Experimental Study on the Effect of Liquid Properties on Fluid Behavior and Liquid Loading in Vertical Natural Gas Wells

Liquid loading is one of the major challenges for natural gas wells, as it can decrease and eventually cease the production of hydrocarbons. Characterizing liquid loading in gas wells and quantifying flow dynamics throughout this process is crucial. The literature shows a lack of experiments for oil-air low liquid loading flow, especially in vertical pipes. This demonstrates a lack of experimental data to comprehend the effect of liquid properties on two-phase flow behavior. In order to simulate a vertical natural gas well, oil-air tests were carried out in this work in a vertical pipe with a 0.0508-m ID (Inner-Diameter) under low liquid loading conditions. The aim of this study is to better comprehend the effect of liquid characteristics on flow behavior and the onset of liquid loading. The results of the experiments reveal that changing the liquid phase from water to oil reduces pressure loss, liquid holdup, and the surface gas velocity (vSg) at which liquid loading occurs. Liquid holdup and pressure loss of the water-air and oil-air flow with medium gas rates show notable variances within the churn flow. Additionally, compared to the minimum-pressure drop technique, the positive pressure technique offers considerably more reliable predictions for the start of liquid loading. Two models-the unified Tulsa University Fluid Flow Projects (TUFFP) and OLGA-are used to investigate the flow pattern, liquid holdup, and pressure loss predictions. The liquid film reverse, liquid droplet models, and the inflection point technique were compared to experimental data for similar liquid loading onset models by Alsanea et al. (2022). For the data in this investigation, the OLGA model and Coleman's correlation perform the best at predicting the onset of liquid loading. Given that both the unified model and OLGA use slug flow models, the results show that a separate model must be created for churn flow to accurately estimate the pressure drop and liquid holdup.