Use of Raw Capacitance Tomography Data for Flow Pattern Control

Laurent F.C. Jeanmeure1, Tomasz Dyakowski1, William B.J. Zimmerman2, Glen Baker3

1 Department of Chemical Engineering, University of Manchester Institute of Science and Technology, UMIST, P.O. Box 88, Manchester M60 1QD, England. Laurent.Jeanmeure@umist.ac.uk

2 Department of Chemical and Process Engineering, University of Sheffield, Newcastle Street, Sheffield S1 3JD, England

3 The School of Chemical, Environmental and Mining Engineering, University of Nottingham,

University Park, Nottingham, NG7 2RD, England

ABSTRACT

This paper presents an application of Electrical Capacitance Tomography (ECT) for the monitoring of the flow pattern ahead of a T-junction separator when the gas liquid separation process of an air/kerosene mixture is under investigation. As the oil industry favours the use of pre-calibrated equipment, thus avoiding set-up and calibration overheads; the control system design proposed in this work relies on parameter values obtained from simulation of flow-patterns likely to be encountered on the experimental rig. Speed of decision making is a crucial issue for online monitoring applications. In order to satisfy this concern, the flow pattern classification is achieved without the need for time consuming image reconstruction and analysis. The distinguishable features of a given flow-pattern are extracted straight from the capacitance measurements and translated into pattern information via a set of identification parameters and an associated decision tree. During the course of experimentation, unrelated additional information has been collected on slug formation in pipes by monitoring the online image that has been kept for validation and additional information.

Keywords Flow-pattern, online monitoring, ECT, direct classification