Pressure Drop in Large Diameter Geothermal Two-Phase Pipelines

Abstract

Most conventional geothermal wells produce two-phase flow of water, steam and gas at the well head. This fluid normally travels at the surface in the geothermal fluid transmission system toward the separator where steam and water are separated. Early transmission system used short runs of small diameter pipelines to avoid the slug flow regime, which caused significant problems of increased pressure drop and water hammer effects to the pipelines. However, significant work was been dedicated over the years to develop and use empirical, theoretical and phenomenological correlations to calculate pressure drop in the two-phase pipelines. However, most of these correlations were developed from work on relatively small diameter pipelines. As most of the new geothermal development uses large diameter pipelines carrying fluid from several wells to centralized separation stations, problems have been encountered in many fields. This work evaluates some of the existing correlations for estimating pressure drop in two-phase geothermal pipelines applied to field data from the Lahendong geothermal field, Indonesia. Actual two-phase pressure drop data were measured from a range of horizontal pipes (10-22“) to be compared with the existing correlations. The result shows that the annular flow is the most common flow regime in the pipelines. Generally, Friedel’s correlation is the best suited for large diameter (>18“) pipelines, with 5–15% average deviation. The homogenous and Harrison-Freeston methods better suit smaller pipe sizes (10”–18”), with an average deviation of 5–28%. The Lockhart-Martinelli, Zhao-Freeston and Brill-Mukherjee correlations were found to be less accurate in predicting the two-phase pressure drop in the same range of pipe sizes. Based on the results, a correction factors chart is proposed to reduce discrepancies in the estimated pressure drops for the given range of pipe sizes.