CN103984233B - A kind of quadrotor dual-granularity method for diagnosing faults based on mixed model - Google Patents
A kind of quadrotor dual-granularity method for diagnosing faults based on mixed model Download PDFInfo
- Publication number
- CN103984233B CN103984233B CN201410196572.1A CN201410196572A CN103984233B CN 103984233 B CN103984233 B CN 103984233B CN 201410196572 A CN201410196572 A CN 201410196572A CN 103984233 B CN103984233 B CN 103984233B
- Authority
- CN
- China
- Prior art keywords
- model
- quadrotor
- granularity
- mixed model
- dual
- 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.)
- Active
Links
Abstract
The present invention relates to a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model, it includes all kinds of physical effects of four rotor flying of analyzing influence, to determine the modeling category of prior model and nonparametric model；For nonparametric model, nonlinear degree is measured；Suitable parameter identification method is selected according to nonlinear degree, quadrotor mixed model is set up；According to the granularity level of mixed model partition process variable；The passage occurred according to coarse grain level failure judgement and the components and parts that failure generation is determined by fine granularity rank, are achieved in dual-granularity fault diagnosis.The present invention takes full advantage of the construction featuress of mixed model, it is adaptable to the process detection of four-rotor helicopter structure failure and the feasibility checking for diagnosing.
Description
Technical field
The present invention relates to aviation aircraft fault diagnosis technology field, more particularly to a kind of four rotations based on mixed model
Rotor aircraft dual-granularity method for diagnosing faults.
Background technology
Quadrotor is a kind of typical modern complication system, and compared to fixed wing airplane, it has more complicated
Aerodynamic characteristic and more specifically state of flight, need the mathematical model and more sane control law of higher precision to ensure flight product
Matter and flight safety.
The traditional modeling method of quadrotor is roughly divided into modelling by mechanism and data-driven modeling.Modelling by mechanism parameter
Interpretability is strong, and model extensionality is good, but for the modern times such as multivariate, non-linear and close coupling complication system, then it is difficult to model
Degree is larger；Data-driven modeling does not need the priori of process object, but model accuracy and generalization ability to be highly dependent on and build
Modulus evidence.Therefore, rely solely on single modeling means to be difficult to obtain high-quality object module.
Quadrotor exports three attitude angle signals by four executors, belongs to overdrive system, can be effective
Raising structural load ability and response speed, it in running can inevitably receive external disturbance or break down
Deng；The failure refers to that larger deviation occur at least one characteristic of drive system or parameter, beyond acceptable scope.This
When system performance be significantly lower than its normal level；The classification of the failure can be carried out in terms of different, from the portion that failure occurs
From the point of view of position, actuator failures, sensor fault and structure failure can be divided into.
Problems with is primarily present for the method for diagnosing faults of quadrotor：
(1) in terms of diagnostic method, mainly based on the method for diagnosing faults based on model, diagnosis process have ignored in a large number
The data characteristic and partial fault information of process variable；
(2) in terms of fault type, most of documents consider the fault diagnosis of quadrotor under actuator failures, few
Amount document considers the fault diagnosis of sensor, and structure failure is due to complicated fault model and diversified fault type
And it is few studied.
In sum, the deficiencies in the prior art how are overcome to become in current aviation aircraft fault diagnosis technology field
One of emphasis difficult problem urgently to be resolved hurrily.
The content of the invention
The purpose of the present invention is to overcome the shortcomings of to provide a kind of four rotations based on mixed model existing for prior art
Rotor aircraft dual-granularity method for diagnosing faults, the present invention in recurring structure failure, using based on physical effect analysis and
The mixed model set up by nonlinear metric, by the granularity level of refining data, can improve the accuracy of fault diagnosis, be suitable for
In the process detection and the feasibility checking for diagnosing of four-rotor helicopter structure failure.
According to a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model proposed by the present invention, its
It is characterised by that it comprises the following specific steps that：
Step A：The influence degree of quadrotor is analyzed using all kinds of physical effects, will according to influence degree
Which is divided into major influence factors and minor effect factor, determines prior model and non-ginseng in quadrotor mixed model
The modeling category of exponential model, and prior model is set up according to major influence factors；
Step B：For all kinds of nonlinear terms and coupling terms in minor effect factorial analysiss nonparametric model, tolerance four is revolved
The nonlinear degree of rotor aircraft；
Step C：According to the influence degree and every nonlinear degree of physical effect, select suitable with indistinct logic computer
Parameter identification, linearization technique, set up the mixed model of quadrotor；
Step D：According to source and the own physical meaning of data, the granularity level of quadrotor is divided；
Step E：Using the process monitoring method based on pivot analysis, first for the data of coarse grain level, failure is determined
The passage of generation, then for the data of fine granularity rank, position the component of generation of being out of order, be achieved in dual-granularity failure
Diagnosis.
Its remarkable advantage is the present invention compared with prior art：One is that the mixed model that the present invention sets up takes into full account
Influence degree of all kinds of physical effects to quadrotor, while avoiding the model caused due to nonlinear terms and coupling terms
Difficulty in analysis, process；Two is the data message that dual-granularity method for diagnosing faults makes full use of each granularity level, is improve
The accuracy and reliability of diagnosis；Three is the data using coarse grain level, is accurate to the other fault diagnosis of channel level, is different
It is convenient that the modification of passage faults-tolerant control rule is provided；Four is the data using fine granularity rank, is accurate to the fault diagnosis of part
Method, in efficiently avoid conventional fault diagnosis method, structure failure models complexity, there is a polynary difficult problem, preferably expands
The scope of fault diagnosis.
Description of the drawings
Fig. 1 is a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model proposed by the present invention
Step block diagram.
Fig. 2 is Z=X × Y type 3-D view schematic diagrams.
Fig. 3 isType 3-D view schematic diagram.
Fig. 4 is the general process schematic diagram of hybrid modeling.
Fig. 5 is fault detect (attitude angle) schematic diagram for coarse grain level.
Fig. 6 is fault detect (aerodynamic moment) schematic diagram for coarse grain level.
Fig. 7 is contribution rate schematic diagram of the attitude angle to failure.
Fig. 8 is total contribution rate schematic diagram of the attitude angle to failure.
Fig. 9 is contribution rate schematic diagram of the aerodynamic moment to failure.
Figure 10 is total contribution rate schematic diagram of the aerodynamic moment to failure.
Figure 11 is contribution rate schematic diagram of the fuselage gyroscopic couple to failure.
Figure 12 is total contribution rate schematic diagram of the fuselage gyroscopic couple to failure.
Figure 13 is fault detect (voltage) schematic diagram for fine granularity rank.
Figure 14 is fault detect (rotor gyroscopic couple) schematic diagram for fine granularity rank.
Figure 15 is contribution rate schematic diagram of the rotor gyroscopic couple to failure.
Figure 16 is total contribution rate schematic diagram of the rotor gyroscopic couple to failure.
Specific embodiment
Below in conjunction with the accompanying drawings the specific embodiment of the present invention is done and is further described in detail.
With reference to Fig. 1, a kind of quadrotor dual-granularity fault diagnosis side based on mixed model proposed by the present invention
Method, it includes all kinds of physical effects of four rotor flying of analyzing influence, to determine the modeling model of prior model and nonparametric model
Farmland；For nonparametric model, nonlinear degree is measured；Suitable parameter identification method is selected according to nonlinear degree, four are set up
Rotor craft mixed model；According to the granularity level of mixed model partition process variable；According to coarse grain level failure judgement
The passage of generation, determines the components and parts that failure occurs according to fine granularity rank；Specific implementation step is as follows：
Step A：The influence degree of quadrotor is analyzed using all kinds of physical effects, will according to influence degree
Which is divided into major influence factors and minor effect factor, determines prior model and non-ginseng in quadrotor mixed model
The modeling category of exponential model, and prior model is set up according to major influence factors；Wherein：Described all kinds of physical effects, mainly
The form of aerodynamic moment is shown as, the aerodynamic moment is that the pulling force that produced by quadrotor rotor wing rotation and resistance cause
, it is the topmost torque type that aircraft bears, including rolling moment, pitching moment and yawing, belongs to non-ginseng
The modeling category of exponential model；It is major influence factors by described aerodynamic moment, prior model formula can be set up as follows：
In above formula, φ, θ,The angle of pitch, roll angle and yaw angle are represented respectively；Respectively x, y, in z-axis
Aerodynamic moment；L is distance of the motor center point to zero；Kf、Kt,cIt is the constant system between electric moter voltage and torque
Number；V1、V3、V2、V4It is the voltage of the four motor generations in front, rear, left and right；Rolling is represented respectively
Turn passage, pitch channel, jaw channel with regard to x, y, the rotary inertia of z-axis；Jxx,Jyy,JzzRoll channel, pitching are represented respectively
Passage, jaw channel is with regard to x, y, the rotary inertia of z-axis.
Choosing quantity of state isControlled quentity controlled variable u=[V1 V3 V2 V4]T, then described with state equation form
Quadrotor prior model formula be：
Y=Cx
In above formula：
D=0；
In addition to aerodynamic moment this major influence factors, although the gyroscopic couple suffered by system affect relative weak but also
It is very important.Its presence determines the partial approximation performance of system, belongs to minor effect factor, is the modeling of nonparametric model
Category, mainly includes fuselage gyroscopic couple and rotor gyroscopic couple, the above analysis, can set up quadrotor more accurate
Mechanism model formula it is as follows：
After minor effect factor is considered, nonlinear terms and cross-couplings item in modular form, are occurred in that.
Step B：For all kinds of nonlinear terms and coupling terms in minor effect factorial analysiss nonparametric model, tolerance four is revolved
The nonlinear degree of rotor aircraft；Wherein：Described nonlinear degree is referred to：For input signalOne stable
Causal system N:Ua→ Y, its nonlinear degreeIt is defined as non-negative equation：
In above formula：G:Ua→ Y be linear operator, norm | | | |VTIt is defined asT is the cycle；inf
Infimum and supremum are represented respectively with sup；Choose a series of sinusoidal signal composition input set ULB：
ULB=<U | u (t)=A sin (ω t), A ∈ A, ω ∈ Ω>
A=<A∈R+|Amin≤A≤Amax>
Ω=<ω∈R+|ωmin≤ω≤ωmax>
Then stable state output can be expressed as following formula：
According to the definition of nonlinear degree, derive that the computing formula of nonlinear degree lower bound is：
In the modeling process of quadrotor, gyroscopic effect this class minor effect factor belongs to nonparametric model
Modeling category；As the above analysis, first the nonlinear degree of its each several part should be measured.
Now by taking roll channel as an example, the calculating process of nonlinear degree is illustrated.For fuselage gyroscopic effect, its original table
Up to formula it isIt is the coupling terms of two quantity of states.Therefore, can be this class gyroscopic effect abstract for x × y types；It is right
In rotor gyroscopic couple, its original expression isΩiFor the rotating speed of rotor, it is proportional to control
Amount voltage ViSquare, i.e.,Therefore, rotor gyroscopic couple is substantially the coupling terms of quantity of state and controlled quentity controlled variable, can be
Class gyroscopic effect is abstract is for thisType.Fig. 2 and Fig. 3 describe the 3-D view of two class abstract models, intuitively can reflect
Respective nonlinear degree；X and y is represented with different amplitudes, sinusoidal signal Asin (α t) of phase angle and Bsin (β t) respectively, then
X × y types further write A sin (α t) × B sin (β t) types,Type is further writeFor two
Class abstract model chooses enough representational amplitudes and phase angle, can calculate respective nonlinear degree.Table 1 give with
The nonlinear degree result of calculation of fuselage gyroscopic couple and rotor gyroscopic couple as a example by roll channel, wherein, table 1 first,
The result of calculation of two rows represents the nonlinear degree by nonlinear terms Taylor series expansion to the i-th one and Section 2 respectively.
1 gyroscopic effect nonlinear degree of table is calculated
In order to determine suitable linearization technique, after the nonlinear degree for drawing two class gyroscopic effects, comprehensive examining is needed
Consider model accuracy and algorithm complexity.The representational linearization technique of four classes is：Least-squares algorithm linear fitting (linear
Fit, LF), equilibrium point Taylor series expansion (series expansion, SE), T-S Fuzzy Inference Model (Takagi-
Sugeno, TS) and subspace system identification (parameter identification, PI).
Although least-squares algorithm linear fitting algorithm recognizes efficiency high, Identification Errors are larger, are a kind of rough data
Processing method, it is adaptable to the weaker situation of nonlinear degree；Although subspace system identification algorithm accuracy is higher, identification effect
Rate is low, realizes that process is complicated, it is adaptable to the stronger situation of nonlinear degree；According to the respective nonlinear degree of variable and linearly
The applicable elements of change method, can set up fuzzy rule, as shown in table 2 between nonlinear degree and linearization technique；Choose
Used as input quantity, appropriate linearization technique sets up mamdani types as output to the mean square deviation of nonlinear degree and output
Fuzzy Inference Model.
2 fuzzy rule base of table
For the gyroscopic effect suffered by each passage of quadrotor, by situational variables relation, show that two class gyros are imitated
The influence area of response state equation；Through fuzzy reasoning, respective linearization technique is exported from fuzzy rule observation window such as
Shown in table 3.
The principal linear optimization method of 3 nonparametric model of table
According to table 3, linearization process is carried out to nonparametric model each several part, and combines prior model, draw mixed model
State equation.In sum, Fig. 4 gives the general process of hybrid modeling.
Step C：According to the influence degree and every nonlinear degree of physical effect, select suitable with indistinct logic computer
Parameter identification, linearization technique, set up the mixed model of quadrotor；Such as using the method for fuzzy reasoning, non-linear
Foundation mapping between degree and parameter identification, linearization technique, it is final to obtain based on physical effect analysis and nonlinear metric
Quadrotor hybrid guided mode pattern is as follows：
In above formula, fSE<·>,fTS<·>,fPI<·>Represent respectively and use SE, tri- kinds of parameter identification methods process of TS, PI "<
>" in all kinds of gyroscopic effect expression formulas, it is concrete as shown in table 6.
To expect attitude angle as 1 °, reference voltage is UbiasAs a example by the operating mode of=2V, the state equation matrix of mixed model
Formula is as follows：
Step D：According to source and the own physical definition of data, the granularity level of quadrotor is divided；
Specific embodiment includes：
Mixed model is routinely believed except obtaining the output in prior model, quantity of state etc. compared with single model
Breath is outer, can also obtain some unknown parameters by intelligent algorithms such as parameter identification, fuzzy reasonings from nonparametric model, with reference to
The mechanism knowledge of prior model modeling process and controlled device can obtain the physical significance of these unknown parameters；Only from quantity of information
From the point of view of, the data volume of mixed model output is more, it is meant that obtain fault message more comprehensive；However, from fault diagnosis
Angle considers that efficient fault diagnosis needs to divide the granularity level of data.
In mixed model, prior model holds the global property of system, low to the degree of refinement of mixed model, therefrom obtains
Obtain the data message of coarseness level, such as system output and quantity of state；Nonparametric model has good partial approximation performance, right
The degree of refinement of mixed model is high, therefrom obtains the data message of fine granularity level, the such as unknown parameter in prior model.For four
Rotor craft, can choose rolling, the attitude angle of three passages of pitching and driftage and aerodynamic moment as the number of coarse grain level
According to the data of selection fuselage gyroscopic couple, rotor gyroscopic couple and electric moter voltage as fine granularity rank.
Step E：Using the process monitoring method based on pivot analysis, first for the data of coarse grain level, failure is determined
The passage of generation, then for the data of fine granularity rank, position the component of generation of being out of order, be achieved in dual-granularity failure
Diagnosis；Specific embodiment includes：
Using the process monitoring method based on pivot analysis, using the method for diagnosing faults based on traditional contribution plot, with flat
Square forecast error statistic (SPE) is evaluation index；First with quadrotor rolling, pitching and three attitude angle of going off course
The data of the coarse grain levels such as angular velocity determine the approximate range that failure occurs, that is, determine the passage that failure occurs.
In reference voltage 3.5V, expect that attitude angle is under 0.3 ° of operating mode, gather the data of 15 groups of nominal situations, per group
155 sampled points, for setting up Principal Component Analysis Model；12 groups of data, per group of 155 sampled points are gathered under fault condition.
By taking a kind of fault type as an example, fault detect is carried out with the data of coarse grain level first；From Fig. 5 and Fig. 6,
Three attitude angle and aerodynamic moment are different degrees of beyond control limit, therefore detect failure generation；
Secondly, with the method for diagnosing faults based on traditional contribution plot, for attitude angle, aerodynamic moment and fuselage gyro
The data of torque this three classes coarse grain level, Fig. 7, Fig. 8 and Fig. 9 are given in different sampling point variables pair in the form of broken line graph
The variation tendency of failure contribution rate；Figure 10, Figure 11 and Figure 12 have more intuitively counted each variable in the form of block diagram simultaneously
Total contribution rate of failure.
The diagnostic result of comprehensive Fig. 7 to Figure 12, roll angle, rolling moment and roll channel fuselage gyroscopic couple are to failure
Contribution rate be maximum；Therefore, may infer that roll channel there occurs failure according to the fault diagnosis result of coarse grain level.
For roll channel, the above-mentioned failure of physical quantity (rotor gyroscopic couple and electric moter voltage) repetition to fine granularity rank
Diagnosis algorithm；As shown in Figure 13, without departing from control limit, voltage shows that motor does not break down.And be directed to rotor gyroscopic couple and enter
During row fault detect, SPE statistics as shown in figure 14, illustrate that the part that failure occurs is rotor beyond control limit.To sum up institute
State, the Contribution Rate according to Figure 15 and Figure 16, the part for breaking down is front and back rotors.
On this basis, the part that failure occurs finally is determined according to the data of the fine granularity rank such as fuselage gyroscopic effect,
Quadrotor dual-granularity method for diagnosing faults based on mixed model is realized finally.
Jing validation trials of the present invention, achieve satisfied application effect.
Claims (6)
1. a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model, it is characterised in that the method includes
Following concrete steps：
Step A：The influence degree of quadrotor is analyzed using all kinds of physical effects, all kinds of physical effect roots
Major influence factors and minor effect factor are divided into according to its influence degree, the priori in quadrotor mixed model is determined
The modeling category of model and nonparametric model, and prior model is set up according to major influence factors；Wherein：All kinds of physics effects
The form for showing as aerodynamic effect, fuselage gyroscopic effect and rotor gyroscopic effect should be referred to；The major influence factors refer to gas
Kinetic moment, the aerodynamic moment are that the pulling force that produced by quadrotor rotor wing rotation and resistance cause, and aerodynamic moment is winged
The topmost torque type that row device bears, including the form of rolling moment, pitching moment and yawing, belong to nonparametric mould
The modeling category of type；Minor effect factor referred in addition to aerodynamic moment, the impact of the gyroscopic couple suffered by system, mainly including machine
Body gyroscopic couple and rotor gyroscopic couple, are the modeling categories of nonparametric model；
Step B：For all kinds of nonlinear terms and coupling terms in minor effect factorial analysiss nonparametric model, four rotors of tolerance fly
The nonlinear degree of row device；
Step C：According to the influence degree and every nonlinear degree of physical effect, recognize with indistinct logic computer selection parameter,
Linearization technique, sets up the mixed model of quadrotor；
Step D：According to source and the own physical definition of data, the granularity level of quadrotor is divided；
Step E：Using the process monitoring method based on pivot analysis, first for the data of coarse grain level, determine that failure occurs
Passage, then for the data of fine granularity rank, positioning is out of order the component of generation, is achieved in dual-granularity failure and examines
It is disconnected.
2. a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model according to claim 1,
It is characterized in that described aerodynamic moment is major influence factors, prior model formula is set up as follows：
In above formula, φ, θ,The angle of pitch, roll angle and yaw angle are represented respectively；Respectively x, y, the gas in z-axis
Kinetic moment；L is distance of the motor center point to zero；KfIt is rotor torque constant coefficient；Kt,cIt is to push away when rotor is reversely rotated
Constant coefficient between power and torque；V1、V3、V2、V4It is the voltage of the four motor generations in front, rear, left and right；Jxx,Jyy,JzzRespectively
Expression roll channel, pitch channel, jaw channel is with regard to x, y, the rotary inertia of z-axis.
3. a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model according to claim 1,
It is characterized in that the nonlinear degree described in step B is referred to：For input signalOne stable causal system N:
Ua→ Y, the nonlinear degree of the causal systemIt is defined as non-negative equation：
G in above formula is the linear operator in causal system N；G [u] represents causal system linear parts G with regard to input signal u
Function；N [u] represents the function of stable causal system N with regard to input signal u；Norm | | | |VTIt is defined as
T is the cycle；Inf and sup represent infimum and supremum respectively；Again to the non-of the nonlinear terms and coupling terms in nonparametric model
After linear degree tolerance, linearisation, parameter identification method are selected, linearization process is carried out to nonlinear terms and coupling terms.
4. a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model according to claim 2,
It is characterized in that the mixed model described in step C is referred to：Using the method for fuzzy reasoning, nonlinear degree and parameter identification,
Mapping is set up between linearization technique, final acquisition is mixed based on the quadrotor of physical effect analysis and nonlinear metric
Modular form is as follows：
In above formula, ΩiThe rotating speed of four rotors of expression, i=1,2,3,4；F<g(x)>Represent with corresponding parameter identification or linear
Method F of change process bracket "<>" in expression formula g (x) after the result that obtains；F selects one of SE, tri- kinds of methods of TS, PI, SE,
TS, PI represent equilibrium point Taylor series expansion, T-S Fuzzy Inference Models and subspace system identification respectively, and g (x) is all kinds of
Represent the expression formula of gyroscopic effect, JrRepresent the rotary inertia of rotor motor.
5. a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model according to claim 1,
It is characterized in that the granularity level described in step D carries out division referring to：In mixed model, prior model holds the overall situation of system
Characteristic, compared with nonparametric model, prior model is low to the degree of refinement of mixed model, therefrom obtain include system output with
The data message of the coarseness level of quantity of state；Nonparametric model has good partial approximation performance, compared with prior model, non-
Parameter model is high to the degree of refinement of mixed model, therefrom obtains the data of the fine granularity level for including unknown parameter in prior model
Information.
6. a kind of quadrotor dual-granularity method for diagnosing faults based on mixed model according to claim 1,
It is characterized in that the dual-granularity fault diagnosis described in step E is referred to：Using the process monitoring method based on pivot analysis, with flat
Square forecast error statistic carries out fault detect for evaluation index, while carry out failure using the method based on traditional contribution plot examining
It is disconnected, with the larger variable of contribution rate as causing trouble occur the reason for variable, specifically include：
Step E1：Using quadrotor rolling, the number of the coarse grain level of the angular velocity of three attitude angle of pitching and driftage
According to the scope for determining that failure occurs, so that it is determined that the passage that failure occurs；
Step E2：It is determined that on the basis of faulty channel, finally being determined according to the data of the fine granularity rank of fuselage gyroscopic effect
The component that failure occurs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410196572.1A CN103984233B (en) | 2014-05-09 | 2014-05-09 | A kind of quadrotor dual-granularity method for diagnosing faults based on mixed model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410196572.1A CN103984233B (en) | 2014-05-09 | 2014-05-09 | A kind of quadrotor dual-granularity method for diagnosing faults based on mixed model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103984233A CN103984233A (en) | 2014-08-13 |
| CN103984233B true CN103984233B (en) | 2017-03-29 |
Family
ID=51276252
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410196572.1A Active CN103984233B (en) | 2014-05-09 | 2014-05-09 | A kind of quadrotor dual-granularity method for diagnosing faults based on mixed model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103984233B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104290919A (en) * | 2014-09-25 | 2015-01-21 | 南京航空航天大学 | Direct self-repairing control method for four-rotor aircraft |
| CN106444813A (en) * | 2016-10-26 | 2017-02-22 | 成都市优艾维机器人科技有限公司 | A quadrotor attitude control method based on a T-S fuzzy model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010068225A1 (en) * | 2008-12-12 | 2010-06-17 | Karem Aircraft, Inc. | Swashplateless rotorcraft with fault tolerant linear electric actuator |
| CN102126554A (en) * | 2011-01-28 | 2011-07-20 | 南京航空航天大学 | Unmanned air vehicle with multiple rotary wings in plane-symmetry layout |
| CN102945311A (en) * | 2012-10-08 | 2013-02-27 | 南京航空航天大学 | Method for diagnosing fault by functional fault directed graph |
| CN103135553A (en) * | 2013-01-21 | 2013-06-05 | 南京航空航天大学 | Four-rotor aircraft fault-tolerant control method |
| CN103149928A (en) * | 2013-03-24 | 2013-06-12 | 西安费斯达自动化工程有限公司 | Fault diagnosing and tolerance control method for aircraft large-angle-of-attack movement ternary number model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9280146B2 (en) * | 2011-02-14 | 2016-03-08 | The Mathworks, Inc. | Multi-model, multi-objective tuning of control systems |
-
2014
- 2014-05-09 CN CN201410196572.1A patent/CN103984233B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010068225A1 (en) * | 2008-12-12 | 2010-06-17 | Karem Aircraft, Inc. | Swashplateless rotorcraft with fault tolerant linear electric actuator |
| CN102126554A (en) * | 2011-01-28 | 2011-07-20 | 南京航空航天大学 | Unmanned air vehicle with multiple rotary wings in plane-symmetry layout |
| CN102945311A (en) * | 2012-10-08 | 2013-02-27 | 南京航空航天大学 | Method for diagnosing fault by functional fault directed graph |
| CN103135553A (en) * | 2013-01-21 | 2013-06-05 | 南京航空航天大学 | Four-rotor aircraft fault-tolerant control method |
| CN103149928A (en) * | 2013-03-24 | 2013-06-12 | 西安费斯达自动化工程有限公司 | Fault diagnosing and tolerance control method for aircraft large-angle-of-attack movement ternary number model |
Non-Patent Citations (2)
| Title |
|---|
| 四旋翼飞行器的非线性PID姿态控制;宿敬亚等;《北京航空航天大学学报》;20110930;第37卷(第9期);第1054-1058页 * |
| 旋翼飞行机器人故障诊断与容错控制技术综述;齐俊桐等;《智能系统学报》;20070430;第2卷(第2期);第31-39页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103984233A (en) | 2014-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Potsdam et al. | Computational investigation and fundamental understanding of a slowed UH-60A rotor at high advance ratios | |
| CN109829468B (en) | Bayesian network-based civil aircraft complex system fault diagnosis method | |
| CN111639467B (en) | Aero-engine service life prediction method based on long-term and short-term memory network | |
| CN108196532B (en) | Fault detection and separation method for longitudinal flight control system of unmanned aerial vehicle based on nonlinear adaptive observer | |
| CN103499921A (en) | Fault diagnosis method for variable structure fuzzy system sensor and application thereof in flight control system | |
| CN111680875B (en) | Unmanned aerial vehicle state risk fuzzy comprehensive evaluation method based on probability baseline model | |
| CN111460676B (en) | Method for evaluating health performance of multi-rotor aircraft under atmospheric turbulence disturbance | |
| CN103995529A (en) | On-line diagnosis method for mechanical installation faults of rotor craft | |
| CN104571087B (en) | Spacecraft control diagnosability determination method under a kind of influence of noise | |
| CN114004023A (en) | Aircraft pneumatic parameter identification method based on recurrent neural network | |
| CN103984233B (en) | A kind of quadrotor dual-granularity method for diagnosing faults based on mixed model | |
| Wang et al. | Data-driven anomaly detection of UAV based on multimodal regression model | |
| Jacobellis et al. | Investigation of blade loads on a modern high-speed lift-offset coaxial helicopter using coupled computational fluid dynamics/computational structural dynamics | |
| Gao et al. | UAV sensor fault diagnosis technology: A survey | |
| CN113128047A (en) | Aero-engine thrust calculation method, storage medium and calculation device | |
| Tong et al. | Machine Learning for UAV Propeller Fault Detection based on a Hybrid Data Generation Model | |
| McMillen et al. | Application of an artificial neural network as a flight test data estimator | |
| Battipede et al. | Neural networks for air data estimation: Test of neural network simulating real flight instruments | |
| EP2189962B1 (en) | Method for real-time simulation of a helicopter rotor | |
| Carmichael et al. | An integrated approach to the analysis and design of wings and wing-body combinations in supersonic flow | |
| US20220319257A1 (en) | Aircraft motion observer configured for use in electric aircraft | |
| Bae et al. | PALS-based analysis of an airplane multirate control system in Real-Time Maude | |
| Tran et al. | Developing an approach for fault detection and diagnosis of angular velocity sensors | |
| CN116956471B (en) | Aerodynamic force prediction method, device, equipment and medium for large-scale conveyor | |
| Li et al. | Study on blade fault diagnosis of free flying UAV based on multiparameter and deep learning method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |