Model, analysis debottleneck complicated field gathering system

Ken J. Vargas
Falcon EDF Ltd.
Calgary

Modern predictive computer models and analysis methods can aid in debottlenecking a complicated oil production facility to find the maximum oil that can flow under multiphase limitations through a facility's flow lines.

In the case examined, the gathering flow line system had been modified and expanded numerous times without consideration for operability and fully integrated operations. Consequently, a good understanding of how best to operate the field was lacking.

The study reviewed all facilities, obtained well production data, interviewed operations personnel, and updated all well, manifold, and flow line maps and process flow drawings. From the information gathered, a methodology was developed comprising the following steps:

Develop forms to conduct gathering system surveys.
Take pressure surveys at predetermined flowing conditions.
Match the survey conditions with multiphase flow simulations.
Assemble optimum gathering system flows.
Prepare a report with recommended operating parameters such as how the flow lines should operate.

Gathering system

Fig. 1 [40707 bytes] shows the wells and flow lines available to produce the field. The wells are either produced into a high or low-pressure system.

The advantage of producing wells at higher pressure is to reduce gas breakout. However, at lower operating pressure on the wellhead, more oil can be produced. Thus, the strategy was to find the maximum oil flow given the multiphase limitations through the flow lines.

Fig. 1 shows test/group lines and individual flow lines to the battery. The lines of interest are in the following categories:

Dashed lines-Dedicated flow lines from wells to battery.
Dotted lines-Flow lines from wells to satellite
Solid thin lines-Test lines from satellites to battery
Solid thick lines-Group lines from satellites to battery.

The field is broken down into north and south portions. The demarcation is an east-west line cutting the battery (Fig. 1).

The north section consists of one satellite at Well 14-3 and wells and flow lines to the battery. The satellite has a 6-in. group line. and a 4-in. test line from the satellite to the battery.

The south section consists of three satellites at Wells 9-25, 1-36, and 16-33 and associated flow lines. The southernmost satellite at Well 9-25 has one 8-in. group line to the next satellite at Well 1-36 and one 4-in. test line running directly to the battery.

The middle satellite has one 8-in. group line into the next satellite at Well 16-33 and a dedicated 4-in. test line to the battery.

The closest satellite to the battery at Well 16-33, again has an 8-in. line into the battery, together with dedicated 4-in. test lines. All 8-in. group lines have common flow to the battery.

The well test data (Fig. 2 [72277 bytes]) show how all the wells tie into the satellites and the battery and include the latest well test data.

The manifolds of the typical satellite and battery are shown in the manifold sketches of Fig. 3 [79088 bytes]. The manifold sketches have been simplified, because the actual battery manifold required four 11 x 17-in. isometric drawings, to define the flows completely.

Engineering information

To start the analysis, all available field and battery drawings were compiled. After receiving the drawings, field visits verified the accuracy of the drawings and determined the new drawings required. A complete set of the following drawings were produced:

Satellite and battery manifold process and instrumentation drawings (P&IDs)
Battery production flow diagram (PFD) with flow rates and line sizes and equipment capacities
Flow line maps with line sizes and designated line use.

Production data

During the field visit, all the most current well test data, throughputs, various historical production records, and plant operator interviews were gathered. The production data were organized in a three-ring binder and collated into convenient forms.

The wells' plant log (Fig. 2), shows the most current well test data.

Pressure surveys

All active well sites, satellites, flow line intermediate points, and inlet manifold pressure gathering locations were visited with a plant operations representative. These visits confirmed the presence of a nipple with an isolating valve for recording pressures at each required point.

If no fitting was found, hot taps were made to install the appropriate fitting and valve to take pressure surveys. All pressure data points were identified and tagged to expedite the field pressure survey gathering trips. One pressure gauge was kept for checking the consistency of pressure readings.

Survey checklist

A survey checklist form eased data collection. The forms showed all wells, manifolds, and flow lines in a logical and realistic configuration.

Fig. 2 shows a sample form filled out with the pressure line data in the first column.

Study methodology

The methodology used to perform the debottlenecking study comprised the following six steps:

Build a multiphase model on the computer, using Neotechnology's Pipeflo program. The model for this gathering system was made up of the following legs:
- A 4-in. individual high-pressure line of varying lengths to model all individual wells going into the battery. These wells feed a high-pressure separator, connected to a 500 psig (suction) compressor.
- A 1.023 mile, 6-in. group line to model all flows from the north. This group line runs at a low pressure of about 150 psig.
- A 1.971 mile, 8-in. group line with side-stream flows at each of the satellites to collect flows at each satellite before entering the battery. Satellite inlet pressure is about 150 psig
- A 4-in. individual, low pressure, about 150 psig, line of varying lengths to model all lower pressure wells flowing individually into the battery.
Table 1 [74309 bytes] gives a sample printout of the 8-in. line.
The data used for the line models was organized as well flow from the latest well test and pressures from operations and survey logs.
Perform pressure surveys for different production scenarios. The most important item to remember is to try and load all the lines as much as possible. In this case, operations felt that the 8-in. group line was the key; therefore, this line was loaded up as much as possible. Five different data sets were obtained, under different field conditions (Table 1 [74309 bytes]).
After building the models and obtaining significant pressure data, collate the data and run the models with flows from well tests and pressures from the survey.
Run computer simulations and organize by survey data and line diameter.
Match simulation results with actual plant flows for the corresponding survey data. Once a match is obtained, the models are validated and one can proceed to predict flows.
Make computer runs to determine the maximum possible flows, given the conditions found in the surveys.

Results

In the case examined, the simulation determined the following:

For the 4-in., high-pressure wells flowing in dedicated high-pressure lines into the battery, all high-pressure lines must enter the battery at 500 psig. Only one well was included for predicting the pressure drop, Well 11-25, because it is the furthest from the battery and has the severest multiphase flows in the field. The program predicted an 8 psi/1,000 ft pressure drop.
For the 4-in., low-pressure wells flowing in low pressure dedicated lines into the battery, the field pressures and the computer-simulated pressures could not be matched. The reason for this was that probably all low-pressure wells were tested at much higher pressures than the actual operating pressures.
A match was obtained after doubling the gas and oil flows. The conclusion was that the wells could produce twice as much as during the flow tests.
For the 6-in. group line from the satellite at Well 14-3, this line was never at full capacity because the north sector of the field has ample flow line capacity.
Two computer cases predicted the pressure drop with all possible wells flowing through this group line. Using a starting pressure of 210 psig, as recorded in the surveys, the runs predicted a total flow of 642 b/d into the battery at a 147 psig inlet.
For the 8-in. group line, the biggest potential bottleneck, a careful match was obtained for all survey results (Table 1).
Predicted results from the well test were 4,378 bo/d, 7,718 bw/d, and 59 MMscfd. The flow line model predicted 4,013 bo/d, 7,076 bw/d, and 69 MMscfd.
The gas-handling capacity at the battery was thus identified as the production bottleneck.
After compression was installed at the battery, the total oil produced was 3,900 b/d. Hence, a good prediction was made, (3,900/4,013 = 97%).

The Author

Ken J. Vargas is the president of Falcon EDF Ltd., an engineering consulting company in Calgary. His specialties are process and mechanical design and project management. Vargas has a BS from the U.S. Air Force Academy. He is a member of ASME and is a registered professional engineer in Canada and the U.S.