as follows:
Figure BDA0003295255180000108
by (27), when k is 0 and k is z-1, and applying formula (23) we have:
Figure BDA0003295255180000109
ISS gain of
Figure BDA00032952551800001010
To obtain the minimum value of the gain, conversion is made into a solution such asThe convex optimization problem of the following form:
Figure BDA00032952551800001011
and the above convex optimization needs to satisfy the following matrix inequalities (30) to (32):
Figure BDA00032952551800001012
Figure BDA00032952551800001013
Figure BDA00032952551800001014
while
Figure BDA00032952551800001015
Figure BDA0003295255180000111
Where P isi,l,m，Gi,l,m，Zi,l,mAre all positive definite matrices, scalars
Figure BDA0003295255180000112
And delta represents the mutual transposition of the symmetric positions in the matrix. And system (16)
Figure BDA0003295255180000113
The associated ISS gain is
Figure BDA0003295255180000114
Furthermore, the gain matrix obtained from the observer estimator is:
Figure RE-GDA0003496439210000115
here, the
Figure BDA0003295255180000116
M, n ∈ {1,..., M }. The optimal solutions under the constraints (30) - (32) of the above Linear Matrix Inequalities (LMIs) can be solved with the MATLAB LMI toolbox.
③: the optimal solution satisfying the conditions of the closed-loop ISS system (18) and the fault-tolerant control system (19) is solved, and the Liya Ponuff equation (20) is considered
Figure BDA0003295255180000117
And
Figure BDA0003295255180000118
Figure BDA0003295255180000119
m, n ∈ {1,. said, M } satisfies the following (33) - (36) matrix inequalities:
||x(k)||2≤Vi,l(x(k),ρk)≤κ-1||x(k)||2 (34)
||x(k)||2≤Vi,τd(x(k),ρk)≤κ-1||x(k)||2 (35)
Figure BDA00032952551800001110
Figure BDA00032952551800001111
if a symmetric matrix Q existsi,l,m，Qi,τd,m，Ui,l,m,Ui,τd,mScalar τd＞0，
Figure BDA00032952551800001112
0<Kappa.ltoreq.1 such that:
Figure BDA00032952551800001113
Figure BDA00032952551800001114
Qi,l,m≥κI,Qi,τd,m≥κI (40)
here, the
Figure BDA00032952551800001115
To pair
Figure BDA00032952551800001116
Figure BDA0003295255180000121
M, n ∈ {1,..., M } are all satisfied, solving for system (19) concerns d (k), ef(k)，ex(k) ISS gain of
Figure BDA0003295255180000122
And the gain matrix for the corresponding controller is obtained as follows:
Figure RE-GDA0003496439210000123
here Ki,l,m＝Ui,l,mQ-1 i,l,m，
Figure BDA0003295255180000124
m,n∈{1,...,M}。
Analyzing the parameters of the Fault Detection Filter (FDF) and obtaining the gain matrix based on the observer estimator
Figure BDA0003295255180000125
And fault tolerant controller gain matrix Ki(ρk)。
The above description is only an embodiment of the present invention, but the design concept of the present invention is not limited thereto, and any insubstantial modifications made by using the design concept should fall within the scope of infringing the present invention.

Claims (7)

1. A method for estimating and compensating for faults of an asynchronous motor actuator is characterized by comprising the following steps:
building a five-order nonlinear mathematical model of the asynchronous induction motor based on the assumption of a linear magnetic circuit, improving the built mathematical model of the asynchronous induction motor according to the switching characteristic of the asynchronous motor to obtain an improved nonlinear mathematical model of the asynchronous induction motor, and discretizing the improved nonlinear mathematical model of the asynchronous induction motor;
establishing a fault detection filter, generating a residual signal through the fault detection filter, and judging whether the motor has an actuator fault according to the value of the residual signal;
designing a fault estimator for estimating the fault size of the actuator;
and compensating the actuator fault through fault reconstruction.
2. The method for estimating and compensating the fault of the asynchronous motor actuator according to claim 1, wherein a five-order nonlinear mathematical model of the asynchronous induction motor is constructed on the basis of the assumption of a linear magnetic circuit, and the established squirrel-cage induction motor mathematical model is improved according to the switching characteristic of the asynchronous motor to obtain the improved nonlinear mathematical model of the asynchronous induction motor; the method specifically comprises the following steps:
a five-order nonlinear mathematical model of the asynchronous induction motor is constructed on the assumption of a linear magnetic circuit, and the mathematical model comprises the following steps:
Figure FDA0003295255170000011
the nonlinear mathematical model of the improved asynchronous induction motor is as follows:
Figure FDA0003295255170000012
where ω is the rotor speed, the letters are marked ". cndot." and
Figure FDA0003295255170000013
respectively representing the differential of the corresponding letter physical quantity and defining the left side as the right side; tau isLRepresenting a load moment disturbance;
Figure FDA0003295255170000014
the state vector is represented by a vector of states,
Figure FDA0003295255170000015
which represents the magnetic flux of the rotor,
Figure FDA0003295255170000016
which is representative of the stator current,
Figure FDA0003295255170000017
representing the stator voltage and as a control system input vector; the upper right hand "T" indicates transpose.
Figure FDA0003295255170000021
And the rotor speed range is omega epsilon [ omega ]min,ωmax]＝[-110,110]rad/s；a1＝npLsr/(DmLr)，a2＝-Rm/Dm，a3＝-1/Dm，a4,i＝-1/Tr,i，a5,i＝Lsr/Tr,i，a6,i＝Lsr/(Tr,iσLsLr)，a7＝npLsr/(σLsLr)，a8＝1/(σLs)，
Figure FDA0003295255170000022
wherein LsIs a stator inductance, LrIs the rotor inductance, LsrIs mutual inductanceσ is the leakage factor, Rs,iIs stator resistance, Rr,iIs rotor resistance, DmIs the moment of inertia, RmIs a viscous damping constant, npIs the number of pole pairs.
3. The method for estimating and compensating for the fault of the asynchronous motor actuator according to claim 2, wherein the improved nonlinear mathematical model of the asynchronous induction motor is discretized, and specifically:
Figure FDA0003295255170000023
where x (k) represents system state, y (k) represents system output, u (k) represents control input, d (k) represents unknown input, f (k) represents fault input, and matrix Aσk(ρk)，Bσk(ρk),Cσk(ρk) Is a known function of p of a measurable parameter variable, BdIs a given constant matrix of appropriate dimensions, BfIs a matrix representing the impact of each fault on the system; switching signal sigmakThe residence time constraint is satisfied and,
Figure FDA0003295255170000024
n represents the number of subsystems in the system; let σ for conveniencekI.e. i
Figure FDA0003295255170000025
4. The method for estimating and compensating for faults of an asynchronous motor actuator according to claim 3, wherein a fault detection filter is established, and a residual signal is generated by the fault detection filter, specifically:
Figure FDA0003295255170000026
Figure FDA0003295255170000027
is an estimate of x (k) and,
Figure FDA0003295255170000028
is a residual estimation signal, and the superscript "-" indicates the estimation of the physical quantity represented by the letter; a. theF,i(ρk)，BF,i(ρk)，CF,i(ρk)，DF,i(ρk) Is the parameter matrix of the fault detection filter.
5. The method for estimating and compensating for the actuator fault of the asynchronous motor according to claim 4, wherein whether the actuator fault exists in the motor is judged according to the magnitude of the residual signal value, and specifically, the method comprises the following steps:
the residual estimation equation and its critical values are:
Figure FDA0003295255170000031
Figure FDA0003295255170000032
sLis the time window of the residual estimate, k0Representing an initial estimated value, f and d representing a fault and an external disturbance respectively; l2Represents the euclidean 2 norm with values of 0, ∞; sup denotes the supremum, i.e. the minimum upper bound.
6. The method for estimating and compensating the fault of the asynchronous motor actuator according to claim 5, wherein a fault estimator is designed to estimate the fault magnitude of the actuator, and specifically comprises the following steps:
Figure FDA0003295255170000033
wherein ,
Figure FDA0003295255170000034
in order to be able to estimate the state,
Figure FDA0003295255170000035
in order to be able to estimate the fault,
Figure FDA0003295255170000036
and
Figure FDA0003295255170000037
is the observer gain matrix.
7. The method for estimating and compensating for the fault of the asynchronous motor actuator according to claim 6, wherein the fault of the actuator is compensated through fault reconstruction, specifically:
Figure FDA0003295255170000038
ur(k) is a reference input, Ki(ρk) Is the LPV feedback gain matrix to be determined,
Figure FDA0003295255170000039
is used to compensate for the effect of the fault on the system.
CN202111176407.6A 2021-10-09 2021-10-09 Estimation and compensation method for faults of asynchronous motor actuator Active CN114142782B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111176407.6A CN114142782B (en) 2021-10-09 2021-10-09 Estimation and compensation method for faults of asynchronous motor actuator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111176407.6A CN114142782B (en) 2021-10-09 2021-10-09 Estimation and compensation method for faults of asynchronous motor actuator

Publications (2)

Publication Number Publication Date
CN114142782A true CN114142782A (en) 2022-03-04
CN114142782B CN114142782B (en) 2023-08-29

Family

ID=80394250

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111176407.6A Active CN114142782B (en) 2021-10-09 2021-10-09 Estimation and compensation method for faults of asynchronous motor actuator

Country Status (1)

Country Link
CN (1) CN114142782B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106569487A (en) * 2016-11-16 2017-04-19 北京理工大学 Fault detection method for servo system based on observer
CN108845495A (en) * 2018-04-03 2018-11-20 南通大学 Intermittent fault diagnosis and Active Fault-tolerant Control Method based on the double-deck Kalman filter
KR20190078120A (en) * 2017-12-26 2019-07-04 (주)유진에코씨엘 Extended Luenberger-Sliding Mode Observer Capable of Estimating Rotor flux and Rotor Resistance for Three Phase Induction Motor
CN110460058A (en) * 2018-05-07 2019-11-15 南京理工大学 A kind of control method of non-linear THE UPFC
CN113467240A (en) * 2021-07-02 2021-10-01 湖州师范学院 Asynchronous filter optimization method based on T-S fuzzy Markov jump system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106569487A (en) * 2016-11-16 2017-04-19 北京理工大学 Fault detection method for servo system based on observer
KR20190078120A (en) * 2017-12-26 2019-07-04 (주)유진에코씨엘 Extended Luenberger-Sliding Mode Observer Capable of Estimating Rotor flux and Rotor Resistance for Three Phase Induction Motor
CN108845495A (en) * 2018-04-03 2018-11-20 南通大学 Intermittent fault diagnosis and Active Fault-tolerant Control Method based on the double-deck Kalman filter
CN110460058A (en) * 2018-05-07 2019-11-15 南京理工大学 A kind of control method of non-linear THE UPFC
CN113467240A (en) * 2021-07-02 2021-10-01 湖州师范学院 Asynchronous filter optimization method based on T-S fuzzy Markov jump system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
姜陶然;李涛;王丽华;: "基于UIO方法的非线性Markov跳跃系统故障重构", 东南大学学报(自然科学版), no. 1 *
王旭;沈艳霞;吴定会;: "一类非线性系统的故障重构与容错控制", 测控技术, no. 10 *

Also Published As

Publication number Publication date
CN114142782B (en) 2023-08-29

Similar Documents

Publication Publication Date Title
Li et al. Robust simultaneous fault estimation and nonfragile output feedback fault-tolerant control for Markovian jump systems
CN113078644B (en) Self-adaptive event trigger control method of permanent magnet synchronous generator coupled chaotic network system
CN106788044B (en) A kind of adaptive non-singular terminal sliding-mode control of permanent magnet synchronous motor based on interference observer
Wang et al. Event-based disturbance compensation control for discrete-time SPMSM with mismatched disturbances
Sun et al. Torque ripple reduction for a 12/8 switched reluctance motor based on a novel sliding mode control strategy
Boufadene et al. Adaptive nonlinear observer augmented by radial basis neural network for a nonlinear sensorless control of an induction machine
Zhang et al. Multiple-model-based diagnosis of multiple faults with high-speed train applications using second-level adaptation
Boukhlouf et al. Hardware-in-the-loop implementation of an unknown input observer for synchronous reluctance motor
Gundogdu et al. Comparison of Si-ANN and extended Kalman filter-based sensorless speed controls of a DC motor
CN114142782A (en) Fault estimation and compensation method for asynchronous motor actuator
Djelamda et al. Field-oriented control based on adaptive neuro-fuzzy inference system for PMSM dedicated to electric vehicle
CN114840969B (en) Active fault-tolerant control method of nonlinear electromechanical system
Stender et al. Accurate torque estimation for induction motors by utilizing a hybrid machine learning approach
Chacko et al. Rotor resistance estimation of vector controlled induction motor drive using GA/PSO tuned fuzzy controller
Brescia et al. Parameter identification of pmsms considering vsi nonlinearity with coupled adaline nns
Wang et al. A high performance permanent magnet synchronous motor servo system using predictive functional control and Kalman filter
Khodamoradi et al. Design of Robust H∞ Controller based on Nonlinear Observer for Sensorless PMSM using LMIs
Yi et al. Intelligent detection of stator and rotor faults of induction motor based on improved backstepping sliding mode observer
KR20030075446A (en) Precision speed and position control method of actuator system
Ghariani et al. Online parameters estimation scheme using artificial intelligence for vector Controlled Speed Sensorless induction motor drive
Allouche et al. Takagi-sugeno fuzzy observer design for induction motors with immeasurable decision variables: State estimation and sensor fault detection
Pan et al. A high-order sliding mode stator flux observer for asynchronous motor based on feedback linearization
Teler et al. Analysis of the stator current prediction capabilities in induction motor drive using the LSTM network
Lan et al. Disturbance Observer‐Based Complementary Fractional‐Order Sliding Mode Control for PMSM Drive System
CN114172436B (en) Asynchronous motor instruction filtering discrete control method based on observer

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
GR01 Patent grant
GR01 Patent grant