CN115356118A - Engine condition maintenance method based on flight parameter data and vibration data - Google Patents
Engine condition maintenance method based on flight parameter data and vibration data Download PDFInfo
- Publication number
- CN115356118A CN115356118A CN202210819751.0A CN202210819751A CN115356118A CN 115356118 A CN115356118 A CN 115356118A CN 202210819751 A CN202210819751 A CN 202210819751A CN 115356118 A CN115356118 A CN 115356118A
- Authority
- CN
- China
- Prior art keywords
- data
- flight parameter
- vibration
- engine
- parameter data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
- G01M15/04—Testing internal-combustion engines
- G01M15/12—Testing internal-combustion engines by monitoring vibrations
Landscapes
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Combustion & Propulsion (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention provides an engine condition maintenance method based on flight parameter data and vibration data, which comprises the steps of S calculating a kurtosis value of the flight parameter data, and carrying out anomaly detection on the flight parameter data; establishing a vibration abnormity detection model according to the vibration data, and performing fault prediction on the vibration data; and establishing a confidence coefficient hypothesis model, taking the abnormal detection result and the fault prediction result as input vectors, outputting the engine health state result, and giving the prediction probability of the engine for maintenance according to the situation. The method designed by the invention effectively solves the problems of low data utilization rate, few statistical analysis frame times, unobvious support and maintenance guarantee effects and the like, can provide powerful support for the engine to change from passive maintenance guarantee to active maintenance guarantee, is beneficial to improving the attendance rate and maintenance efficiency of the airplane, and provides support for scientific and reasonable inspection and maintenance.
Description
Technical Field
The invention relates to the technical field of avionics, in particular to an engine condition maintenance method based on flight parameter data and vibration data.
Background
In the flight process of the airplane, the flight parameter system records engine system parameters and unit operation data, and the flight parameter data contains important operation information of engine operation, which is an important judgment basis for engine safety monitoring.
The engine safety monitoring usually adopts experts to judge the criterion of a single parameter or a few parameter combinations, so that the effect of fault prediction and early warning is achieved.
At present, failure prediction is also performed by means of visual maintenance of an unmanned aerial vehicle, the visual maintenance is a maintenance method which is widely researched in recent years, and based on analysis of failure mechanism, adjustment, maintenance and replacement are performed when a maintenance object has a potential failure according to a result of an undissociated test, so that occurrence of a functional failure is avoided. But the mode of current unmanned aerial vehicle maintenance according to the circumstances is that single non-parameter of using is judged according to or the vibration data, has the judgement inaccuracy, and the confidence coefficient is low problem.
Disclosure of Invention
The invention aims to design an engine visual maintenance method based on flight parameter data and vibration data.
The technical scheme for realizing the purpose of the invention is as follows: an engine condition maintenance method based on flight parameter data and vibration data comprises the following steps:
s1, calculating a kurtosis value of the flight parameter data, and carrying out anomaly detection on the flight parameter data;
s2, establishing a vibration abnormity detection model according to the vibration data, and performing fault prediction on the vibration data;
s3, establishing a confidence coefficient hypothesis model, taking an abnormal detection result and a fault prediction result as input vectors, and outputting an engine health state result;
and S4, according to the engine health state result, giving the prediction probability of the engine maintenance according to the situation.
Further, in step S1, calculating a kurtosis value of the flight parameter data, and performing anomaly detection on the flight parameter data, includes:
s101, calculating a kurtosis value of the flight parameter data of each frame to form the flight parameter data characteristics of the frame;
and S102, based on the dtw algorithm, taking the characteristic of the flight parameter data which is arranged for a plurality of times as an input vector, and carrying out abnormity detection on the flight parameter data.
Further, in step S2, a vibration anomaly detection model is established according to the vibration data, and the fault prediction on the vibration data includes:
s201, extracting vibration data characteristics by adopting a manifold learning model;
s202, establishing a vibration abnormality detection model based on the deep learning DNN model, taking vibration data characteristics as input vectors of the vibration abnormality detection model, and performing fault prediction on the vibration data.
Furthermore, in step S201, the manifold learning model further performs dimension reduction processing on the vibration data, and the extracted vibration data features are the vibration data features after dimension reduction.
Further, in the step S1, the flight parameter data is obtained by preprocessing the original flight parameter data, and in the step S2, the vibration data is obtained by preprocessing the original vibration data, and both the flight parameter data and the vibration data are stored in the distributed server.
Furthermore, the preprocessing of the original flight parameter data and the original vibration data comprises engineering value calculation, null value processing and messy code processing.
Compared with the prior art, the invention has the beneficial effects that: according to the engine condition-based maintenance method based on the flight parameter data and the vibration data, the value information stored in the data is mined, the flight parameter data and the vibration data are combined, powerful support is provided for the engine to be changed from passive maintenance support to active maintenance support, the attendance rate and the maintenance efficiency of the airplane are improved, support is provided for scientific and reasonable inspection and maintenance, and the problems of low data utilization rate, few statistical analysis frames, unobvious support and maintenance support effects and the like can be effectively solved.
Drawings
In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings used in the description of the embodiment will be briefly introduced below.
FIG. 1 is a flow chart of an engine visual maintenance method based on flight parameter data and vibration data according to the present invention.
Detailed Description
The invention will be further described with reference to specific embodiments, and the advantages and features of the invention will become apparent as the description proceeds. These examples are illustrative only and do not limit the scope of the present invention in any way. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention, and that such changes and modifications may be made without departing from the spirit and scope of the invention.
The engine condition-based maintenance method comprises the steps of establishing a storage cluster of flight parameters and vibration data, establishing a distributed computing frame, a deep learning frame and the like, extracting the characteristics of the flight parameters, extracting the abnormal data, carrying out abnormal detection on the flight parameters, establishing a vibration abnormal detection model according to the vibration data, carrying out fault prediction, inputting an abnormal detection result and a fault prediction result into a confidence coefficient hypothesis model, and giving a health state result and a condition-based maintenance prediction probability of the engine.
In the following method, the dtw algorithm is Dynamic Time Warping, namely a Dynamic Time reduction algorithm; DNN, deep neural network, is an open source portal and content management framework.
In the present embodiment, referring to the architecture diagram of the engine on-the-fly maintenance method based on flight reference data and vibration data shown in fig. 1, the method comprises the following steps:
s1, calculating a kurtosis value of the flight parameter data, and carrying out anomaly detection on the flight parameter data;
s2, establishing a vibration abnormity detection model according to the vibration data, and performing fault prediction on the vibration data;
s3, establishing a confidence coefficient hypothesis model, taking an abnormal detection result and a fault prediction result as input vectors, and outputting an engine health state result;
and S4, according to the engine health state result, giving the prediction probability of the engine maintenance according to the situation.
In step S1, calculating a kurtosis value of the flying parameter data, and performing anomaly detection on the flying parameter data includes:
s101, calculating the kurtosis value of the flight parameter data of each frame to form the flight parameter data characteristics of the frame. In the step, important parameters (such as flight parameters recorded in a hydraulic system and an engine system) are selected from flight parameter data files acquired in each frame, kurtosis values of flight parameter data corresponding to the important parameters are calculated, and a data feature matrix is formed according to the kurtosis values calculated in the frame, namely the flight parameter data features of the frame.
And S102, based on the dtw algorithm, taking the characteristic of the flight parameter data which is arranged for a plurality of times as an input vector, and carrying out abnormity detection on the flight parameter data. In the step, an abnormal section related to the engine flight parameter data is found out according to the characteristics of the flight parameter data with multiple frames through a dtw algorithm, and an abnormal detection result is obtained.
In step S2, a vibration anomaly detection model is established according to the vibration data, and the failure prediction of the vibration data includes:
s201, extracting vibration data characteristics by adopting a manifold learning model.
Furthermore, the manifold learning model in this step also performs dimension reduction processing on the vibration data, and the extracted vibration data features are vibration data features after dimension reduction.
S202, establishing a vibration abnormality detection model based on the deep learning DNN model, taking vibration data characteristics as input vectors of the vibration abnormality detection model, and performing fault prediction on the vibration data.
The specific method for establishing the confidence coefficient hypothesis model in the step S3, taking the abnormal detection result and the failure prediction result as input vectors, and outputting the engine health state result includes: and performing hypothesis test on the maintenance judgment result based on the mass historical normal data, wherein the hypothesis test method gives the deviation degree of the data points, namely the confidence coefficient of the final detection result based on Gaussian distribution.
In order to improve the accuracy of the abnormal detection result and the failure prediction result and further accurately calculate the prediction probability of the engine maintenance according to the situation, in an improved embodiment of the invention, the flight parameter data in the step S1 is obtained by preprocessing the original flight parameter data, the vibration data in the step S2 is obtained by preprocessing the original vibration data, and the flight parameter data and the vibration data are both stored in the distributed server. Specifically, the preprocessing of the original flight parameter data and the original vibration data includes engineering value calculation, null value processing, and scrambling code processing, and the preprocessing methods of the engineering value calculation, the null value processing, the scrambling code processing, and the like are conventional general methods, and detailed description thereof will not be provided in this embodiment.
The engine condition maintenance method provided by the specific embodiment utilizes the distributed computing platform to provide the health state assessment of the aircraft engine, is suitable for engine condition maintenance, fully utilizes internal information contained in engine flight parameter data and vibration data compared with the traditional method, can integrate manifold learning and deep learning to comprehensively judge the health state of the engine, and provides a new idea for condition maintenance.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (6)
1. An engine condition maintenance method based on flight parameter data and vibration data is characterized by comprising the following steps:
s1, calculating a kurtosis value of the flight parameter data, and carrying out anomaly detection on the flight parameter data;
s2, establishing a vibration abnormity detection model according to the vibration data, and performing fault prediction on the vibration data;
s3, establishing a confidence coefficient hypothesis model, taking an abnormal detection result and a fault prediction result as input vectors, and outputting an engine health state result;
and S4, according to the engine health state result, giving out the prediction probability of the engine maintenance according to the situation.
2. The engine on-the-fly maintenance method according to claim 1, characterized in that: in the step S1, calculating a kurtosis value of the flight parameter data, and carrying out anomaly detection on the flight parameter data, wherein the method comprises the following steps:
s101, calculating a kurtosis value of the flight parameter data of each frame to form the flight parameter data characteristics of the frame;
and S102, based on a dtw algorithm, taking the characteristics of the flight parameter data with multiple frames as input vectors, and carrying out anomaly detection on the flight parameter data.
3. The engine on-the-fly maintenance method according to claim 1, characterized in that: in step S2, a vibration anomaly detection model is established according to the vibration data, and the failure prediction of the vibration data includes:
s201, extracting vibration data characteristics by adopting a manifold learning model;
s202, establishing a vibration abnormality detection model based on the deep learning DNN model, taking vibration data characteristics as input vectors of the vibration abnormality detection model, and performing fault prediction on the vibration data.
4. The engine on-demand maintenance method according to claim 3, characterized in that: in step S201, the manifold learning model further performs dimension reduction on the vibration data, and the extracted vibration data features are the vibration data features after dimension reduction.
5. The engine on-the-fly maintenance method according to claim 1, characterized in that: in the step S1, the flight parameter data is obtained by preprocessing original flight parameter data, in the step S2, the vibration data is obtained by preprocessing original vibration data, and the flight parameter data and the vibration data are both stored in the distributed server.
6. The engine on-the-fly maintenance method according to claim 5, characterized in that: the preprocessing of the original flight parameter data and the original vibration data comprises engineering value calculation, null value processing and messy code processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210819751.0A CN115356118A (en) | 2022-07-11 | 2022-07-11 | Engine condition maintenance method based on flight parameter data and vibration data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210819751.0A CN115356118A (en) | 2022-07-11 | 2022-07-11 | Engine condition maintenance method based on flight parameter data and vibration data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115356118A true CN115356118A (en) | 2022-11-18 |
Family
ID=84031983
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210819751.0A Pending CN115356118A (en) | 2022-07-11 | 2022-07-11 | Engine condition maintenance method based on flight parameter data and vibration data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115356118A (en) |
-
2022
- 2022-07-11 CN CN202210819751.0A patent/CN115356118A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110377984B (en) | Industrial equipment residual effective life prediction method and system and electronic equipment | |
| CN109783461B (en) | Method for data flow analysis of an aircraft and flight data processing system | |
| US8959065B2 (en) | System and method for monitoring distributed asset data | |
| CN109063308B (en) | Health assessment method based on deep quantum learning | |
| CN111241694B (en) | Big data processing-based aircraft fleet health assessment method, equipment and readable storage medium | |
| US11816590B2 (en) | Monitored machine performance as a maintenance predictor | |
| JP2013507611A (en) | Platform health monitoring system | |
| CN110119787B (en) | Working condition detection method and equipment for rotary mechanical equipment | |
| WO2016059809A1 (en) | Ship propulsion performance prediction device and method, and ship operation assist system | |
| CN113247041B (en) | Wheel set axle box device state monitoring method based on digital twin model | |
| CN110852509A (en) | Fault prediction method and device of IGBT module and storage medium | |
| US20210118242A1 (en) | Predicting a repair and maintenance activity for an aircraft system | |
| CN112816898B (en) | Battery failure prediction method and device, electronic equipment and storage medium | |
| KR102130272B1 (en) | Method for optimizing predictive algorithm based empirical model | |
| Ren et al. | DLformer: A dynamic length transformer-based network for efficient feature representation in remaining useful life prediction | |
| Zheng et al. | An unsupervised transfer learning method based on SOCNN and FBNN and its application on bearing fault diagnosis | |
| CN115356118A (en) | Engine condition maintenance method based on flight parameter data and vibration data | |
| Cerquitelli et al. | Enabling predictive analytics for smart manufacturing through an IIoT platform | |
| CN116933656A (en) | Ship power system fault diagnosis method based on deep learning | |
| CN116521406A (en) | Method for detecting anomaly of non-overrun flight parameter data of aero-engine based on residual gate GRU-VAE model | |
| CN111079227A (en) | Health assessment modeling method for spacecraft thermal control function system | |
| CN116910574A (en) | Mechanical equipment diagnosis method and device based on meta-learning and time convolution network | |
| Luo et al. | A novel method for remaining useful life prediction of roller bearings involving the discrepancy and similarity of degradation trajectories | |
| Krupp et al. | A hybrid framework for bearing fault diagnosis using physics-guided neural networks | |
| CN113807704A (en) | Intelligent algorithm platform construction method for urban rail transit data |
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 |