A novel integral extension LMD method based on integral local waveform matching. Issue 3 (April 2016)

• Record Type: Journal Article
• Authors: Liu, W.; Zhou, L.; Hu, N.; He, Z.; Yang, C.; Jiang, J.
• Is Part Of: Neural computing & applications. Volume 27:Issue 3(2016)
• Page Start: 761
• Page End: 768
• Publication Date: 2016-04
• Subjects: Integral extension -- Integral local waveform matching -- Local mean decomposition (LMD) -- End effect; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-1894-z
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract This paper proposed a novel integral extension local mean decomposition (IELMD) method, which can be widely applied in wind turbine fault diagnosis. Firstly, the characteristic waveform and its corresponding integral values are calculated. Then, the similar waveform is established according to signal extreme points and their integral values. The similar waveform and characteristic waveform are compared in order to obtain an optimal waveform, which can match the characteristic waveform in the best way. Finally, the optimal waveform is used to extend the left side and right side of the original signal. According to the simulation experimental analysis, the novel IELMD method is reasonable and effective in suppressing LMD end effect, and this method can be applied in wind turbine fault diagnosis.