Title :
Fault detection using multi-order conditional probability density distribution
Author :
Kumagai, Shusuke ; Takayama, Jim-Ya ; Ohyama, Shinji ; Kobayashi, Akira
Author_Institution :
Dept. of Control & Syst. Eng., Tokyo Inst. of Technol., Japan
Abstract :
The vibration of machines in abnormal condition contains some fault symptoms. In this research, in order to detect the fault as soon as possible, a signal processing method is proposed which focuses on the conditional probability of the fault signal value in the time domain. Furthermore, the proposed method is applied to an experiment, and its validity is confirmed.
Keywords :
fault diagnosis; probability; signal processing; abnormal condition; fault detection; fault signal; fault symptom; fault waveform; multiorder conditional probability density distribution; signal processing method; time domain; Control systems; Fault detection; Gaussian processes; Performance analysis; Shape; Signal analysis; Signal generators; Signal processing; Tiles; Vibration control;
Conference_Titel :
SICE 2002. Proceedings of the 41st SICE Annual Conference
Print_ISBN :
0-7803-7631-5
DOI :
10.1109/SICE.2002.1195723
