Sistem Pemodelan Gempa Vulkanik Berbasis Integrasi Multi-Event

Asrori, Nur Dia (2026) Sistem Pemodelan Gempa Vulkanik Berbasis Integrasi Multi-Event. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2042221073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Pemodelan gempa vulkanik merupakan bagian penting dalam pemantauan aktivitas vulkanik, namun masih menghadapi keterbatasan akurasi dan kurangnya integrasi antar metode analisis. Oleh karena itu, proyek akhir ini mengembangkan sistem pemodelan gempa vulkanik berbasis integrasi multi-event dengan pendekatan hybrid artificial intelligence. Sistem ini menggabungkan metode Ant Colony Optimization (ACO), Genetic Algorithm (GA), Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), dan Naive Bayes untuk memproses dan menganalisis data historis aktivitas vulkanik. Data yang digunakan berasal dari Volcanoes & Earthquakes dan MIROVA dengan parameter waktu kejadian, lintang, bujur, magnitudo, kedalaman, serta radiasi termal. ACO digunakan untuk menentukan pusat aktivitas gempa vulkanik dan luas area terdampak. Selanjutnya, GA digunakan untuk memprediksi arah dan sudut pergerakan aktivitas gempa. LSTM berfungsi menyimpan hasil pemodelan setiap kejadian gempa serta mendeteksi adanya anomali. Hasil tersebut kemudian digunakan oleh CNN untuk menghasilkan prediksi arah dan sudut pergerakan yang lebih optimal. Selanjutnya, Naive Bayes digunakan untuk mengevaluasi kinerja sistem prediksi. Hasil pengujian pada wilayah Jawa Timur menunjukkan bahwa sistem mampu menghasilkan pemodelan risiko dan pemetaan area terdampak yang lebih terintegrasi dan konsisten, dengan akurasi klasifikasi Naive Bayes sebesar 90%. Sistem ini dapat dikembangkan lebih lanjut sebagai dasar sistem peringatan dini gempa vulkanik.
==============================================================================================================================================================
Volcanic earthquake modeling is an important part of volcanic activity monitoring, but still faces limitations in accuracy and a lack of integration between analysis methods. Therefore, this final project develops a volcanic earthquake modeling system based on multi-event integration with a hybrid artificial intelligence approach. This system combines Ant Colony Optimization (ACO), Genetic Algorithm (GA), Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and Naive Bayes methods to process and analyze historical volcanic activity data. The data used comes from Volcanoes & Earthquakes and MIROVA with parameters of event time, latitude, longitude, magnitude, depth, and thermal radiation. ACO is used to determine the epicenter of volcanic earthquake activity and the extent of the affected area. Next, GA is used to predict the direction and angle of movement of earthquake activity. LSTM functions to store the modeling results of each earthquake event and detect anomalies. These results are then used by CNN to produce more optimal predictions of direction and angle of movement. Furthermore, Naive Bayes is used to evaluate the performance of the prediction system. Testing results in East Java demonstrated that the system was capable of producing more integrated and consistent risk modeling and mapping of affected areas, with a Naive Bayes classification accuracy of 90%. This system can be further developed as the basis for a volcanic earthquake early warning system.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Gempa Vulkanik, Sistem Pemodelan, Multi-Event , Peringatan Dini, Hybrid AI, Volcanic Earthquakes, Modeling Systems, Multi-Event , Early Warning, Hybrid AI
Subjects:	H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA402.5 Genetic algorithms. Interior-point methods. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Vocational > Instrumentation Engineering
Depositing User:	Nur Dia Asrori
Date Deposited:	31 Jul 2026 08:39
Last Modified:	31 Jul 2026 08:39
URI:	http://repository.its.ac.id/id/eprint/140506

Actions (login required)

View Item