Title :
The improved grey correlation fault diagnosis for oil pumping well
Author :
Xiufang Wang ; Chuang Guan ; Running Gao
Author_Institution :
Sch. of Electr. Eng. & Inf., Northeast Pet. Univ., Daqing, China
Abstract :
Grey correlation fault diagnosis of oil pumping well though indicator diagram is the simple and practicable method at present. But it is easy to lead misjudgment of some failures whose geometric characteristics are similar. In order to solve this problem, weighted entropy is introduced into this paper. In this method weighted value is calculated by the information entropy of each fault feature vector, so the minor differences are enlarged, which can reduces the rate of mistakes. It is proved that through field experimental this method can effectively reduce the misjudgment of similar failures, and it can also improve the diagnostic accuracy.
Keywords :
condition monitoring; entropy; fault diagnosis; fuel pumps; geometry; grey systems; oil technology; diagnostic accuracy; failure misjudgment; fault feature vector; geometric characteristics; grey correlation fault diagnosis; indicator diagram; information entropy; oil pumping well; weighted entropy; weighted value; Extraterrestrial measurements; Gray-scale; Pistons; Valves; Fault feature vector; Grey correlation; Indicator diagram; Information entropy;
Conference_Titel :
Measurement, Information and Control (ICMIC), 2013 International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4799-1390-9
DOI :
10.1109/MIC.2013.6758170