Reza, Ahmad Andi (2025) Inversi Data Gravitasi Untuk Estimasi Kedalaman Basement Relief Menggunakan Multi-Objective Evolutionary Algorithm Base On Decomposition (MOEA/D). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5001211078-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5001211078-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy |
Abstract
Data gravitasi diinversi untuk mendapatkan kedalaman basement relief melalui proses inversi data gravitasi. Dalam melakukan proses inversi, algoritma yang efektif dan robust diperlukan agar proses inversi dilakukan secara akurat dan cepat. Dalam penelitian ini, inversi data gravitasi menggunakan algoritma Multi-Objective Evolutionary Algorithm Based on Decomposition (MOEA/D) diterapkan pada data gravitasi untuk mengestimasi kedalaman basement. Melalui penelitian ini, performa algoritma MOEA/D diuji dalam inversi data gravitasi sintetik (noise-free dan yang ditambah Gaussian noise 10%) dan observasi untuk mengestimasi kedalaman basement. Karena solusi dari inversi data gravitasi ini non-unique dan keberadaan noise dalam data lapangan, penilaian ketidakpastian (uncertainty) dilakukan menggunakan analisis interquartile range (IQR) dari distribusi solusi Pareto. Berdasarkan penilaian ketidakpastian parameter model yang dihasilkan, algoritma MOEA/D akurat untuk menentukan kedalaman basement dari data gravitasi sintetik dengan RMSE 0,1663-0,3018% (noise-free) dan 7,5965-10,5489% (Gaussian noise 10%). Selanjutnya, inversi data gravitasi observasi dilakukan pada gravitasi residual dari daerah Aegean Graben, Anatolia Barat, Turki. Algoritma MOEA/D dibandingkan dengan menggunakan algoritma differential evolution (DE) dan data sumur bor. Hasil estimasi kedalaman basement dari inversi MOEA/D data Profil 4 di lokasi sumur bor bernilai 0,7328 ± 0,0022 km. Adapun hasil inversi dengan algoritma DE bernilai 0,6088 km. Sedangkan kedalaman sumur bor bernilai 0,7300 km. Artinya, hasil dari MOEA/D lebih mendekati hasil data bor. Selanjutnya, algoritma MOEA/D digunakan untuk inversi data gravitasi yang terukur dari Gunung Galunggung. Algoritma MOEAD menghasilkan nilai kedalaman basement berkisar 0,2087-0,3408 km dengan ketidakpastian sebesar ±0,0031 km. Hasil inversi tersebut sesuai dengan akumulasi berbagai jenis endapan sedimen sekitar Gunung Galunggung yang dapat mencapai lebih dari 215 m pada hasil penelitian terdahulu. Dengan demikian, inversi data gravitasi menggunakan algoritma MOEA/D cukup akurat untuk estimasi kedalaman basement relief.
===================================================================================================================================
Gravity data is inverted to obtain basement relief depth through a gravity data inversion process. An effective and robust algorithm is required to perform the inversion accurately and quickly. In this study, gravity data inversion using the Multi-Objective Evolutionary Algorithm Based on Decomposition (MOEA/D) is applied to gravity data to estimate basement depth. The performance of the MOEA/D algorithm is tested on synthetic gravity data (noise-free and with 10% Gaussian noise added) and observed data for estimating basement depth. Due to the non-unique nature of gravity data inversion solutions and the presence of noise in field data, uncertainty assessment is performed using an interquartile range (IQR) analysis of the Pareto solution distribution. Based on the uncertainty assessment of the resulting model parameters, the MOEA/D algorithm is accurate for determining basement depth from synthetic gravity data with RMSE values of 0.1663-0.3018% (noise-free) and 7.5965-10.5489% (10% Gaussian noise). Furthermore, observed gravity data inversion is applied to the residual gravity from the Aegean Graben, Western Anatolia, Turkey. The MOEA/D algorithm is compared with the differential evolution (DE) algorithm and borehole data. The basement depth estimate from MOEA/D inversion of Profile 4 at the borehole location is 0.7328 ± 0.0022 km. Meanwhile, the inversion result using the DE algorithm is 0.6088 km, while the borehole depth is 0.7300 km. This indicates that the MOEA/D result is closer to the borehole data. Subsequently, the MOEA/D algorithm is used to invert measured gravity data from Mount Galunggung. The MOEA/D algorithm yields basement depth values ranging from 0.2087 to 0.3408 km with an uncertainty of ±0.0031 km. These inversion results are consistent with the accumulation of various types of sedimentary deposits around Mount Galunggung, which previous studies indicate can exceed 215 m. Thus, gravity data inversion using the MOEA/D algorithm is sufficiently accurate for estimating basement relief depth.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Inversi gravitasi, MOEA/D, basement relief, optimasi multiobjektif, ketidakpastian model. Gravity inversion, MOEA/D, basement relief, multi-objective optimization, uncertainty model. |
| Subjects: | G Geography. Anthropology. Recreation > G Geography (General) Q Science Q Science > QE Geology Q Science > QE Geology > QE471 Sedimentary rocks. Sedimentology Q Science > QE Geology > QE601 Geology, Structural |
| Divisions: | Faculty of Mathematics and Science > Physics > 45201-(S1) Undergraduate Thesis |
| Depositing User: | Ahmad Andi Reza |
| Date Deposited: | 30 Jul 2025 08:21 |
| Last Modified: | 30 Jul 2025 08:21 |
| URI: | http://repository.its.ac.id/id/eprint/123982 |
Actions (login required)
 |
View Item |