Non-Stationary Signal Analysis Using Time Frequency Transform
In this paper, visual detection and classification of the non-stationary power signals are demonstrated by well-known transform called generalized synchrosqueezing transform.The wavelet based time frequency representation gives poor quality and understandability, hence the proposed synchrosqueezing transform is an effective method to get better quality and readability of the wavelet-based TFR by summarizing along the frequency axis. Different feature vectors have been extracted from the frequency contour of the generalized synchrosqueezing transform and these feature vectors applied as input to the Reformulated Fuzzy C-Means algorithm for automatic classification.
KeywordsContinuous wavelet transform (CWT) Generalized synchrosqueezing transform (GST) Reformulated fuzzy C-Means algorithm
- 1.Li, C., Liang, M.: A generalized synchrosqueezing transform for enhancing signal time frequency representation. Signal Process. 92(9), 2264−2274 (2012)Google Scholar
- 6.Biswal, B., Biswal, M., Mishra, S., jalaja, R.: Automatic classification of power quality disturbances with balanced neural tree” IEEE. Trans. Ind. Electron. 61(1), 521–530 (2014)Google Scholar
- 7.Biswal, B., Mishra, S.: Detection and classification of disturbances in non-stationary signals using modified frequency slice wavelet transform”. IET Gener. Transm. Distrib. 7(9), (2013)Google Scholar
- 9.Daubechies, I., Maes, S.: A nonlinear squeezing of the continuous wavelet transform based on auditory nerve models. In: Aldroubi, A., Unser, M. (eds.) Wavelets in Medicine and Biology, pp. 527−546. CRC Press, Boca Raton (1996)Google Scholar
- 10.Biswal, B., Dash, P.K., Panigrahi, B.K.: Power quality disturbance classification using Fuzzy C-means algorithm and adaptive particle swarm optimization. IEEE Trans. Ind. Electron. 56(1), 212–220 (2009)Google Scholar