Identification of flow regimes using back-propagation networks trained on simulated databased on a capacitance tomography sensor

Abstract

Non-invasive techniques such as electrical capacitance tomography (ECT) are beginning tomake promising contributions to control systems and are well fitted for flow-regimeidentification in opaque pipes or conduits. A new method of two-component flow-regimeidentification based on a neural network and an eight-electrode ECT sensor is proposed inthis paper. Time-consuming image reconstruction and analysis are avoided. Ten featureparameters are extracted straight from the capacitance measurements and translated intoregime information via a back-propagation (BP) network. The extraction of featureparameters, the architecture and the training of the BP network are given. Simulationresults show that the new identification method has good precision and high speed. Theuse of feature parameters and the BP network for flow-regime identification ispromising.