Accuracy of droplet models for liquid loading prediction: Analysis of production well parameters

Daniyar Bopbekov, Peyman Pourafshary*, Randy Hazlett

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Liquid loading is a common problem in gas wells. Gas field operators are interested in knowing liquid loading onset time in advance since it helps to minimize liquid removal costs and reduce production loss. Knowing critical gas velocity is crucial for the accurate prediction of liquid loading. Analytical models for critical velocity estimation are based on either film or droplet reversal theories, and droplet models are easier to use. However, the accuracy of droplet models is not consistent across various datasets. Available studies showed that the accuracy of droplet models depends on wellhead pressure and producing gas-water ratio. This paper tests seven droplet models with an extensive well databank collected from several publications. Models are tested for various wellhead pressure and gas-water ratio ranges. Paper selects best and worst models for several ranges of wellhead pressure and gas-water ratio, and recommendations are based on the accuracy of the models in a given range. It is found that the accuracy of the models in wellhead pressures below 600 psi does not change with the gas-water ratio. In this wellhead pressure range, the same recommendations apply for the entire gas-water ratio range (0–1000 Mscf/scf). In higher wellhead pressure ranges, the accuracy of models significantly changes with the gas-water ratio, and it is essential to know the gas-water ratio for model selection. It is also found that all studied models are accurate in wells with wellhead pressure above 2000 psi and a gas-water ratio above 100 Mscf/scf. The results in studied parameter ranges prove that selecting an appropriate model strongly depends on the operational parameters of the gas well. The results of the study and recommendations are verified by a newly acquired gas-condensate field data set.

Original languageEnglish
Article number104391
JournalJournal of Natural Gas Science and Engineering
Volume98
DOIs
Publication statusPublished - Feb 2022

ASJC Scopus subject areas

  • Fuel Technology
  • Geotechnical Engineering and Engineering Geology
  • Energy Engineering and Power Technology

Fingerprint

Dive into the research topics of 'Accuracy of droplet models for liquid loading prediction: Analysis of production well parameters'. Together they form a unique fingerprint.

Cite this