CN108205310A - Gas path failure recognition methods in a kind of aero-engine envelope curve based on ELM filtering algorithms - Google Patents
Gas path failure recognition methods in a kind of aero-engine envelope curve based on ELM filtering algorithms Download PDFInfo
- Publication number
- CN108205310A CN108205310A CN201810013686.6A CN201810013686A CN108205310A CN 108205310 A CN108205310 A CN 108205310A CN 201810013686 A CN201810013686 A CN 201810013686A CN 108205310 A CN108205310 A CN 108205310A
- Authority
- CN
- China
- Prior art keywords
- elm
- filtering algorithm
- engine
- sample
- weight 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.)
- Granted
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims description 44
- 239000013598 vector Substances 0.000 claims description 40
- 239000011159 matrix material Substances 0.000 claims description 27
- 238000012360 testing method Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 3
- 238000013095 identification testing Methods 0.000 claims description 3
- 238000011423 initialization method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 9
- 238000003909 pattern recognition Methods 0.000 abstract description 4
- 230000036541 health Effects 0.000 abstract description 2
- 238000012423 maintenance Methods 0.000 abstract description 2
- 230000001737 promoting effect Effects 0.000 abstract 1
- 241000051616 Ulmus minor Species 0.000 description 9
- 230000006870 function Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000009530 blood pressure measurement Methods 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses gas path failure recognition methods in a kind of aero-engine envelope curve based on ELM filtering algorithms, this method includes：ELM model topology parameters are trained using Kalman filtering algorithm；By the trained ELM models of filtering algorithm for gas path failure identification in engine envelope curve.The present invention is solved in existing envelope curve in Fault Diagnosis of Aircraft Engine Gas Path, the engine diagnosis of conventional data-driven is not strong in different operating point generalization ability, the problem of precision is not high, suitable for the engine failure pattern-recognition of the different operating point in flight envelope, play the role of actively promoting for engine health control, reduction maintenance cost.
Description
Technical Field
The invention belongs to the technical field of aeroengine gas circuit fault diagnosis, and particularly relates to an aeroengine envelope gas circuit fault identification method based on an ELM filtering algorithm.
Background
The aircraft engine is used as the heart of the aircraft, and has a complex structure and a severe working environment. The engine fault diagnosis technology is an important means for ensuring the performance and reliability of the engine and reducing the use and maintenance cost. During the service life of an aircraft engine, the performance of the components can slowly degrade. Furthermore, sudden changes in component health parameters may also occur. Gas path component failures affect aircraft engine performance and reliability and require diagnostics. An Extreme Learning Machine (ELM) is a fast Learning method developed on the basis of a neural network theory, and is widely applied to the fields of data mining, pattern recognition and the like. The essence of ELM is a Single hidden layer Feed-forward neural network (SLFNs). The traditional feedforward neural network consists of an input layer, a hidden layer and an output layer. The ELM model is different from the traditional SLFNs in that the input weight and the bias corresponding to the ELM are generated through random assignment, so that a linear system with fixed parameters is formed, and the linear system is solved by a least square method. Huang et al have verified through theory and extensive simulation experiments that ELM is an efficient and effective learning algorithm.
The ELM algorithm also has some defects, and the way of randomly generating input weight and bias brings certain randomness to the ELM, and different initialization parameters can cause different learning effects. And the least square algorithm has low precision, so that the stability and generalization capability of the ELM model are influenced. One of the methods to remedy this deficiency is to use the idea of combinatorial neural networks, but such network structures would be redundant with increasing computational complexity. Obtaining the optimal hidden node parameters by using a genetic algorithm or a cross-validation method is also a measure for improving the stability of the ELM, but the algorithms need a large amount of iterative computation, so that the learning efficiency is not high.
The kalman filter algorithm is a recursive form of linear minimum variance estimation. In stochastic estimation theory, the linear minimum variance estimate is the best of all linear estimates. Based on the Kalman filtering algorithm, the traditional ELM solving algorithm is replaced by a filtering recursion estimation method, all training samples only need one filtering iteration to obtain the output weight of the ELM, and the stability of the ELM model is improved. On the basis, the invention provides an aeroengine gas circuit fault identification method based on an ELM filtering training algorithm, and the identification precision of the fault modes of the engine at different working points in a flight envelope is improved.
Disclosure of Invention
Aiming at the technical problems, the invention provides the aeroengine envelope inner gas circuit fault identification method based on the ELM filtering algorithm, aiming at the instability of the learning effect caused by the randomness of the ELM network topology parameters, the Kalman filtering algorithm is adopted to replace the traditional least square solving method, and the stability of the ELM model is effectively improved. The identification method based on the ELM filtering training algorithm is designed aiming at the fault modes of different working points of the aircraft engine in the envelope, and the identification precision of the fault modes of the different working points of the engine in the flight envelope is effectively improved.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the technical scheme that:
an aeroengine envelope inner gas circuit fault identification method based on an ELM filtering algorithm comprises the following steps:
step 1) training an ELM (ensemble empirical mode model) by adopting a Kalman filtering algorithm according to fault mode data of boundary points of an engine in a flight envelope;
and 2) using the ELM model trained by the filtering algorithm for identifying the fault modes of other common working points in the envelope.
Further, the step 1) of training the ELM model by using a Kalman filtering algorithm according to the fault mode data of the boundary point of the engine in the flight envelope specifically comprises the following steps:
step 1.1) carrying out standardization processing on fault data of an engine model at boundary points to obtain a training sample, wherein the fault mode data consists of measurement parameters of each sensor;
step 1.2) the input layer weight and hidden layer bias of the ELM model are initialized randomly, the hidden layer activation function selects a linear correction function ReLU, the linear correction function ReLU is substituted into the training sample obtained in the step 1.1, and the hidden layer output vector is calculated;
and step 1.3) iteratively updating the ELM model output weight vector by adopting a Kalman filtering algorithm.
Further, the step 1.3) of iteratively updating the output weight vector of the ELM model by using a kalman filter algorithm specifically comprises the following steps:
step 1.3.1), the output weight vector β is randomly initialized to a small random variable with a zero mean value and uniform or normal distribution;
step 1.3.2), the output vector error variance matrix initialization method comprises the following steps:
C＝ξI
wherein, I is a unit matrix, C is an output weight vector error variance matrix, ξ is an artificially set decimal number, and the range is [0.001,0.1 ];
step 1.3.3) substituting into training sample { xk,ykWhere x is the input data and y is the desired output. Calculating a one-step predicted value of the weight vector corresponding to the kth sample according to the discrete state equation of the ELM modelAnd output layer node outputWherein, the ELM model discrete state equation is written as:
wherein h iskOutputting vectors, ω, for hidden layer nodesTAnd upsilonTThe method is characterized in that the method is artificially introduced process noise and observation noise vectors which are both white noise;
and step 1.3.4) utilizing a Kalman filtering algorithm principle to iteratively update the output weight vector.
Further, the step 1.3.4) of utilizing the kalman filter algorithm principle to iteratively update the output weight vector specifically includes the following steps:
step 1.3.4.1) calculating a filter gain matrix GkThe concrete formula is as follows:
Gk＝Ck-1hk·[R+hk TCk-1hk]-1
wherein, Ck-1An error variance matrix estimation value corresponding to the (k-1) th sample, and R is a symmetrical positive definite variance matrix of corresponding observation noise;
step 1.3.4.2) updating the weight vector estimation value and the error variance matrix estimation value corresponding to the kth sample, wherein the specific formula is as follows:
wherein,is the weight vector estimation value corresponding to the (k-1) th sample,is the weight vector estimation value corresponding to the kth sample; ck-1Is the error variance matrix estimate, C, for the k-1 samplekIs the error variance matrix estimate corresponding to the kth sample.
Step 1.3.4.3) iteratively updates until all sample training is complete.
Further, the specific step of applying the ELM model trained by the filtering algorithm to the fault mode recognition of other common working points in the envelope in the step 2) is as follows:
step 2.1), carrying out standardized processing on fault data of other common working points of the engine in an envelope to obtain a test sample, wherein the fault mode data consists of measurement parameters of each sensor;
and 2.2) substituting the standardized fault data into the ELM model trained by the filter algorithm to carry out engine fault mode identification test.
Has the advantages that: 1. the method for identifying the gas circuit fault in the aircraft engine envelope based on the ELM filtering algorithm solves the problem that the fault diagnosis precision in the flight envelope is low in the traditional data-based fault diagnosis method. Compared with the common ELM, the KF-ELM algorithm improves the accuracy and ensures the stability and generalization capability of the model. On the basis, the method for identifying the fault of the aero-engine can well identify the fault mode of the aero-engine gas circuit in the flight envelope.
Drawings
FIG. 1 is a flow chart of an aeroengine gas circuit fault identification method based on an ELM filter training algorithm.
FIG. 2 is a cross-sectional view of a turbofan engine.
FIG. 3 is a strategy for selecting reference training points and test verification points within a flight envelope.
FIG. 4 is a schematic diagram comparing the filter training algorithm (KF-ELM) of the extreme learning machine with the classification Accuracy (Accuracy) of the conventional ELM algorithm on the UCI standard classification problem data set.
FIG. 5 is a comparison of classification accuracy of two algorithms against the trend of training sample number and hidden node number on a Page data set.
FIG. 6 is a comparison of classification accuracy of two algorithms against the trend of training sample number and hidden node number on a vessel data set.
Detailed Description
The following further describes embodiments of the present invention with reference to the drawings.
The invention discloses an aeroengine envelope inner gas circuit fault identification method based on an ELM filtering algorithm, which specifically comprises the following steps:
step 1) training an ELM (ensemble empirical mode model) by adopting a Kalman filtering algorithm according to fault mode data of boundary points of an engine in a flight envelope;
step 1.1) carrying out standardization processing on fault data of an engine model at boundary points to obtain a training sample, wherein the fault mode data consists of measurement parameters of each sensor;
step 1.2) the input layer weight and hidden layer bias of the ELM model are initialized randomly, the hidden layer activation function selects a linear correction function ReLU, the linear correction function ReLU is substituted into the training sample obtained in the step 1.1, and the hidden layer output vector is calculated;
and step 1.3) iteratively updating the ELM model output weight vector by adopting a Kalman filtering algorithm.
Step 1.3.1), the output weight vector β is randomly initialized to a small random variable with a zero mean value and uniform or normal distribution;
step 1.3.2), the output vector error variance matrix initialization method comprises the following steps:
C＝ξI
wherein, I is a unit matrix, C is an output weight vector error variance matrix, ξ is an artificially set decimal number, and the range is [0.001,0.1 ];
step 1.3.3) substituting into training sample { xk,ykWhere x is the input data and y is the desired output. Calculating a one-step predicted value of the weight vector corresponding to the kth sample according to the discrete state equation of the ELM modelAnd output layer node outputWherein, the ELM model discrete state equation is written as:
wherein h iskOutputting vectors, ω, for hidden layer nodesTAnd upsilonTThe method is characterized in that the method is artificially introduced process noise and observation noise vectors which are both white noise;
and step 1.3.4) utilizing a Kalman filtering algorithm principle to iteratively update the output weight vector.
Step 1.3.4.1) calculating a filter gain matrix GkThe concrete formula is as follows:
Gk＝Ck-1hk·[R+hk TCk-1hk]-1
wherein, Ck-1An error variance matrix estimation value corresponding to the (k-1) th sample, and R is a symmetrical positive definite variance matrix of corresponding observation noise;
step 1.3.4.2) updating the weight vector estimation value and the error variance matrix estimation value corresponding to the kth sample, wherein the specific formula is as follows:
wherein,is the weight vector estimation value corresponding to the (k-1) th sample,is the weight vector estimation value corresponding to the kth sample; ck-1Is the error variance matrix estimate, C, for the k-1 samplekIs the error variance matrix estimate corresponding to the kth sample.
Step 1.3.4.3) iteratively updates until all sample training is complete.
And 2) using the ELM model trained by the filtering algorithm for identifying the fault modes of other common working points in the envelope.
Step 2.1), carrying out standardized processing on fault data of other common working points of the engine in an envelope to obtain test data, wherein the fault mode data consists of measurement parameters of each sensor;
and 2.2) substituting the standardized fault data into the ELM model trained by the filter algorithm to carry out engine fault mode identification test.
In order to verify the effectiveness of the aircraft engine gas circuit fault diagnosis algorithm based on the ELM filtering training algorithm, digital simulation of gas circuit fault identification in an envelope is carried out in an MATLAB environment.
Firstly, the ELM filtering training algorithm provided by the invention is verified on a UCI standard data set. In order to show the effectiveness and the innovation of the algorithm, the algorithm is compared with the common ELM algorithm and the BP neural network. Data description and simulation results are shown in table 1, table 2 and fig. 4. FIGS. 5 and 6 are graphs that verify the stability of both algorithms by varying the number of training samples and the number of hidden layer nodes on the Page and vessel data sets. As can be seen, the ELM filter training algorithm (KF-ELM) is superior to the common ELM in precision. And the KF-ELM training time is significantly reduced compared to BPNN. This shows that the KF-ELM algorithm proposed by the invention has advantages in precision compared with the common ELM and is obviously improved in training time compared with BPNN.
TABLE 1UCI Standard Classification problem data set description
TABLE 2ELM vs KF-ELM training time and test time comparisons on Classification problems
Aiming at the problem of identifying the fault modes of different working points of the aircraft engine in the envelope, firstly 8 reference training points are selected in the flight envelope, and the selection strategy and the specific description of the training points and the test points are shown in fig. 3 and table 4. It can be seen that the training points are all located on the boundary of the flight envelope, and the test points are some typical common operating points of the engine within the envelope. At the selected operating point, 13 failure modes of the turbofan engine are injected (as shown in table 3), resulting in data samples for training consisting of various sensor measurement parameters, including mainly low pressure rotation speed NL, high pressure rotation speed NH, and various cross-sectional pressure and temperature measurements (T22, P22, T3, P3, T43, P43, T6, P5). The working section of the turbofan engine gas circuit applied to the embodiment of the invention is shown in figure 2. After the training samples are obtained, an ELM model is trained by adopting a Kalman filtering algorithm. After the model training is completed, the fault mode data of the selected 8 training points and 4 test points are respectively substituted into the trained model to perform fault mode identification, the identification results are shown in tables 5, 6, 7 and 8, and tables 9 and 10 are comparison of the average identification accuracy of the two algorithms at each working point.
TABLE 3 turbofan Engine failure mode description
TABLE 4 description of training and test points within the flight envelope
TABLE 5 failure mode recognition results (correct number/total number of samples) for ELM at training points
TABLE 6KF-ELM Fault Pattern recognition results at training points (correct number/total number of samples)
TABLE 7 failure mode identification results (correct number/total number of samples) for ELM at test points
TABLE 8KF-ELM Fault Pattern recognition results at test points (correct number/total number of samples)
TABLE 9 comparison of average recognition accuracy of ELM and KF-ELM at training points
TABLE 10 comparison of average identification accuracy of ELM and KF-ELM at test points
As can be seen from tables 9 and 10, the KF-ELM method provided by the present invention improves the average recognition accuracy of the failure modes of each operating point compared with the common ELM. The average recognition precision of the KF-ELM algorithm at 8 training points is 91.12%, and the average recognition precision of the common ELM algorithm is 88.25%; the average identification precision of KF-ELM is 86.82% and the average identification precision of common ELM is 83.25% at 4 test points. The KF-ELM algorithm provided by the invention has the advantages and better generalization capability compared with the common ELM algorithm, and can better meet the requirement of the fault mode identification work of the engine in the flight envelope.
The method for identifying the gas circuit fault in the aircraft engine envelope based on the ELM filtering algorithm solves the problem that the fault diagnosis precision in the flight envelope is low in the traditional data-based fault diagnosis method. Compared with the common ELM, the KF-ELM algorithm improves the accuracy and ensures the stability and generalization capability of the model. On the basis, the method for identifying the fault of the aero-engine can well identify the fault mode of the aero-engine gas circuit in the flight envelope.
It should be noted that the above mentioned embodiments are only specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any changes and substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (5)
1. An aeroengine envelope inner gas circuit fault identification method based on an ELM filtering algorithm is characterized in that: the method comprises the following steps:
step 1) training an ELM (ensemble empirical mode model) by adopting a Kalman filtering algorithm according to fault mode data of boundary points of an engine model in a flight envelope;
and 2) using the ELM model trained by the filtering algorithm for fault mode recognition of other working points in the flight envelope.
2. The aeroengine on-line air circuit fault identification method based on the ELM filtering algorithm as claimed in claim 1, wherein: the method comprises the following specific steps of training an ELM model by using fault mode data of boundary points in a flight envelope in the step 1) and adopting a Kalman filtering algorithm:
step 1.1) carrying out standardization processing on fault mode data of a boundary point of an engine model in an envelope line to obtain a training sample, wherein the fault mode data consists of measurement parameters of each sensor;
step 1.2) the input layer weight and hidden layer bias of the ELM model are initialized randomly, the hidden layer activation function selects a linear correction function ReLU, the linear correction function ReLU is substituted into the training sample obtained in the step 1.1, and the hidden layer output vector is calculated;
and step 1.3) iteratively updating the ELM model output weight vector by adopting a Kalman filtering algorithm.
3. The aeroengine on-line air circuit fault identification method based on the ELM filtering algorithm as claimed in claim 2, wherein: the step 1.3) of iteratively updating the ELM model output weight vector by adopting a Kalman filtering algorithm comprises the following specific steps:
step 1.3.1), the output weight vector β is randomly initialized to a small random variable with a zero mean value and uniform or normal distribution;
step 1.3.2), the output vector error variance matrix initialization method comprises the following steps:
C＝ξI
wherein, I is a unit matrix, C is an output weight vector error variance matrix, ξ is an artificially set decimal number, and the range is [0.001,0.1 ];
step 1.3.3) substituting into training sample { xk,ykWhere x is the input data and y is the desired output. Calculating a one-step predicted value of the weight vector corresponding to the kth sample according to the discrete state equation of the ELM modelAnd output layer node outputWherein, the ELM model discrete state equation is written as:
wherein h iskOutputting vectors, ω, for hidden layer nodesTAnd upsilonTThe method is characterized in that the method is artificially introduced process noise and observation noise vectors which are both white noise;
and step 1.3.4) utilizing a Kalman filtering algorithm principle to iteratively update the output weight vector.
4. The aeroengine on-line air circuit fault identification method based on the ELM filtering algorithm as claimed in claim 3, wherein: the step 1.3.4) of utilizing the Kalman filtering algorithm principle to iteratively update the output weight vector specifically comprises the following steps:
step 1.3.4.1) calculating a filter gain matrix GkThe concrete formula is as follows:
Gk＝Ck-1hk·[R+hk TCk-1hk]-1
wherein, Ck-1An error variance matrix estimation value corresponding to the (k-1) th sample, and R is a symmetrical positive definite variance matrix of corresponding observation noise;
step 1.3.4.2) updating the weight vector estimation value and the error variance matrix estimation value corresponding to the kth sample, wherein the specific formula is as follows:
wherein,is the weight vector estimation value corresponding to the (k-1) th sample,is the weight vector estimation value corresponding to the kth sample; ck-1Is the error variance matrix estimate, C, for the k-1 samplekIs the error variance matrix estimation value corresponding to the kth sample;
step 1.3.4.3) iteratively updates until all sample training is complete.
5. The aeroengine on-line air circuit fault identification method based on the ELM filtering algorithm as claimed in claim 1, wherein: the specific steps of applying the ELM model trained by the filtering algorithm to the fault mode recognition of other common working points in the envelope in the step 2) are as follows:
step 2.1), carrying out standardization processing on fault mode data of other working points of the engine model in an envelope to obtain a test sample, wherein the fault mode data consists of measurement parameters of each sensor;
and 2.2) substituting the standardized fault data into the ELM model trained by the filter algorithm to carry out engine fault mode identification test.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810013686.6A CN108205310B (en) | 2018-01-08 | 2018-01-08 | Aeroengine envelope internal gas circuit fault identification method based on ELM filtering algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810013686.6A CN108205310B (en) | 2018-01-08 | 2018-01-08 | Aeroengine envelope internal gas circuit fault identification method based on ELM filtering algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108205310A true CN108205310A (en) | 2018-06-26 |
| CN108205310B CN108205310B (en) | 2020-03-17 |
Family
ID=62606108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810013686.6A Active CN108205310B (en) | 2018-01-08 | 2018-01-08 | Aeroengine envelope internal gas circuit fault identification method based on ELM filtering algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108205310B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109163911A (en) * | 2018-09-21 | 2019-01-08 | 昆明理工大学 | A kind of fault of engine fuel system diagnostic method based on improved bat algorithm optimization ELM |
| CN109635318A (en) * | 2018-11-01 | 2019-04-16 | 南京航空航天大学 | A kind of aero-engine sensor intelligent analytic redundancy design method based on KEOS-ELM algorithm |
| CN110084324A (en) * | 2019-05-10 | 2019-08-02 | 杭州电子科技大学 | Kalman filtering parameter adaptive update method based on extreme learning machine |
| CN110672328A (en) * | 2019-11-05 | 2020-01-10 | 大连理工大学 | Turbofan engine health parameter estimation method based on random configuration network |
| CN112364446A (en) * | 2020-09-30 | 2021-02-12 | 南京航空航天大学 | Engine whole performance attenuation prediction method based on EC-RBELM algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103983453A (en) * | 2014-05-08 | 2014-08-13 | 南京航空航天大学 | Differentiating method of fault diagnosis of executing mechanism of aeroengine and sensor |
| CN104008432A (en) * | 2014-06-03 | 2014-08-27 | 华北电力大学 | Micro-grid short-term load forecasting method based on EMD-KELM-EKF |
| CN104200062A (en) * | 2014-08-04 | 2014-12-10 | 南京航空航天大学 | Aircraft engine gas path fault fusion diagnosis method |
| CN107045575A (en) * | 2017-04-14 | 2017-08-15 | 南京航空航天大学 | Aero-engine performance model modelling approach based on self-adjusting Wiener model |
-
2018
- 2018-01-08 CN CN201810013686.6A patent/CN108205310B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103983453A (en) * | 2014-05-08 | 2014-08-13 | 南京航空航天大学 | Differentiating method of fault diagnosis of executing mechanism of aeroengine and sensor |
| CN104008432A (en) * | 2014-06-03 | 2014-08-27 | 华北电力大学 | Micro-grid short-term load forecasting method based on EMD-KELM-EKF |
| CN104200062A (en) * | 2014-08-04 | 2014-12-10 | 南京航空航天大学 | Aircraft engine gas path fault fusion diagnosis method |
| CN107045575A (en) * | 2017-04-14 | 2017-08-15 | 南京航空航天大学 | Aero-engine performance model modelling approach based on self-adjusting Wiener model |
Non-Patent Citations (4)
| Title |
|---|
| 尤成新等: "航空发动机传感器信号重构的K-ELM方法", 《航空动力学报》 * |
| 许梦阳等: "用于航空发动机动态辨识的MSMEA-ELM算法", 《传感器与微系统》 * |
| 陶金伟等: "航空发动机传感器故障隔离与气路健康估计方法", 《自动化应用》 * |
| 顾嘉辉等: "航空发动机健康估计的神经网络修正卡尔曼滤波算法", 《推进技术》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109163911A (en) * | 2018-09-21 | 2019-01-08 | 昆明理工大学 | A kind of fault of engine fuel system diagnostic method based on improved bat algorithm optimization ELM |
| CN109635318A (en) * | 2018-11-01 | 2019-04-16 | 南京航空航天大学 | A kind of aero-engine sensor intelligent analytic redundancy design method based on KEOS-ELM algorithm |
| CN109635318B (en) * | 2018-11-01 | 2023-07-25 | 南京航空航天大学 | Intelligent analysis redundancy design method for aero-engine sensor |
| CN110084324A (en) * | 2019-05-10 | 2019-08-02 | 杭州电子科技大学 | Kalman filtering parameter adaptive update method based on extreme learning machine |
| CN110084324B (en) * | 2019-05-10 | 2021-05-04 | 杭州电子科技大学 | Kalman filtering parameter self-adaptive updating method based on extreme learning machine |
| CN110672328A (en) * | 2019-11-05 | 2020-01-10 | 大连理工大学 | Turbofan engine health parameter estimation method based on random configuration network |
| CN110672328B (en) * | 2019-11-05 | 2020-08-14 | 大连理工大学 | Turbofan engine health parameter estimation method based on random configuration network |
| CN112364446A (en) * | 2020-09-30 | 2021-02-12 | 南京航空航天大学 | Engine whole performance attenuation prediction method based on EC-RBELM algorithm |
| CN112364446B (en) * | 2020-09-30 | 2023-05-12 | 南京航空航天大学 | Engine whole-engine performance attenuation prediction method based on EC-RBELM algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108205310B (en) | 2020-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108205310B (en) | Aeroengine envelope internal gas circuit fault identification method based on ELM filtering algorithm | |
| CN108416086B (en) | Aero-engine model self-adaptive correction method based on deep learning | |
| CN108647428B (en) | Turbofan engine adaptive component level simulation model construction method | |
| US6539783B1 (en) | Methods and apparatus for estimating engine health | |
| CN108829928B (en) | Turboshaft engine adaptive component-level simulation model construction method | |
| WO2019144337A1 (en) | Deep-learning algorithm-based self-adaptive correction method for full-envelope model of aero-engine | |
| CN109162813B (en) | One kind being based on the modified Aeroengine Smart method for controlling number of revolution of iterative learning | |
| CN109635318B (en) | Intelligent analysis redundancy design method for aero-engine sensor | |
| CN109612738B (en) | Distributed filtering estimation method for improving gas path performance of turbofan engine | |
| CN109800449B (en) | Neural network-based aeroengine compression component characteristic correction method | |
| CN112364446B (en) | Engine whole-engine performance attenuation prediction method based on EC-RBELM algorithm | |
| WO2021088160A1 (en) | Stochastic configuration network-based health parameter estimation method for turbofan engine | |
| CN107357176B (en) | Modeling method for test run data of aero-engine | |
| CN111581763B (en) | Method for evaluating air path fault diagnosis result of aero-engine | |
| CN110516395A (en) | A kind of aero-engine control method based on non-linear mould predictive | |
| EP1656597A1 (en) | Intelligent database for performance predictions | |
| CN115081484A (en) | Aircraft engine sensor fault diagnosis method based on CRJ-OSELM algorithm | |
| CN114815619A (en) | Robot tracking method based on Kalman filtering under random measurement loss | |
| CN110985216B (en) | Intelligent multivariable control method for aero-engine with online correction | |
| Huang et al. | Gas path deterioration observation based on stochastic dynamics for reliability assessment of aeroengines | |
| CN110084324B (en) | Kalman filtering parameter self-adaptive updating method based on extreme learning machine | |
| CN112801267A (en) | Multiple fault diagnosis device for aircraft engine with dynamic threshold value | |
| CN111289231A (en) | Rotor system health monitoring method and system based on incomplete B-spline data fitting | |
| Zhu et al. | Application of adaptive square root cubature Kalman filter in turbofan engine gas path performance monitoring | |
| CN112749789A (en) | Aero-engine multiple fault diagnosis device based on self-association neural network |
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 |