Reservoir characterization using pre-stack simultaneous inversion of 3D seismic data from Timor Sea, Australia

การหาลักษณะเฉพาะของแหล่งกักเก็บโดยใช้การผกผันแบบผสมผสานก่อนรวมสัญญาณคลื่นของข้อมูลคลื่นไหวสะเทือน 3 มิติ จากทะเลติมอร์ ออสเตรเลีย

Name: Namfon Pooksook

LCSH: Seismic waves

Classification :.DDC: 551.22

LCSH: Seismic prospecting

Abstract: The Vulcan Sub-basin is located in the Western Timor Sea, an area of particular seismic imaging difficulties. This has led to continuous challenges when building reliable subsurface models in this area. This study will discuss a workflow to improve the delineation of the reservoir properties in the petroleum exploration area, using a seismic reservoir characterization approach. The study workflow comprise of five main steps; (1) rock physics analysis, (2) well tie and wavelet extraction, (3) low frequency modelling, (4) pre-stack simultaneous inversion, and (5) lithofacies classification. The rock physics analysis was carried out using four wells and mainly obtained reservoir characterization feasibility assessment. Lithology discrimination was deemed feasible considering pre-stack seismic inversion for both acoustic impedance (AI) and shear impedance (SI). Extracted wavelets of three seismic partial stacks (near, mid and far angle stacks) were achieved at well locations when well ties performed to optimize time-depth relations. The final averaged wavelets of each stacks were calculated from selected extracted wavelets. The low frequency models are significant to transform the seismic derived relative elastic impedance values to absolute elastic properties. Final low frequency models of elastic properties (AI, SI and density) were combined between the ultra-low frequency models that derived from seismic stacking velocity and low frequency models using well data. The pre-stack simultaneous inversion algorithm was based on Constrained Sparse Spike Inversion, and used three partial stacks with theirs averaged wavelets and also final low frequency models to produce the inverted properties. Finally, lithofacies classification and probability cubes were calculated using g combined absolute inverted AI and Vp/Vs cubes by applying the derived Bayesian 2D probability density functions for each lithofacies type. In conclusion, Vp/Vs was the key parameter to enable seismic reservoir characterization in the study area. Thicker sandstone layers were mostly efficiently classified from the other most common lithology types in the area, such as shale and carbonate. The comparison between lithology logs and lithology cubes showed a good correlation at all wells. However, the seismic detectability in the area was limited by the low acoustic impedance contrast between two different lithology types. It was assumed that individual sand layers thinner than approximately 23 meters could not be accurately classified, based on the seismic reservoir characterization methodology and input data applied in this study.

Chiang Mai University Library

Address: Chiang Mai

Email: cmulibref@cmu.ac.th

Name: Helmut Jakubowicz

Role: Chairman

Name: Pisanu Wongpornchai

Role: Member

Name: Helge Ivar Sognnes

Role: Member

Name: Diako Hariri Naghadeh

Role: Member

Created: 2016

Modified: 2020-01-10

Issued: 2020-01-10

วิทยานิพนธ์/Thesis

application/pdf

Th 551.22 N174R

eng

DegreeName: Master of Science

Level: Master's Degree

Descipline: Petroleum Geophysics

Grantor: Chiang Mai University

©copyrights Chiang Mai University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	geopt80116napo_full.pdf	36.28 MB

ใช้เวลา

0.021983 วินาที