| Autor: |
Ma Liang, Ping Yu, Cong Zhihui, Li Shuo, Jinshan Wang |
| Rok vydání: |
2021 |
| Předmět: |
|
| Zdroj: |
2021 International Conference on Applications and Techniques in Cyber Intelligence ISBN: 9783030791964 |
| DOI: |
10.1007/978-3-030-79197-1_25 |
| Popis: |
This paper proposed and developed fault detection and diagnosis solution based on the ensemble empirical mode decomposition and support vector machine (SVM). The collected field signals are used as the input signals that are decomposed into a series of intrinsic mode functions (IMF) by using the empirical mode decomposition. The dominant eigenmode functions are then selected in the proposed solution and their approximate entropy and energy ratio are calculated as signal features. Finally, the feature vector of the signals is used as the input into the support vector machine for model training to classify different mechanical faults of the wind turbine gearbox. The developed algorithmic solution is extensively verified through a set of experiments and the numerical results confirmed the effectiveness of the proposed solution. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
