Physics-Inspired Data-Driven Method Manages Liquid Loading
Chris Carpenter_
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper IPTC 24769, “Managing Liquid Loading Using a Physics-Inspired Data-Driven Method,” by Prithvi Singh Chauhan and Zhuoran Li, SPE, Xecta Digital Labs, and Wenyue Sun, SPE, China University of Petroleum, et al. The paper has not been peer reviewed. Copyright 2025 International Petroleum Technology Conference.
_
Liquid accumulation in gas wells can result in frequent well shut-ins and reduced productivity, necessitating regular cleanups or lift interventions to minimize disruptions. This work introduces a fast and methodical approach to detect liquid loading using easily available field data, avoiding traditional assumptions. It also determines critical gas rates directly from observed field data and provides a mechanism to recommend and confirm the best artificial lift settings to address liquid-loading problems.
Introduction
As shown in Fig. 1, wells experience a series of flow patterns throughout their life cycle. A vertical gas well may include the following flow regimes:
- Mist flow occurs at high gas velocities and low liquid content.
- Annular flow involves liquid forming a film along the well walls.
- Churn flow involves larger, intermittent liquid surges.
- Slug flow is common at lower gas velocities and often results in fluctuating pressure and flow rates.
Liquid loading typically initiates during the transition from churn flow to slug flow when gas velocity drops, and the inability of the gas to fully entrain the liquid marks the onset of accumulation. In bubble flow, where gas is dispersed within the liquid, the well is typically severely loaded with liquids, indicating a need for more aggressive intervention.
Several liquid-loading-mitigation strategies are used currently. However, successful design and operation of these strategies relies on selecting the right wells with proper understanding of expected critical rates at the right time.
In this work, the authors focus on well cycling, which involves intentionally shutting in the well for a finite time period to allow pressure to build up near the wellbore as a result of recharge from the reservoir. Well cycling is a cost-effective strategy that leverages the natural pressure dynamics of the reservoir, avoiding more-complex and -expensive solutions.
Methodology
The liquid-loading detection, prediction, and optimization workflow is divided into the following four main stages:
- Liquid-loading detection uses frequency-based analysis to determine the onset of liquid loading.
- Critical-rate correction uses a data-driven method to correct the critical rate to explain the pattern of liquid loading.
- Liquid-loading prediction combines material-balance-based reservoir depletion and well-nodal analysis to predict time to liquid load and reservoir recovery.
- Well-cycling optimization uses well shut-in and production durations to maximize cumulative production for active cycling wells.