APLIKASI SEISMIC IMAGE PROCESSING BERBASIS MACHINE LEARNING PADA SEISMIK 2D DALAM PENGOLAHAN INVERSI SEISMIK MODEL-BASED

BIANTARA, WAHYU (2023) APLIKASI SEISMIC IMAGE PROCESSING BERBASIS MACHINE LEARNING PADA SEISMIK 2D DALAM PENGOLAHAN INVERSI SEISMIK MODEL-BASED. Other thesis, UPN "Veteran" Yogyajarta.

[thumbnail of WAHYU BIANTARA DAFTAR ISI.pdf] Text
WAHYU BIANTARA DAFTAR ISI.pdf
Download (165kB)
[thumbnail of SKRIPSI_WAHYU BIANTARA (115 170 030).pdf] Text
SKRIPSI_WAHYU BIANTARA (115 170 030).pdf
Restricted to Repository staff only
Download (8MB)
[thumbnail of DAFTAR PUSTAKA_WAHYU BIANTARA (115 170 030).pdf] Text
DAFTAR PUSTAKA_WAHYU BIANTARA (115 170 030).pdf
Download (142kB)
[thumbnail of COVER_WAHYU BIANTARA(115170030).pdf] Text
COVER_WAHYU BIANTARA(115170030).pdf
Download (130kB)
[thumbnail of abstrak_Wahyu Biantara (115 170 030).pdf] Text
abstrak_Wahyu Biantara (115 170 030).pdf
Download (127kB)
[thumbnail of lembar pengesahan.pdf] Text
lembar pengesahan.pdf
Download (2MB)

Abstract

ABSTRAK
APLIKASI SEISMIC IMAGE PROCESSING BERBASIS
MACHINE LEARNING PADA SEISMIK 2D DALAM
PENGOLAHAN INVERSI SEISMIK MODEL-BASED
Wahyu Biantara
115.170.030
Selama bertahun-tahun, pengolahan seismik telah mengalami kemajuan
yang signifikan, mulai dari tahap awal pengolahan data secara manual hingga era
modern dengan memanfaatkan bahasa pemrograman dan machine learning.
Peralihan ini secara signifikan meningkatkan akurasi dan efisiensi pengolahan
seismik. Di bidang geofisika, machine learning telah menghasilkan banyak
penerapan terbarukan yaitu memprediksi interpolasi data seismik, seismik inversi
untuk aplikasi reservoir, dan lain-lain. Machine Learning menawarkan potensi
untuk mengoptimalkan banyak aspek yang beragam dalam analisis data seismik.
penelitian kali ini, akan dilakukan pendekatan yang berbeda dalam
pemanfaatan machine learning yaitu dengan melakukan seismic image processing.
Penelitian ini menggunakan data dari hasil penelitian Jakfar Husin Almuhdar
(2019) yaitu berupa penampang seismik 2D Formasi Parigi Cekungan Jawa Barat
Utara yang bersifat open sources (bebas akses). Sebagai validasi, dilakukan
pengolahan seismik inversi model-based dengan perhitungan trial and error
menggunakan bahasa pemrograman python untuk mendapatkan nilai impedansi
akustik.
Hasil dari Seismic Image Processing didapatkan penampang seismik
digital yang terbentuk atas diskrit matriks berdimensi 336 x 1140. Sebanyak 1140
kolom diskrit dianggap sebagai trace digital yang berguna sebagai data untuk
pengolahan seismik inversi. Pengolahan inversi model-based menghasilkan grafik
AI yang menunjukkan korelasi yang baik antara hasil inversi AI terhadap data
seismik digital dengan kurva yang cenderung identik dengan koefisien korelasi
yang mendekati 1. Penampang AI yang dihasilkan menunjukkan nilai AI berkisar
antara 0,08 hingga 0,80. Penampang AI kemudian di overlay dengan penampang
seismik digital menunjukkan korelasi yang baik yang mana kemenerusan nilai AI
mengikuti pola lapisan serta perubahan kontras impedansi yang cenderung mirip
pada penampang seismik.
Kata Kunci: Seismic Image Processing, Inversi Model-Based, Impedansi Akustik,
Python
ABSTRACT
APPLICATION OF SEISMIC IMAGE PROCESSING BASED ON
MACHINE LEARNING IN 2D SEISMIC FOR MODEL-BASED
SEISMIC INVERSION PROCESSING
Wahyu Biantara
115.170.030
Over the years, seismic processing has experienced significant progress,
starting from the initial stages of manual data processing to the modern era using
programming and machine learning. This transition significantly improves the
accuracy and efficiency of seismic processing. In geophysics machine learning
has produced many renewable applications such as predicting interpolation of
seismic data, seismic inversion for reservoir applications, and others. Machine
Learning offers the potential to optimize various aspects of seismic data analysis.
In this research, a different approach will be taken in utilizing machine
learning, namely by carrying out seismic image processing. 2D seismic section of
the Parigi Formation of the North West Java Basin which is open source (free
access). As validation, model-based seismic inversion processing was carried out
with trial and error calculations using the Python programming to get the
acoustic impedance value.
The results of Seismic Image Processing obtained a digital seismic section
formed from a discrete matrix with dimensions of 336 x 1140. Total of 1140
discrete columns are considered as digital traces which are useful as data for
seismic inversion processing. Model-based inversion processing produces AI
graphs that show good correlation between AI inversion results on digital seismic
data on curves that to be identical with a correlation coefficient that is close to 1.
The resulting AI cross-section shows AI values ranging from 0.08 to 0.80. Then
the AI cross-section is overlaid with the digital seismic cross-section shows a
good correlation where the continuity of the AI values follows a layer pattern as
well as changes in impedance contrast which tend to be similar to seismic
sections.
Keywords: Seismic Image Processing, Model-Based Inversion, Acoustic
Impedance, Python

Item Type: Thesis (Other)
Uncontrolled Keywords: Seismic Image Processing, Model-Based Inversion, Acoustic Impedance, Python
Subjects: Q Science > QE Geology
Divisions: Faculty of Engineering, Science and Mathematics > School of Engineering Sciences
Depositing User: Eko Yuli
Date Deposited: 17 Nov 2023 08:29
Last Modified: 17 Nov 2023 08:29
URI: http://eprints.upnyk.ac.id/id/eprint/38151

Actions (login required)

View Item View Item