Title :
Algorithmic method of analysis of time series data for definition of prognostic parameters of engine fault
Author :
Kirillov, Alexandr ; Kirillova, Olga ; Kirillov, Sergey
Author_Institution :
GRIDMotors Lab., SmartTech Appl R&D Center, Vitebsk, Belarus
Abstract :
The paper describes an algorithmic method for predicting the failure of internal combustion engines. The method is based on the construction of discrete series of sequences of wavelet coefficients of vibration signals and their analysis in terms of Kolmogorov complexity and entropy. The main principles of the computing program of preventive diagnostics are discussed.
Keywords :
acoustic signal processing; data analysis; entropy; failure analysis; fault diagnosis; internal combustion engines; mechanical engineering computing; time series; vibrations; wavelet transforms; Kolmogorov complexity; algorithmic method; computing program; discrete series; engine fault; entropy; failure prediction; internal combustion engines; preventive diagnostics; prognostic parameters; time series data analysis; vibration signals; wavelet coefficient sequence; Engines; Gold; Zirconium; Kolmogorov complexity; engine; entropy; fault prognosis; preventive diagnostics;
Conference_Titel :
Advanced Computer Control (ICACC), 2011 3rd International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4244-8809-4
Electronic_ISBN :
978-1-4244-8810-0
DOI :
10.1109/ICACC.2011.6016384
