제어로봇시스템학회:학술대회논문집
- 1992.10b
- /
- Pages.170-175
- /
- 1992
Leak detection in a pipeline based on estimation theory
- Jeong, Sang-Hun (Department of Chemical Engineering, Sogang University) ;
- Bang, Sung-Ho (Department of Chemical Engineering, Sogang University) ;
-
Lee, Kwang-Soon
(Department of Chemical Engineering, Sogang University)
- Published : 1992.10.01
Abstract
A leak detection method for diagnosis of the leak position in a pipeline was developed using an estimation theory with the assumption that the measured flow rates and pressures are stochastic processes. A notch filter was designed using power spectral density analysis of measurements to reduce the effects of disturbances. The noise model dimension was determined by hypothesis testing and then recursive extended least square method was applied to estimate the leak position in real time. The proposed method was applied to an experimental system for evaluation of its performance.
Keywords