1. Application of Wavelet Transform in Intermittent Fault Fearture Extraction
- Author
-
Li Huakang, Kehong Lv, Guanjun Liu, and Jing Qiu
- Subjects
0209 industrial biotechnology ,Computer science ,Feature extraction ,0211 other engineering and technologies ,Hardware_PERFORMANCEANDRELIABILITY ,02 engineering and technology ,Fault (power engineering) ,Signal ,Physics::Fluid Dynamics ,Computer Science::Hardware Architecture ,020901 industrial engineering & automation ,Wavelet ,Computer Science::Operating Systems ,Computer Science::Distributed, Parallel, and Cluster Computing ,021103 operations research ,business.industry ,ComputerSystemsOrganization_COMPUTER-COMMUNICATIONNETWORKS ,Process (computing) ,Wavelet transform ,Pattern recognition ,Intermittent fault ,Nonlinear Sciences::Chaotic Dynamics ,ComputerApplications_GENERAL ,Artificial intelligence ,Transient (oscillation) ,business ,human activities - Abstract
For intermittent fault diagnosis, it is important to extract fault features. An intermittent fault feature extraction method based on wavelet transform is proposed. Firstly, the stress-intensity model is used to describe the fault process of intermittent faults, analysing the main part of intermittent fault signal characteristics. Considering the randomness and suddenness of intermittent faults, wavelet transform is used to extract intermittent fault features based on the advantages of wavelet transform in signal mutation and singularity detection. By extracting signal energy through the wavelet coefficients, intermittent fault features are obtained, which can be used to identify and isolate the intermittent faults. Finally, by a simulating circuit, transient intermittent fault response characteristics are obtained by simulation in the circuit, and the features are extracted by wavelet transform for the intermittent fault location. Results show that the intermittent fault features can be extracted by the wavelet transform, and the features can be used for intermittent fault diagnosis.
- Published
- 2018