Magfirah, Vania Iqsora (2018) Karakterisasi Reservoir Minyak dan Gas Bumi Menggunakan Analisis Multi Atribut dan Neural Network pada Lapangan “VAL”, Formasi Bekasap, Cekungan Sumatera Tengah. Sarjana thesis, Universitas Brawijaya.
Abstract
Distribusi batuan reservoir akan lebih mudah diketahui berdasarkan properti fisis yang merepresentasikan karakteristik reservoir secara universal. Properti fisis tersebut diantaranya adalah volume serpih dan porositas. Untuk kepentingan delineasi reservoir, maka perlu dilakukan prediksi properti fisis tersebut pada lapangan VAL, Formasi Bekasap, Cekungan Sumatera Tengah menggunakan analisis multi atribut dan neural network. Data yang dilibatkan adalah data seismik 3D PSTM (Post Stack Time Migration) dan 26 data sumur. Hasil penelitian menunjukkan bahwa, analisis multi atribut dan neural network dapat memprediksi volume serpih secara baik, dengan faktor korelasi 0.76. Akan tetapi, tidak cukup baik dalam memprediksi porositas dengan faktor korelasi sebesar 0.47. Sehingga, prediksi porositas dilakukan dengan analisis single attribute yang menghasilkan faktor korelasi sebesar 0.91. Hasil prediksi tersebut disajikan dalam peta persebaran melalui analisis time slice. Melalui peta persebaran tersebut, kemudian dilakukan interpretasi persebaran volume serpih dan porositas sehingga diketahui distribusi batuan reservoir batu pasir. Hasil integrasi antara peta persebaran volume serpih dan porositas, reservoir batu pasir (sand stone) terdistribusi di arah barat laut hingga tenggara pada horizon Sand A dan barat laut hingga utara horizon Sand C. Di mana sand A memiliki kandungan serpih sebesar 0% - 36% dan nilai porositas 15% - 19% (kualitas baik). Sedangkan Sand C memiliki kandungan serpih 0% - 36% dan nilai porositas 16% - 24% dengan kualitas sangat baik.
English Abstract
Distribution of reservoir rocks will be much easier to discovered based on physical property which represent the characteristics of reservoir in universal. These physical property such as shale volume and porosity. For the purpose of reservoir delineation, it is necessary to predict physical property in “VAL” field, Bekasap Formation, Central Sumatra Basin using multiattributes analysis and neural network. Data which involved are 3D seismic (Post Stack Time Migration) and 26 well’s data. For result, multiattribute analysis and neutral network are able to predict shale volume with correlation factor 0.76 in a good way. However, it is not good enough to predict porosity with correlation factor 0.47. Therefore, prediction of porosity done by single attribute analysis and generate correlation 0.91. This result is presented in distribution map through time slice analysis. Interpretation of distribution shale volume and porosity using distribution map discovered distribution sandstone reservoir rock. Integration’s result between distribution map of shale volume and porosity are sandstone reservoir rock distributed in the northwest to southeast on the Sand A horizon and in the northwest to north on Sand C horizon. In terms of Sand A, it contents 0% - 36% of shale and the value of porosity is 15% - 19% (in good quality). For Sand C, which content 0% - 36% of shale and the value porosity is 16% - 24% (in excellent quality).
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/MIPA/2018/181/051805645 |
| Uncontrolled Keywords: | Multi atribut, neural network, volume serpih, porositas-Multiattributes, neural network, shale volume, porosity |
| Subjects: | 000 Computer science, information and general works > 006 Special computer methods > 006.3 Artificial intelligence > 006.32 Neural nets (neural networks) |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam > Fisika |
| Depositing User: | soegeng sugeng |
| Date Deposited: | 29 Jun 2020 06:28 |
| Last Modified: | 23 Dec 2020 05:39 |
| URI: | http://repository.ub.ac.id/id/eprint/168636 |
Actions (login required)
 |
View Item |