A robust method for estimating the two-phase flow rate of oil and gas using wellhead data

Ehsan Khamehchi,Mohammad Reza Mahdiani,Mohammad Zolfagharroshan

doi:10.1007/s13202-020-00897-2

Ehsan Khamehchi, Mohammad Reza Mahdiani + Show 1 more

Open Access

https://doi.org/10.1007/s13202-020-00897-2

Copy DOI

Abstract

Flow rate prediction of multiphase flow in the oil and gas wellbores is more complicated than single-phase flow avoiding direct measurements such as using flowmeters or well logging. This study offers an approach to find the accurate two-phase flow rates, applicable in extensive cases of two-phase wells/pipelines. When in a production well, the wellhead data are accessible except for flow rate, and bottom hole conditions, computing the pressure and temperature profiles through the wellbore can be brought about by replacing different values for flow rates, and lead us to probable accurate answers. This aim can be achieved by hiring a heuristic solver to find the most accurate answers as quickly as possible. This approach is flexible and practical depending on the statement of the problem. So, in this study, it has been applied to some vertical two-phase flow wells, which their well survey data was available to avoid future loggings, the wells modeled. Two models were developed, where each one predicted the flow rate by an error of less than 2%. Considering the final results for vertical wells, in this study, the model in which a mechanistic method for predicting pressure gradient applied in proposed compared with experiment-based methods.

Highlights

Accurate rate estimation of two-phase flow wells is one of the crucial concerns of petroleum engineers
Some of them are rule-based like the study Kabir et al (2008) and Izgec et al (2010), which used wellhead pressure and wellhead temperature data, and Hasan and Kabir (2010) or Zolfagharroshan and Khamehchi (2020), which provided methods to model two-phase flow wells
Fluids and well geometry properties and two approximations of pressure and temperature gradient are read by program as inputs, and the program starts searching flow rates

Summary

Introduction

Accurate rate estimation of two-phase flow wells is one of the crucial concerns of petroleum engineers. Some of them are rule-based like the study Kabir et al (2008) and Izgec et al (2010), which used wellhead pressure and wellhead temperature data, and Hasan and Kabir (2010) or Zolfagharroshan and Khamehchi (2020), which provided methods to model two-phase flow wells. There are numerous studies on the ways to model pressure and temperature traverse in two-phase flow wells. Applying artificial intelligence tools like genetic programming, artificial neural network, etc., on big data sets led to strong models, which simulate production conditions like Wang et al (2016), Pouryoussefi and Zhang (2015) or Alizadehdakhel et al (2009)

Objectives

Results

Conclusion