Analisis Pengaruh Robust Processing Terhadap Koherensi Sinyal Pada Pemodelan 2D Data Magnetotelurik di Daerah Prospek Geotermal Gunung Endut

Nasution, Fajar Perdana (2021) Analisis Pengaruh Robust Processing Terhadap Koherensi Sinyal Pada Pemodelan 2D Data Magnetotelurik di Daerah Prospek Geotermal Gunung Endut. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 03411740000033-Undergraduate_Thesis.pdf] Text
03411740000033-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (6MB) | Request a copy

Abstract

Dalam eksplorasi geotermal, magnetotelurik (MT) sering digunakan karena keefektifannya dalam memetakan komponen sistem geotermal berdasarkan parameter resistivitas. Data MT memiliki frekuensi sangat luas (10-5104 Hz) dan rentan terkontaminasi oleh noise outlier. Analisis koherensi sinyal dalam bentuk multiple coherence efektif untuk mendeteksi kontaminasi noise outlier. Penelitian ini menjelaskan tentang analisis perbandingan data MT dengan nilai multiple coherence dari tiga parameter robust processing, yaitu no weight, rho variance, dan ordinary coherency. Data yang telah dibandingkan diproses lebih lanjut seperti seleksi crosspower, rotasi tensor impedansi, koreksi static-shift, dan pemodelan inversi 2-D. Parameter no weight memberikan nilai rata-rata multiple coherence sebesar 72,89%, rho variance 83,08%, dan ordinary coherency 82,68%. Setelah dilakukan seleksi crosspower nilai multiple coherence meningkat untuk no weight menjadi 83,95%, rho variance 92,11%, dan ordinary coherency 92,01%. Nilai rms error setelah pemodelan inversi 2-D lintasan 1 untuk no weight sebesar 4,438%, rho variance 1,7147%, serta ordinary coherency 1,742935%, sedangkan untuk lintasan 2 untuk no weight 8,1460%, rho variance 1,9536%, serta ordinary coherency 1,742935%. Robust processing berpengaruh terhadap peningkatan multiple coherence dengan mengeliminasi noise outlier penyebab bias pada tensor impedansi. Parameter robust processing rho-variance paling efektif untuk mengeliminasi noise, dengan multiple coherence tertinggi daripada parameter lain. Pada proses inversi 2-D, data hasil robust processing memiliki nilai rms error relatif lebih rendah, serta persebaran resistivitas yang dihasilkan lebih representatif terhadap informasi geologi. Model hasil parameter rho-variance menghasilkan rms error terkecil dan penampang resistivitas yang paling representatif terhadap informasi geologi.
=======================================================================================================
In geothermal exploration, magnetotelluric (MT) is often used because of its effectiveness in mapping the components of a geothermal system based on resistivity parameters. MT data has a very wide frequency (10-5104 Hz) and prone to contamination by noise outliers. Signal coherence analysis in the form of multiple coherence is effective for detecting outlier noise contamination. This study describes the comparative analysis of MT data with multiple coherence values from three robust processing parameters, namely no weight, rho variance, and ordinary coherency. The compared data is further processed such as crosspower selection, impedance tensor rotation, static-shift correction, and 2-D inversion modeling. The no weight parameter gives an average multiple coherence value of 72.89%, rho variance 83.08%, and ordinary coherency 82.68%. After crosspower selection, the value of multiple coherence increased for no weight to 83.95%, rho variance 92.11%, and ordinary coherency 92.01%. The rms error value after modeling the 2-D inversion of line 1 for no weight is 4.438%, rho variance is 1.7147%, and ordinary coherency is 1.742935%, while for line 2 for no weight is 8.1460%, rho variance is 1.9536 %, and ordinary coherency 1.742935%. Robust processing has an effect on increasing multiple coherence by eliminating outlier noise that causes bias in the impedance tensor. The robust processing rho-variance parameter is the most effective for eliminating noise, with the highest multiple coherence compared to other parameters. In the 2-D inversion process, the robust processing data has a relatively lower rms error value, and the resulting resistivity distribution is more representative of geological information. The rho-variance parameter model results in the smallest rms error and the most representative resistivity cross section of geological information.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Magnetotelurik, Geotermal, Multiple Coherence, Robust Processing, Pemodelan 2-D, Magnetotelluric, Geothermal, Multiple Coherence, Robust Processing, 2-D Modeling.
Subjects: Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics)
Q Science > QC Physics > QC20.7.F67 Fourier transformations
Q Science > QC Physics > QC610.3 Electric conductivity
Q Science > QC Physics > QC718 Magnetohydrodynamics.
Q Science > QE Geology > QE601 Geology, Structural
Divisions: Faculty of Civil Engineering and Planning > Geophysics Engineering > 33201-(S1) Undergraduate Thesis
Depositing User: Fajar Perdana Nasution
Date Deposited: 26 Aug 2021 07:42
Last Modified: 26 Aug 2021 07:42
URI: http://repository.its.ac.id/id/eprint/89653

Actions (login required)

View Item View Item