Multimodal Tomometry for Slug Detection in two Phase Flow

Abstract

Electrical Capacitance and Resistance Tomometric (ECTm/ ERTm) approaches are attractive for measurement and control applications in the process industries. By using the time series of raw resistance and capacitance values R(i,j,t) and C(i,j,t) ( i=1, …, NR, j = 1, 2, …, NR and i ≠ j for ERT modules with NR electrodes; i=1, …, NC, j = 1, 2, …, NC and i ≠ j for ECT modules with NC electrodes), it is shown in this paper (with NR=16,NC=12) how the intermittent occurrence of slugs in a liquid/gas two phase flow can be detected using the time series data consisting of the measured resistances and capacitances R(i,j,t) and C(i,j,t). In a pipe transporting oil and gas, the slugs can be characterized by their sizes, occurring instants and frequencies. The signal processing employed to estimate these parameters associated with the slugs is based on the analysis of the pulse generated from the ECT/ERT module, both in time and frequency domains. Some measurements are also performed using high speed camera, thus making this effort multimodal. Due to the very much reduced estimation times involved in this process as compared to tomographic image processing, this tomometric method has many advantages in real time measurement and control applications. Signal processing of the volume ratio data of tomography systems and image analysis of high speed camera images can be done to extract various slug parameters. Slug length, velocity and frequency in a slightly inclined upward two phase water/air flow are estimated using fusion of data from the multimodal system. The results obtained are compared to verify the various parameters estimated using different modalities.