Inversi Data Magnetotellurik (MT) 1-D Menggunakan Algoritma Ensemble Kalman Inversion (EKI)

Wijdannysa, Jasinda (2024) Inversi Data Magnetotellurik (MT) 1-D Menggunakan Algoritma Ensemble Kalman Inversion (EKI). Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5001201147_Undergraduate_Thesis.pdf] Text
5001201147_Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (9MB) | Request a copy

Abstract

Metode magnetotellurik (MT) merupakan metode eksplorasi geofisika pasif yang dapat menjangkau kedalaman lapisan batuan yang sangat dalam, sehingga cocok digunakan untuk penentuan kedalaman basemen. Untuk menentukan kedalaman basemen, data MT diinversikan agar mendapatkan estimasi parameter model (resistivitas dan ketebalan lapisan batuan). Pada penelitian ini, algoritma Ensemble Kalman Inversion (EKI) digunakan untuk melakukan inversi data MT 1-D. Algoritma EKI dilakukan uji coba pada data sintetik tipe kurva sounding (A, K, H, Q, D, Q) serta data sintetik kasus cekungan sedimen. Setelahnya, hasil inversi pada setiap data sintetik tersebut dianalisis posterior distribution model (PDM) dan principal component analysis (PCA). Analisis PDM dilakukan untuk mengestimasi ketidakpastian parameter model terbaik hasil inversi, sedangkan analisis PCA digunakan untuk mengkorelasikan parameter model sebenarnya pada data sintetik dengan parameter model hasil inversi. Hasil analisis PDM pada data sintetik A, K, H, Q, D, dan Q menunjukkan estimasi model parameter yang sesuai dengan model sebenarnya, sedangkan hasil analisis PCA menunjukkan model sebenarnya dan model terbaik yang berhimpit untuk setiap tipe kurva dan berada pada nilai fungsi objektif yang minimum. Hal ini mengindikasikan bahwa algoritma EKI terbukti robust dalam uji coba data sintetik tipe kurva sounding. Analisis yang sama juga dilakukan pada data MT sintetik untuk kasus cekungan sedimen. Namun, untuk data ini model terbaik pada analisis PDM menunjukkan hasil yang berbeda dari model sebenarnya, serta jarak model sebenarnya dengan model terbaik hasil inversi yang berjauhan pada ruang PCA. Hal ini menunjukkan bahwa algoritma EKI belum akurat dalam menentukan estimasi model parameter pada kasus cekungan sedimen sintetik, khususnya dalam penentuan parameter model resistivitas lapisan basemen.
=================================================================================================================================
The magnetotelluric (MT) method is a passive geophysical exploration technique that can probe deep rock layers, making it ideal for determining basement depths. To achieve this, inversion modeling of MT data is performed to estimate model parameters such as resistivity and rock layer thickness. This research utilizes the Ensemble Kalman Inversion (EKI) algorithm for 1-D MT data inversion. The EKI algorithm was tested on synthetic sounding curve data types (A, K, H, Q, D, Q) and synthetic data in the case of sedimentary basins. The inversion results for each synthetic dataset were analyzed using posterior distribution model (PDM) and principal component analysis (PCA). PDM analysis estimates the uncertainty of the best model parameters from the inversion, while PCA correlates the actual model parameters of synthetic data with the inversion results. For synthetic sounding curve data, PDM analysis indicated that the estimated model parameters matched the actual model, and PCA analysis showed the actual and best models coinciding at the minimum objective function value. This demonstrates the robustness of the EKI algorithm for synthetic sounding curve data. For synthetic MT data representing sedimentary basins, the PDM analysis showed discrepancies between the best model and the actual model, with a significant distance between them in PCA space. This suggests that the EKI algorithm is less accurate in estimating model parameters for synthetic sedimentary basin cases, particularly in determining basement layer resistivity model.

Item Type: Thesis (Other)
Uncontrolled Keywords: EKI, Kedalaman Basemen, Magnetotellurik 1-D, PCA, PDM. Basement Depth, 1-D Magnetotelluric
Subjects: Q Science
Q Science > QA Mathematics > QA274.2 Stochastic analysis
Q Science > QA Mathematics > QA274.7 Markov processes--Mathematical models.
Q Science > QA Mathematics > QA279 Response surfaces (Statistics). Analysis of covariance.
Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory.
Q Science > QA Mathematics > QA401 Mathematical models.
Q Science > QA Mathematics > QA402.3 Kalman filtering.
Q Science > QA Mathematics > QA9.58 Algorithms
Q Science > QC Physics
Q Science > QC Physics > QC610.3 Electric conductivity
Q Science > QC Physics > QC665.E38 Electric fields.
Q Science > QC Physics > QC881.2.I6 ionosphere
Q Science > QC Physics > QC973.4.T76 Tropospheric radio waves Including absorption, propagation, etc.
Q Science > QE Geology > QE1.F557 Magnetic anomalies--Measurement.
Q Science > QE Geology > QE471 Sedimentary rocks. Sedimentology
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.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis
Depositing User: Jasinda Wijdannysa
Date Deposited: 14 Aug 2024 07:26
Last Modified: 14 Aug 2024 07:26
URI: http://repository.its.ac.id/id/eprint/115180

Actions (login required)

View Item View Item