Moving From Wireline Logging to Logging-While-Drilling in an Aggressive North Sea Drilling and Formation Evaluation Program-A Case Study of the Brenda Accumulation

摘要

LWD was integrated into an aggressive drilling program to assess the Brenda field from a previously un-appraised discovery in the Central North Sea. In a six well drilling program in 2004, shifting the logging program from wireline to LWD facilitated a rapid but detailed evaluation of the Brenda accumulation with considerable time and cost savings. Originally single, vertical wells with wireline were planned but practice evolved into drilling 3-legged directional wells using full LWD suites to maximize information garnered per well. Using LWD enabled logging suites to reach TD in directional wells on first attempt and meant logging in-gauge hole rather than the washed-out hole suffered by the wireline tools due to drilling reactive shales with water based mud. Sophisticated LWD toolstrings including a formation pressure tester and a slow shear sonic tool were used in addition to the standard triple-combo (resistivity/density/neutron). Applying advanced data processing methods including phase velocity processing and a physics-based dispersion correction of the sonic as well as thin-bed processing of the resistivity facilitated quality control and quantitative use of the LWD data. The LWD formation tester was used to estimate the oil gravity and to accurately identify the depths of the oil-water contacts. This was required to identify the main Brenda East and Brenda West stratigraphically-trapped accumulations and also for reserves estimates. Downhole fluid sampling was not necessary in these wells. The LWD sonic measurements were used to refine the seismic models with accurate measurement of compressional and slow shear velocities in the overburden. This improved the top reservoir pick which is very important in a stratigraphic trap such as Brenda. The compressional and shear velocities measured in the reservoir were used to refine the acoustic impedance models. This meant that a costly VSP/Check shot survey was not required. This was fortunate because the only check shot survey that was run became stuck in the reactive formations above the reservoir. Thin-bed modeling of the resistivity logs helped derive a more accurate reserves estimate by providing horizontal and vertical true resistivities in a series of thin-beds in the "abandonment facies" above the main reservoir. Thin-bed modeling effectively increased the resolution of the LWD resistivity tool.