Cyclostationarity in electrical resistance tomography data from gas/liquid two-phase flow

Cyclostationarity matches the rhythmic feature of the cycle exciting cycle measurement strategy of electrical resistance tomography (ERT) and thus provides an effective signal processing for ERT data. This paper concerns with the effectiveness and application of cyclostationarity in analysing the ERT data and decribing the characteristic of multiphase flow. To this objective, basic theories on cyclostationarity analysis for signal feature representation are reviewed. Experimental data are obtained from a gas/liquid two-phase flow rig through ERT system. The independent components obtained from ERT data are verified to be second-order cyclostationary signals, whose cyclic autocorrelation functions detect second-order periodicities. It is induced by cycle-exciting-cycle-measurement strategy of adjacent data collection protocol adopted by ERT system. The experiment results have shown that the power spectral of the independent components after the elimination of cyclostationary characteristic can effectively characterize the interface fluctuation of gas/liquid two-phase flow in horizontal pipe.