An abnormal detection method for helicopter maneuvering parts based on generative confrontation network
A component abnormality and detection method technology, applied in the direction of biological neural network model, neural learning method, registration/indicating vehicle operation, etc., can solve problems such as poor engineering applicability, dependence on abnormal judgment threshold setting, and high demand for fault sample data. Achieve the effect of reducing requirements, improving data utilization, and reducing false alarm rate
Active Publication Date: 2021-12-07
BEIHANG UNIV
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The problem to be solved by the present invention is that the current abnormality detection method of helicopter moving parts has the disadvantages of high requirements on the amount of fault sample data, the setting of the abnormality judgment threshold depends on expert experience, and it is difficult to update the threshold online with the use of the model. The problem of poor applicability in actual engineering
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0081] In the present invention, the effectiveness of the method is verified by taking the helicopter flight data actually measured by the HUMS system as an example. The data sampling rate is 32768, the sampling duration is 2 seconds, and contains 65536 points in total. The vibration sensor is arranged on the helicopter accessory casing, wherein the normal data is collected when the helicopter is healthy, and the abnormal data is collected when the tail rotor output shaft of the helicopter fails.
[0082] Step 1. Helicopter health data preprocessing S1
[0083] figure 2 It is the complete original vibration signal sample diagram used in the embodiment of the present invention, figure 2 Display the raw data measured by HUMS for a total of 2 seconds, the flight vibration signal, the horizontal axis is the number of data points, and the vertical axis is the amplitude of the vibration signal.
[0084] image 3 The original sample diagram of a single vibration signal obtained...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a method for abnormal detection of helicopter moving parts based on a generative confrontation network. The method comprises the following steps: step 1, preprocessing of helicopter health data; step 2, generating confrontation network design and components; step 3, generating confrontation network unsupervised training; step 4, alarm threshold setting; step 5, abnormality detection; step 6, threshold adaptive online adjustment; the present invention can make full use of the health data of massive helicopter moving parts, and use the generative confrontation network to perform unsupervised learning on the distribution of health data, Improve data utilization, set anomaly detection alarm thresholds without a large number of abnormal samples, reduce the requirements for data collection, and adjust the abnormality detection alarm thresholds online as the model is used and combined with continuously replenished healthy and abnormal samples , in order to meet the needs of abnormal detection, reduce the false alarm rate and improve the detection rate.
Description
technical field [0001] The invention relates to the technical field of abnormal detection of helicopter moving parts, in particular to a method for detecting abnormalities of helicopter moving parts based on a generative confrontation network. Background technique [0002] Due to its unique performance and wide range of uses, helicopters have received widespread attention in various fields. The transmission system is a basic and important key system in the helicopter, which undertakes the power transmission function of the helicopter. The structural feature of the helicopter transmission system is that there are a large number of rotating parts. The power output from the engine is transmitted to the main rotor, tail rotor and other helicopter subsystems through the transmission system. The power or load transmission path is single, and the rotation is always maintained during flight. Abnormal function or failure of any one of the components may lead to catastrophic accident...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G07C5/08G06N3/04G06N3/08
CPCG07C5/0808G06N3/088G06N3/045
Inventor 程玉杰马梁丁宇李商羽陶来发马剑吕琛
Owner BEIHANG UNIV