Screening Criteria to Adopt The Best Multiphase Flow Correlation
Ramana Palisetti, Petroleum Information/Dwights LLC & Lloyd Heinze, Texas Tech University
Multiphase flow in pipes is defined as concurrent movement of free gases and liquids in the pipes. Flow may be in any direction. The gas and liquid may exist as a homogeneous mixture, or the liquid may be in slugs with gas displacement (which is pushing behind slugs). The liquid and gas may flow parallel to each other, or other combinations of flow patterns may be present. The gas may be flowing with two liquids (normally oil and water), and the possibility exists that the two liquids may be emulsified. The prediction of pressure gradients occurring during the simultaneous flow of gas and liquid in pipes is necessary for the proper tubing size selection, design of artificial lift installations and many other production systems in the petroleum and chemical industries. Petroleum engineers encounter multiphase flow more frequently in well tubing and flowlines. The ability to accurately and analytically predict the pressure at any point in a flow string is essential in determining optimum production string dimensions and in the design of gas-lift and other kinds of production equipment installations. This information is invaluable for predicting bottomhole pressure in flowing wells. As with any correlation, the correlations developed are often misused and applied to cases outside the range of the database from which it was developed. Even though the range of the correlation's application can be extrapolated, it must be used with caution. Hence, a decision has to be made as to which correlation should be used to suit the given set of well data. The importance of being able to assess the accuracy of calculating methods or previously developed correlations is demonstrated in this paper. In fact, their range of validation in the light of the variety of conditions is discussed. These set of tested ranges are used as tools for obtaining a criteria in order to determine the suitability of different correlations towards the given data. This paper is an extraction of work done in relation to the masters" thesis by Palisetti.