PENGEMBANGAN MODEL KECEPATAN GELOMBANG SEISMIK BERBASIS ARTIFICIAL NEURAL NETWORK (ANN) BERDASARKAN DATA GEMPA VULKANIK GUNUNG SINABUNG

NAJWA KAILA NUR ALIF, . (2026) PENGEMBANGAN MODEL KECEPATAN GELOMBANG SEISMIK BERBASIS ARTIFICIAL NEURAL NETWORK (ANN) BERDASARKAN DATA GEMPA VULKANIK GUNUNG SINABUNG. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (635kB)
	Text BAB 1_compress.pdf Download (73kB)
	Text BAB 2.pdf Restricted to Registered users only Download (913kB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (278kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (274kB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (144kB) | Request a copy
	Text Daftar Pustaka.pdf Download (188kB)
	Text Lampiran.pdf Restricted to Registered users only Download (242kB) | Request a copy

Abstract

Model kecepatan gelombang seismik merupakan salah satu parameter penting dalam analisis lokasi hiposenter serta interpretasi struktur bawah permukaan gunung api. Ketelitian model kecepatan sangat dipengaruhi oleh kualitas data seismik dan metode pemodelan yang digunakan. Penelitian ini bertujuan mengidentifikasi waktu tiba gelombang P dan S sebagai data masukan, mengembangkan model kecepatan gelombang seismik berbasis Artificial Neural Network (ANN), serta mengevaluasi performa model yang dihasilkan. Data yang digunakan berupa rekaman aktivitas seismik Gunung Sinabung periode Januari 2017 yang diperoleh dari Badan Riset dan Inovasi Nasional (BRIN) dan Pusat Vulkanologi dan Mitigasi Bencana Geologi (PVMBG). Tahapan penelitian meliputi preparasi data, deteksi event menggunakan metode Short Term Average/Long Term Average (STA/LTA), phase picking menggunakan SeisGram2K80, estimasi hiposenter awal menggunakan metode Geiger Adaptive Damping (GAD), evaluasi kualitas picking menggunakan Diagram Wadati, penyusunan dataset, normalisasi data, serta pelatihan model ANN. Hasil Diagram Wadati menunjukkan rasio Vp/Vs sebesar 1,718 dengan koefisien determinasi (R²) sebesar 0,797, yang menunjukkan konsistensi hasil picking. Evaluasi model ANN pada dua skenario model kecepatan awal menunjukkan bahwa Model 2 memberikan performa terbaik dengan nilai MSE sebesar 0,01333, RMSE sebesar 0,11544 km/s, MAE sebesar 0,09040 km/s, MAPE sebesar 3,203%, dan R² sebesar 0,793, sedangkan Model 1 menghasilkan MSE sebesar 0,01616, RMSE sebesar 0,12712 km/s, MAE sebesar 0,10396 km/s, MAPE sebesar 3,836%, dan R² sebesar 0,749. Model kecepatan hasil ANN menunjukkan pola peningkatan nilai Vp terhadap kedalaman yang masih konsisten dengan karakteristik geologi bawah permukaan Gunung Sinabung. Hasil penelitian menunjukkan bahwa pendekatan ANN mampu memodelkan hubungan nonlinier antara parameter seismik dan kecepatan gelombang P sehingga berpotensi menjadi alternatif dalam pengembangan model kecepatan awal pada studi seismik gunung api. Kata kunci: Artificial Neural Network, Gunung Sinabung, Diagram Wadati, Geiger Adaptive Damping, Model Kecepatan, Gelombang Seismik. ***** Seismic velocity models play an important role in hypocenter analysis and subsurface structural interpretation in volcanic areas. The accuracy of a velocity model depends on the quality of seismic data and the modeling approach employed. This study aims to identify P-wave and S-wave arrival times as input data, develop a seismic velocity model using an Artificial Neural Network (ANN), and evaluate the performance of the proposed model. The dataset consists of Mount Sinabung seismic records acquired during January 2017 from the National Research and Innovation Agency (BRIN) and the Center for Volcanology and Geological Hazard Mitigation (PVMBG). The methodology includes seismic data preparation, event detection using the Short-Term Average/Long-Term Average (STA/LTA) method, phase picking using SeisGram2K80, initial hypocenter estimation using the Geiger Adaptive Damping (GAD) method, picking quality assessment through Wadati analysis, dataset construction, data normalization, and ANN model training. The Wadati diagram produced a Vp/Vs ratio of 1.718 with a coefficient of determination (R²) of 0.797, indicating consistent phase picking results. Performance evaluation using two initial velocity model scenarios showed that Model 2 achieved the best performance with an MSE of 0.01333, RMSE of 0.11544 km/s, MAE of 0.09040 km/s, MAPE of 3.203%, and R² of 0.793, while Model 1 yielded an MSE of 0.01616, RMSE of 0.12712 km/s, MAE of 0.10396 km/s, MAPE of 3.836%, and R² of 0.749. The ANN-derived velocity model exhibits an increasing P-wave velocity with depth, which is consistent with the geological characteristics of Mount Sinabung's subsurface. These results demonstrate that ANN can effectively model the nonlinear relationship between seismic parameters and P-wave velocity, providing a promising alternative approach for developing initial velocity models in volcanic seismic studies.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dr. Bambang Heru Iswanto, M.Si 2). Dr. Mohammad Hasib, M.Sc
Subjects:	Sains > Fisika
Divisions:	FMIPA > S1 Fisika
Depositing User:	Users 36182 not found.
Date Deposited:	12 Aug 2026 03:49
Last Modified:	12 Aug 2026 03:49
URI:	http://repository.unj.ac.id/id/eprint/70107

Actions (login required)

View Item