SOFTWARE-BASED PIPELINE LEAK DETECTION* Presented by: Miguel J. Bagajewicz, James Akingbola**, Elijah Odusina** and David Mannel** University of Oklahoma School of Chemical, Biological and Material Engineering Pipeline leak detection has been a focus of numerous researches in industry. There are several methods based on both the expensive hardware and the inadequate software techniques. Instrument biases greatly affect the leak detection methodology, leading to high rate of false alarms or omitted leaks. As an alternative, a less costly software based method is being proposed. The method takes into account the effects of errors in the form of leaks and instrument biases on the system. It makes use of the measured flows and pressures to infer through data reconciliation and bias detection methodologies whether a leak or a bias is present. In this report, the Generalized Likelihood Ratio (GLR) method proposed by Narasimhan and Mah (1987) is adapted to combine leak detection and instrument bias identification. The methodology is
This is the end of the preview. Sign up
access the rest of the document.
This note was uploaded on 08/31/2011 for the course CHE 4273 taught by Professor Staff during the Spring '10 term at Oklahoma State.