The University of Tulsa  |  McDougall School of Petroleum Engineering  |  Contact TUALP


TUALP Abstract
 

 

 

  

Michinori Miyoshi
Offshore Pipeline Field Data Analysis

This report summarizes the results of an analysis performed on the Texaco Denmark Skjold data.

Comprehensive pressure and temperature data were collected for various flow rates at Skjold , Gord ' D', and Gorm 'C'. The objective of this study was to perform the following analysis:

  1. Compare the calculated and measured pressure drop data using the Beggs and Brill correlation and the Dukler-Eaton correlation.

  2. Calculate the temperature profile using the compositional model and the Beggs and Brill and Dukler-Eaton pressure drop correlations, and to compare the model predictions with the measured data.

  3. Calculate the pressure fluctuations caused by normal slug flow and compare these to the measured pressure fluctuations.

  4. Determine whether severe slug flow ( Geometry dominated slug flow) or normal slug flow exist at flow rates at 3000 B/D, as indicated by the study. In addition to determine the boundary between the normal and severe slug flow for the Skjold-Gorm 'D' pipeline.

  5. Recommend a procedure for the use of the existing control valve or orifice choke to minimize the detrimental effect of slug flow on the separator.

The results of the analysis indicate that pipeline pressure losses were over-predicted by about 30 percent and temperature losses were under-predicted by about 20 percent.

The slug length analysis determined the existence of extremely long (roughly 5 to 6 times as long) liquid slugs which could not be explained by existing correlations. However, a modification to an existing correlation was proposed, based on the simple procedure of interpolating for pipeline diameter effects and extrapolating for pipeline length effects. The agreement between the predictions of this modified correlation and the measured data was reasonably good.

As a result of the flow pattern analysis , we concluded that severe slug flow did not occur under test conditions. However, severe slugging may be experienced when the flow rate falls below 2,000 BBL/D. In that event, installation of a choke was found to be a most effective means of eliminating severe slugging.

Finally, a simple hydrodynamic model was developed which explains the pressure fluctuation observed at the top of the riser pipe.

Download report (TUALP members only)
Comments are welcome
   

Last updated September 04, 2012