Power Quality Disturbances Classification Based on Modified S Transform and Decision Tree

Abstract:

A novel method for power quality disturbances classification is presented using modified S transform (MST) and decision tree. The time-frequency properties of power quality disturbances are analyzed and the effects of window-wide parameter g on the properties are discussed. Four statistical features are extracted from the MST module time-frequency matrix and a decision tree is utilized to recognize 9 power quality disturbances. The simulations are made to illustrate the validity of the method proposed for power quality disturbances recognition.

• Periodical: Advanced Materials Research (Volumes 860-863)
• Pages: 1891-1894
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.860-863.1891
• Online since: December 2013
• Authors: Ji Liang Yi, Ou Yang Qin
• Keywords: Classification, Feature Extraction, Modified s Transform, Power Quality Disturbances

References

[1] H.S. Behera and P.K. Dash: Applied Soft Computing, Vol. 9 (2009), p.1.

[2] M F Faisal and A Mohamed: Journal of Applied Sciences, Vol. 9(2009), p.2688.

[3] Fengzhan Zhao and Rengang Yang: IEEE Tran. on Power Delivery, Vol. 22 (2007), p.944.

[4] C.R. Pinnegar, L. Mansiha: Fluctuation and Noise Letters, Vol. 3(2003), p.357.

[5] Jiliang Yi: Dissertation of Hunan University, (2010) (In Chinese).

[6] Jiliang Yi and Jianchun Peng: Power System Technology, Vol. 33(2009), p.22 (In Chinese).