DocumentCode :
3354083
Title :
Improved Algorithm of Scaleless Band Identification on Correlation Dimension of Vibration Signal
Author :
Du Bi-qiang ; Deng Gui-fu ; Shi Jun-jie
Author_Institution :
Mech. Eng. Dept., North China Electr. Power Univ., Baoding
fYear :
2009
fDate :
27-31 March 2009
Firstpage :
1
Lastpage :
4
Abstract :
An improved algorithm by curve-line-curve fitting piecewise is proposed for fractal scaleless band automatic identification, which is based on the study of the correlation integral algorithm and the distribution rule of the correlation integral series for vibration signal. To get the correlation dimension of vibration signal, a curve-line-curve fitting piecewise on the double logarithmic point series of correlation integral is processed firstly, with a target of least fitting residual square sum, and the middle line section can be treated as scaleless band, thereby the correlation dimension of signal can be got by that line section´s slope. The method was applied to some real vibration signals, which were tested in different operation conditions and different rotate speeds. Finally the region of fractal scaleless band and the correlation dimension of signal are made automatically, which showed the effectiveness of improved algorithm. Meanwhile the result was compared with the calculation results by conventional three-line fitting method, while the comparison shows that the method is more precise.
Keywords :
acoustic signal processing; curve fitting; mechanical engineering computing; vibrations; correlation dimension; correlation integral; correlation integral algorithm; curve-line-curve fitting piecewise; double logarithmic point series; fractal scaleless band automatic identification; least fitting residual square sum; scaleless band identification; three-line fitting method; vibration signal; Curve fitting; Delay effects; Fault diagnosis; Fractals; Mechanical engineering; Power supplies; Power systems; Signal analysis; Signal processing; Vibrations;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Power and Energy Engineering Conference, 2009. APPEEC 2009. Asia-Pacific
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-2486-3
Electronic_ISBN :
978-1-4244-2487-0
Type :
conf
DOI :
10.1109/APPEEC.2009.4918419
Filename :
4918419
Link To Document :
بازگشت