Paper Number 101872-MS
DOI  What's this?10.2118/101872-MS
Title

Well Test by Design: Transient Modelling to Predicting Behaviour in Extreme Wells

Authors

D. Teng, SPE, and B. Maloney, SPE, Woodside Energy Ltd., and J.C. Mantecon, SPE, Scandpower Petroleum Technology

Source

SPE Asia Pacific Oil & Gas Conference and Exhibition, 11-13 September 2006, Adelaide, Australia

Copyright

2006. Society of Petroleum Engineers

Preview

Abstract

Offshore rig rates are at an all time high and wells are becoming bigger and longer, in deeper waters and in more complex reservoirs. Well testing in this environment has become more challenging, where well clean-up and flow assurance issues such as slugging and hydrates can significantly extend the planned duration of well tests. The ability to predict and being prepared to deal with such problems by appropriate design of well test equipment can reduce operational risk, minimise safety hazards and environmental impact and potentially save millions of dollars in rig-time.
Traditional well flow software only models steady-state flow. Predicting the transient behaviour of wells, from the unloading of completion fluids until steady state flow conditions are reached, requires specialised software. Dynamic flow simulation software is a proven tool which has been applied for years by facilities engineers for pipeline and slug-catcher design, but its application for well testing is a new practice. Key outputs from dynamic well simulation include slug sizes and frequency, fluid composition and pressure-temperature trends at any time and at any point in the well. Such information enables optimum design so all parameters are within the equipment’s allowable operating envelope at any time of the well test operation.
This paper describes how dynamic simulation, using the software package OLGA, was applied to a big-bore gas well with 9 5/8” production tubing. The dynamic simulation study:

  1. provided better understanding of well unloading behaviour at different flowrates (by using different choke sizes).
  2. assessed the effect of an emergency shut-down (ESD) during a well clean-up operation.
  3. defined the minimum flowrate required to cleanup the well (for sizing well testing equipment).

Results from the dynamic simulation indicated that a standard well test package may be adequate for cleaning up this big-bore gas well with 9 5/8” production tubing, though the equipment would be operated at or near its limits and would take quite some time for clean-up. A significantly faster
clean-up could be achieved with a high rate well test package at additional cost.

Case Study

The Thylacine and Geographe gas fields are located in the Otway basin, 70 km and 55km from the Victorian coast in South-eastern Australia - Fig.1. Discovered in 2001, gas from these two fields is expected to supply a total of 950 billion cubic feet of raw gas into the domestic market (equivalent to 885 petajoules of sales gas, 12.2 million barrels of condensate and 1.7 million tonnes of LPG).
The first phase of the development will tap into the Thylacine field with four wells from an unmanned platform in 100 m of water. Gas will be sent via a 20” pipeline to a newly built processing plant near Port Campbell - Fig.2. The Geographe field, located 15km north of Thylacine, will be connected by subsea pipelines to the main pipeline in a later development phase.
The joint venture partners in this development are:

  • Woodside Energy Ltd (Operator) 51.55%
  • Origin Energy Resources 30.75%
  • Benaris International NV 12.7%
  • CalEnergy Gas Australia 5.0%

Well description

TM-1 is a big-bore well with a 9 5/8” production tubing – Fig.3. The well is vertical until ~650 m, where it kicks off at a tangent, intersecting the reservoir at a 31o angle. Well depth is about 2600 m measured depth (mMD) or 2300 m true vertical depth (mTVD). Reservoir temperature is ~120oC.

Models Description

The TM-1 well model used to run the simulations was built using the multiphase flow simulator OLGA. The key model building considerations are well geometries, wall materials and layers, fluid PVT and boundary conditions (reservoir pressure and wellhead backpressure).
Clean-up simulations were started from the initial underbalanced conditions using different choke sizes. The well models were allowed to run until the brine, diesel and mud had been displaced from the well and steady state conditions were reached. Mud was included in the model to assess the effect of backproducing mud lost into the formation during drilling.
Details are described in the following sections.

Number of Pages 14
File Size 2,906 KB
Price

Change Currency