Pemodelan Struktur Dike 2D Berdasarkan Data Magnetik untuk Eksplorasi Mineral Menggunakan Ensemble Kalman Inversion dengan Regularisasi Tikhonov

Syahputra, Muhammad Akbar Maulana (2026) Pemodelan Struktur Dike 2D Berdasarkan Data Magnetik untuk Eksplorasi Mineral Menggunakan Ensemble Kalman Inversion dengan Regularisasi Tikhonov. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5017221008-Undergraduate_Thesis.pdf Download (5MB)

Abstract

Pemodelan struktur dike 2D berdasarkan data magnetik merupakan pendekatan yang umum dilakukan dalam eksplorasi mineral. Akan tetapi, pendekatan ini memiliki tantangan karena permasalahan inversi yang ill-posed dan non-uniqueness, terutama dalam kondisi data yang noisy. Studi ini mengusulkan metode inversi data magnetik untuk pemodelan dike berbasis Ensemble Kalman Inversion (EKI) yang ditingkatkan dengan regulasi Tikhonov. Penambahan regularisasi bertujuan untuk meningkatkan stabilitas numerik dan memitigasi sensitivitas terhadap degenerasi ensambel, sehingga memungkinkan kuantifikasi ketidakpastian yang efisien melalui statistik ensambel. Penelitian ini mengimplementasikan EKI dengan regularisasi Tikhonov melalui tahapan uji sensitivitas, eksperimen numerik untuk menentukan ukuran ensambel dan parameter regularisasi optimal, inversi data sintetis, hingga inversi pada data lapangan yang berbeda. Hasil eksperimen numerik terkontrol menunjukkan bahwa penggunaan Ne ≥ 300, dikombinasikan dengan regularisasi (λ) yang efektif mampu mengoptimalkan antara stabilitas konvergensi dan robustness, serta mencegah kegagalan algoritma. Implementasi pada inversi data magnetik sintetis dan lapangan (mineralisasi Cu, Fe, skarn, dan uranium) menunjukkan bahwa EKI yang teregularisasi menghasilkan model dike yang stabil dan konsisten secara geologi dengan hasil penelitian terdahulu serta data pengeboran. Hasil ini menunjukkan bahwa EKI dengan regularisasi Tikhonov dapat direkomendasikan sebagai metode inversi magnetik untuk pemodelan dike dalam eksplorasi mineral ekonomis.
==============================================================================================================================
Magnetic dike modeling is a common approach in mineral exploration. However, this approach faces challenges due to the ill-posed and non-unique in inversion problem, particularly when dealing with noisy data. This study proposes inversion method for 2D magnetic dike modeling based on Ensemble Kalman Inversion (EKI) enhanced with Tikhonov regularization. The addition of regularization aims to improve numerical stability and mitigate sensitivity to ensemble degeneracy, thereby enabling efficient quantification of uncertainty through ensemble statistics. This study has a series of steps, including sensitivity analysis, numerical experiments to determine the optimal ensemble size and regularization parameters, inversion of synthetic data, and inversion of various field data. The results of controlled numerical experiments show that using Ne ≥ 300, combined with effective regularization (λ), can optimize the balance between convergence stability and robustness, as well as prevent algorithm failure. Synthetic and field data (Cu, Fe, skarn, and uranium mineralization) inversion demonstrate that the regularized EKI produces stable dike models that are geologically consistent with previous research findings and drilling data. These results indicate that EKI with Tikhonov regularization can be recommended as a magnetic inversion method for modeling dikes in the exploration of economic minerals.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Dike, Magnetik, Eksplorasi Mineral, Ensemble Kalman Inversion, Regularisasi Tikhonov Dike, Magnetic, Mineral Exploration, Ensemble Kalman Inversion, Tikhonov Regularization Dike, Magnetic, Mineral Exploration, Ensemble Kalman Inversion, Tikhonov Regularization
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TN Mining engineering. Metallurgy > TN269 Prospecting--Geophysical methods
Divisions:	Faculty of Civil Engineering and Planning > Geophysics Engineering > 33201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Akbar Maulana Syahputra
Date Deposited:	09 Jul 2026 07:58
Last Modified:	09 Jul 2026 07:58
URI:	http://repository.its.ac.id/id/eprint/134595

Actions (login required)

View Item