Title :
A new failure detection method and its application in leak monitor of pipeline
Author :
Li, Junhua ; Liu, Wenbai ; Sun, Zhaochen ; Cui, Li
Author_Institution :
Coll. of Ocean Environ. & Eng., Shanghai Maritime Univ., Shanghai
Abstract :
A new failure detection method for abrupt change based on ESPRT (extended sequential probability ratio test) and innovation theory is presented in this paper. This method realizes non-parameters detection by utilizing ESPRT to detect the innovation process generating in the innovation model, which enlarge the applicability of ESPRT. The paper establishes the innovation model by employing the nonlinear time series based on BP neural network and the leakage detection model of long pipe. The validity of the proposed method is verified by experiments. The results indicate that this method can detect leakage in real time.
Keywords :
backpropagation; failure analysis; leak detection; neurocontrollers; pipelines; statistical testing; time series; BP neural network; extended sequential probability ratio test; failure detection; innovation model; innovation theory; leak monitor; leakage detection model; nonlinear time series; nonparameters detection; pipeline; Automatic control; Computerized monitoring; Condition monitoring; Leak detection; Pipelines; Robot control; Robot vision systems; Robotics and automation; Sequential analysis; Technological innovation; innovation; leakage detection; pipeline; sequential test;
Conference_Titel :
Control, Automation, Robotics and Vision, 2008. ICARCV 2008. 10th International Conference on
Conference_Location :
Hanoi
Print_ISBN :
978-1-4244-2286-9
Electronic_ISBN :
978-1-4244-2287-6
DOI :
10.1109/ICARCV.2008.4795688
