CNN-Bagging-based fault diagnosis method for UAV bearing
A fault diagnosis and unmanned aerial vehicle technology, which is applied in computer parts, mechanical parts testing, machine/structural parts testing, etc., can solve the problem that the model accuracy is easily affected by the cardinality, the model training efficiency and fault diagnosis accuracy are low, and there is no Considering the correlation of signal faults and other issues to achieve the effect of improving accuracy and generalization ability, which is conducive to real-time performance and fast training speed
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0048] Example
[0049] For the convenience of description, first explain the relevant professional terms appearing in the specific implementation:
[0050] EMD (Empirical Mode Decomposition): Empirical Mode Decomposition;
[0051] VMD (variational mode decomposition) variational mode decomposition;
[0052] Bagging (bootstrap aggregating): Self-service sampling aggregation;
[0053] CNN: Convolutional Neural Network;
[0054] ANN (Artificial Neutral Network): artificial neural network;
[0055] RNN (Recurrent Neural Networks): Recurrent Neural Network;
[0056] figure 1 It is the flow chart of the UAV bearing fault diagnosis method based on CNN-Bagging of the present invention.
[0057] In this embodiment, as figure 1 As shown, a CNN-Bagging-based UAV bearing fault diagnosis method of the present invention includes the following steps:
[0058] S1, get the signal data set
[0059] Obtain all bearing signals in the UAV to form a signal data set F={f (i) |i∈[1,m]}, f (i) Indicates the signal g...
Example Embodiment
[0096] Instance
[0097] Suppose a UAV has n bearings, f1, f2,..., fn. For f1, the bearing signal of f1 is decomposed into time-domain signal S1 and time-frequency signal F1 based on the variational modal decomposition (VMD), and then the fault diagnosis of f1 is performed through the integrated model CNN. Based on the above process, the fault diagnosis of bearings f2...fn is performed, and the diagnosis process is as follows: Figure 7 As shown, the fault condition of the bearing is finally judged according to the diagnosis result of each bearing
[0098] The model judges the parameters accuracy Acc, precision rate P, recall rate R, F1 and calculation speed and other indicators.
[0099]
[0100]
[0101]
[0102]
[0103] Let S N And F N Indicates the number of time-domain models and time-frequency domain models respectively, and the final test results are shown in Table 1.
[0104] index parameterSN
[0105] Table 1
[0106] Such as Figure 8 As shown, the accuracy of time-domai...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap