Prediksi Log Menggunakan Multi Resolution Graph-Based Clustering (MRGC) Sebagai Data Utama Analisis Distribusi Shale

Rahmatullah, Sayid (2024) Prediksi Log Menggunakan Multi Resolution Graph-Based Clustering (MRGC) Sebagai Data Utama Analisis Distribusi Shale. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5017201078-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (11MB) | Request a copy

Abstract

Kualitas yang baik dari data Well log merupakan pondasi krusial dalam analisis petrofisika sehingga dapat mewakili perhitungan dan gambaran yang valid dan komprehensif untuk formasi penelitian. Keberadaan bad data alias data yang buruk akibat adanya wash out membuat pembacaan log pada interval tersebut menjadi kurang reliable dan tidak dapat sepenuhnya menggambarkan karakterisasi formasi, semakin besar proporsi Shale dalam suatu formasi, semakin tinggi kemungkinan terjadinya zona washout. Sifat inheren serpih membuatnya lebih sensitif pada kondisi pengeboran sehingga harus menganalisis distribusi Shale di lapangan “Sova”, Cekungan Sumatera Tengah. Oleh karena itu, metode Multi Resolution Graph-Based Clustering digunakan untuk menciptakan log sintetik RHOB dan NPHI melalui pendekatan algoritma clustering berdasarkan log model Gamma-ray dengan memanfaatkan KNN dan tahapan lebih lanjut berupa Kernel Representative Index memungkinkan pengguna untuk mengontrol tingkat detail yang diperlukan untuk menentukan electrofacies. MRGC berhasil menghasilkan 25 model clustering dari RHOB dan NPHI, dengan masing-masing 78,278 dan 49,321 sampel dengan nilai korelasi antara log raw dan log sintesis adalah 0.803486 dan 0.943307 secara berurutan. Dengan log RHOB dan NPHI yang telah dikondisikan, menghasil Shale Volume dan total porosity yang konkrit sebagai main input dalam menganalisis distribusi Shale pada Lapangan “Sova” menggunakan Thomas Stieber Plot. Didapatkan net Reservoar yang baik berada pada interval yang didominasi oleh clean sand hingga minim laminated-dispersed shale dengan rentang porositas 0.2 hingga 0.25 V/V dibandingkan laminated-structural shale dengan rentang porositas 0.15 hingga 0.2 V/V, dimana masing-masing dari laminated shale cenderung berada pada interval Shale sedangkan dispersed shale berada pada interval Shaly-sand.
==============================================================================================================================
The good quality of well log data is a crucial foundation in petrophysical analysis, enabling valid and comprehensive calculations and representations of the studied formation. The presence of bad data, such as those caused by washout, makes log readings in those intervals less reliable and unable to fully characterize the formation. The higher the Shale proportion in a formation, the greater the likelihood of washout zones. The inherent properties of Shale make it more sensitive to drilling conditions, necessitating an Analysis of Shale distribution in the “Sova” field, Central Sumatra Basin. Therefore, the Multi Resolution Graph-Based Clustering (MRGC) method was used to create synthetic density and neutron logs through a clustering algorithm approach based on the Gamma-ray log model utilizing KNN and a further stage in the form of a Kernel Representative Index allowing users to control the level of detail required to determine electrofacies. MRGC successfully generated 25 clustering models from RHOB and NPHI, with 78,278 and 49,321 samples respectively, and correlation values between raw logs and synthetic logs of 0.803486 and 0.943307, respectively. With the conditioned density and neutron logs, a concrete Volume of Shale and Porosity was obtained, which can be used to analyze Shale distribution in the “Sova” field using the Thomas Stieber Plot. A good net reservoir was found in intervals dominated by clean sand to minimally laminated-dispersed shale with a porosity range of 0.2 to 0.25 V/V compared to laminated-structural shale with a porosity range of 0.15 to 0.2 V/V, where laminated shale tends to be found in Shale intervals while dispersed shale is found in Shaly-sand intervals.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Analisis Distribusi Shale, Multi Resolution Graph-Based Clustering, Thomas-Stieber, Wash out, Well log, Shale Distribution Analysis
Subjects:	Q Science > QE Geology > QE471 Sedimentary rocks. Sedimentology Q Science > QE Geology > QE601 Geology, Structural
Divisions:	Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geophysics Engineering > 33201-(S1) Undergraduate Thesis
Depositing User:	Sayid Rahmatullah
Date Deposited:	22 Aug 2024 06:08
Last Modified:	22 Aug 2024 06:08
URI:	http://repository.its.ac.id/id/eprint/114554

Actions (login required)

View Item