Inversi Data Gravitasi Menggunakan Random Opposition Based Tunicate Swarm Algorithm Untuk Pemodelan Bijih Dan Mineral

Romadhon, Risky Lucky (2025) Inversi Data Gravitasi Menggunakan Random Opposition Based Tunicate Swarm Algorithm Untuk Pemodelan Bijih Dan Mineral. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5001211088-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Inversi data gravitasi merupakan salah satu metode penting dalam eksplorasi mineral untuk memetakan struktur bawah permukaan. Penelitian ini menggunakan pendekatan optimasi global berbasis algoritma metaheuristik, yaitu Random Opposition-Based Tunicate Swarm Algorithm (ROBTSA). Evaluasi dilakukan secara komprehensif melalui dua skenario: data sintetik (dengan dan tanpa noise 10%) serta data lapangan dari dua wilayah dengan karakteristik geologi berbeda, yaitu Mobrun (Kanada) dan Mundiyawas-Khera (India). Untuk menilai kualitas dan ketidakpastian solusi, digunakan pendekatan Posterior Distribution Model (PDM) dan Principal Component Analysis (PCA). Hasil menunjukkan bahwa pada data sintetik tanpa noise, parameter hasil inversi memiliki distribusi yang sempit dan konvergen, menandakan tingkat ketidakpastian yang rendah. Namun, pada data dengan noise dan data lapangan, distribusi parameter lebih menyebar, terutama pada kasus anomali jamak Mundiyawas, yang mengindikasikan ketidakpastian model yang lebih tinggi akibat tumpang tindih anomali. Analisis PCA mendukung hasil ini dengan menunjukkan sebaran solusi yang lebih luas pada kondisi data bising. Pada studi Mobrun, algoritma ROBTSA menunjukkan performa unggul dengan nilai fungsi objektif rendah dan parameter model yang mendekati nilai referensi, sesuai dengan temuan dalam literatur sebelumnya. Di sisi lain, kasus Mundiyawas menunjukkan keterbatasan dalam memisahkan anomali lemah, meskipun citra anomali utama berhasil dimodelkan dengan dukungan informasi geologi lapangan. Secara keseluruhan, ROBTSA terbukti andal untuk inversi data gravitasi, baik dalam kondisi ideal maupun kompleks.
======================================================================================================================================
Gravity data inversion is one of the essential methods in mineral exploration to map subsurface structures. This study employs a global optimization approach based on a metaheuristic algorithm known as the Random Opposition-Based Tunicate Swarm Algorithm (ROBTSA). The algorithm’s performance was comprehensively evaluated through two scenarios: synthetic data (with and without 10% random noise) and field data from two geologically distinct region —Mobrun (Canada) and Mundiyawas-Khera (India). To assess the quality and uncertainty of the inversion results, the Posterior Distribution Model (PDM) and Principal Component Analysis (PCA) were applied. The results showed that for noise-free synthetic data, the inverted model parameters exhibited narrow and convergent distributions, indicating low levels of uncertainty. However, for noisy synthetic and field data, the parameter distributions became more dispersed—particularly in the multiple-anomaly case of Mundiyawas—reflecting higher model uncertainty due to overlapping anomaly signals. PCA supported these findings by illustrating a wider spread of solutions under noisy conditions. In the Mobrun case study, ROBTSA demonstrated excellent performance, yielding low objective function values and model parameters closely matching the reference values, in line with previous research findings. Conversely, the Mundiyawas case revealed limitations in resolving weak anomalies, although the main anomaly was successfully imaged with the support of geological field information. Overall, ROBTSA proved to be a robust and reliable algorithm for gravity data inversion under both ideal and complex geological conditions.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Inversi Gravitasi, ROBTSA, Algoritma Metaheuristik, Posterior Distribution Model (PDM), Gravity Inversion, ROBTSA, Metaheuristic Algorithms, Posterior Distribution Model (PDM).
Subjects:	Q Science > Q Science (General) > Q180 Gravitation. Q Science > Q Science (General) > Q337.3 Swarm intelligence Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA9.58 Algorithms
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis
Depositing User:	Risky Lucky Romadhon
Date Deposited:	05 Aug 2025 07:54
Last Modified:	05 Aug 2025 07:54
URI:	http://repository.its.ac.id/id/eprint/126969

Actions (login required)

View Item