Time series analysis of gamma densitometry signals
Abstract
Single, narrowbeam densitometry has been developed as a method for determining the flow regime and void fraction for industrial liquid gas experiments at high pressures and temperatures in a vertical, thickwalled, steel vessel. To develop suitable techniques, the experimental conditions were simulated using a transparant air/water column. In the transition region from bubbly to slug flow, a time sequence of four regimes, viz. annular flow, partially developed annular flow, coalescing bubble flow and bubbly flow were visually identified in a given cross section. Gamma rays were used to interrogate a column diameter, and digital time series analysis methods were applied. Amplitude spectral densities were used to determine any periodicity in the gas phase flow. The average void fraction for periodic gas flows was obtained by analysis of probability density distributions (PDD). The time sequence of the flow regimes was obtained from the signal magnitude of the diametral void fractions and the time spent in each regime was measured by the associated probability. The results compared well with those obtained from other methods. In the bubbly flow region, the standard deviation of the PDD exceeded that expected for nuclear counting. This bubble noise was assessed with respect to bubble properties.
 Publication:

Nuclear Instruments and Methods in Physics Research A
 Pub Date:
 January 1986
 DOI:
 10.1016/01689002(86)904717
 Bibcode:
 1986NIMPA.242..582L