Wavelet Denoising of Vehicle Platform Vibration Signal Based on Threshold Neural Network

Gespeichert in:

Bibliographische Detailangaben
Zeitschriftentitel:	Shock and Vibration
Personen und Körperschaften:	Li, Mingzhu, Wang, Zhiqian, Luo, Jun, Liu, Yusheng, Cai, Sheng
In:	Shock and Vibration, 2017, 2017, S. 1-12
Format:	E-Article
Sprache:	Englisch
veröffentlicht:	Hindawi Limited
Schlagwörter:	Mechanical Engineering Mechanics of Materials Geotechnical Engineering and Engineering Geology Condensed Matter Physics Civil and Structural Engineering

Details
Zusammenfassung:	<jats:p>Vehicle Platform Vibration Signal (VPVS) denoising is essential to achieve high measurement accuracy of precise optical measuring instrument (POMI). A method to denoise the VPVS is proposed based on the wavelet coefficients thresholding and threshold neural network (TNN). According to the characteristics of VPVS, a novel thresholding function is constructed, and then its optimized threshold is selected through unsupervised learning of TNN. The original VPVS mixed in trend and random noise is constructed as VPVS model. A VPVS denoising flow is proposed based on the power spectral and energy distribution of the VPVS model. The simulation shows that the proposed denoising method achieves better results, compared to the previous denoising methods using the indexes of SNR and RMSE. The experiment demonstrates that it is efficient for denoising VPVS polluted by the trend and random noise.</jats:p>
Umfang:	1-12
ISSN:	1070-9622 1875-9203
DOI:	10.1155/2017/7962828