American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc.

Abstract

This paper describes results we obtained mathematically modeling vertical flow and heat transfer in geothermal wells. The mathematical model was used to explain observed phenomena and predict performance of wells in various parts of the world performance of wells in various parts of the world including Socal's own geothermal wells.

Equations for heat transfer, vertical pressure drop and pump capacity were combined to simulate the two-phase steady-state flow behaviour of a well producing geothermal water. Studies with a large producing geothermal water. Studies with a large number of wells producing under a variety of conditions showed that the model can accurately match actual field performance.

The modeling approach was found to be useful in the design of the well and its related producing equipment. It can also be used as an operational tool to monitor well and reservoir performance over the life of a geothermal field.

Introduction

The energy crisis and its economic consequences has caused increased attention to be paid to problems relating to geothermal energy development. problems relating to geothermal energy development. In recent years a number of papers have appeared discussing various aspects of the flow of geothermal fluids in wellbores. James discussed factors related to wellbore performance. Dench described well measurement techniques.

Two other papers described calculation techniques. In these steam and water were treated as a well-dispersion mixture of equal velocity. The two-phase friction coefficient was assumed to be constant. Isenthalpic expansion of fluid was assumed such that no heat flow exists between the well fluid and the surrounding rock. The quality (weight percentage) of steam was predicted by relating the percentage) of steam was predicted by relating the temperature of one phase fluid at a selected depth to surface pressures. Since no heat loss was assumed at any flow rate, there was no effect of temperature change on either pressure or quality of the flowing mixture.

The first integrated approach to simulate the combined effects of heat transfer and two phase flow in wells was taken by Gould. Based on a flow regime map and on available data he concluded that Griffith and Wallis/Hagedorn and Brown/Turner Ros formulations are best suited for pressure drop prediction in bubble/slug/annular-mist flow regimes. prediction in bubble/slug/annular-mist flow regimes. He showed the quality profiles in wellbores with time and with heat transfer coefficients as variables.

There exists a multitude of two-phase vertical flow correlations in literature and it is known that each correlation reproduces rather well when applied to data within its experimental domain. It is not our intent to compare several of these with the available data. When the accuracy of the data is subject to question, and with new and modified correlations appearing frequently in literature, we cannot be sure of the outcome from such a comparison.

This content is only available via PDF.
You can access this article if you purchase or spend a download.