Estimasi Profil Kecepatan Gelombang Geser Melalui Inversi Kurva Dispersi Gelombang Rayleigh Berbasis Ensemble Kalman Inversion Pada Kasus Evaluasi Kekuatan Tanggul

Sutidar, Ahmad Zidan (2026) Estimasi Profil Kecepatan Gelombang Geser Melalui Inversi Kurva Dispersi Gelombang Rayleigh Berbasis Ensemble Kalman Inversion Pada Kasus Evaluasi Kekuatan Tanggul. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5017221058-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Karakterisasi bawah permukaan secara non-intrusif merupakan keburuhan mendasar dalam mitigasi kegagalan struktur geoteknik. Tanggul Lusi menjadi kasus kritis karena mengalami kegagalan seperti seepage, collapse, dan overtopping akibat dari sedimen aluvial lunak. Sehingga identifikasi zona lemah secara objektif menjadi krusial. Penelitian ini menggunakan metode Multichannerl Analysis of Surface Waves untuk merekontruksi profil kecepatan gelombang geser (V_{s}) 1D ditanggul LUSI P.79-82. Inversi kurva dispersi gelombang Rayleigh merupakan permasalahan yang ill-posed dan non linear. Untuk mengatasi hal itu dilakukan pendekatan secara Ensemble Kalman Inversion (EKI) yang bersifat derivative-free. Skema asimilasi data Bayesian, EKI mencari solusi global tanpa paramater tunning secara luas dan inheren mengkuantifikasi ketidakpastian melalui statistik posterior ensambel. Untuk validasinya menggunakan tiga skenario model (N, H,dan L) dari free noise dan noisy 5% yang memiliki similarity index 98-99% pada kondisi ideal dan tetap diatas 85% ber-noise Penerapan pada data lapang menghasilkan model empat lapisan yang berhasil mendeteksi zona kecepatan tinggi yang diapit dengan zona rendah dan selaras N-SPT pada domain dangkal. Kerangka Bayesian ini menjadikan EKI-MASW pendekatan yang robust dan objektif untuk karakterisasi kekakuan bawah permukaan tanggul.
================================================================================================================================
Non-intrusive subsurface characterization is a fundamental requirement for mitigating geotechnical structural failure. The Sidoarjo Mud (LUSI) embankment represents a critical case, having suffered repeated failures such as seepage, collapse, and overtopping driven by soft alluvial sediments, making objective identification of weak zones essential. This study applies the Multichannel Analysis of Surface Waves (MASW) method to reconstruct the 1D shear-wave velocity (Vs) profile at the LUSI embankment, point P.79–82. Inversion of the Rayleigh-wave dispersion curve is an ill-posed and non-linear problem; to address this, a derivative-free Ensemble Kalman Inversion (EKI) approach is employed. As a Bayesian data-assimilation scheme, EKI seeks a global solution without extensive parameter tuning and inherently quantifies uncertainty through posterior ensemble statistics. The algorithm is validated on three synthetic scenarios model (N, H, and L) under noise-free and 5% noise-contaminated conditions, yielding a Similarity Index of 98–99% under ideal conditions and remaining above 85% under noise. Application to field data produces a four-layer model that successfully detects a buried low-velocity zone and agrees with N-SPT in the shallow domain. This Bayesian framework establishes EKI-MASW as a robust and objective approach for characterizing the stiffness of the embankment subsurface.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	MASW, EKI, Kegagalan Tanggul LUSI, gelombang Rayleigh, N-SPT, Bayesian. MASW, EKI, LUSI Embankment Failure, Rayleigh Waves, N-SPT, Bayesian.
Subjects:	T Technology > T Technology (General) > T174.5 Technology--Risk assessment. T Technology > T Technology (General) > T57.74 Linear programming T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > T Technology (General) > T57.84 Heuristic algorithms.
Divisions:	Faculty of Civil Engineering and Planning > Geophysics Engineering > 33201-(S1) Undergraduate Thesis
Depositing User:	Ahmad Zidan Sutidar
Date Deposited:	30 Jul 2026 01:00
Last Modified:	30 Jul 2026 01:00
URI:	http://repository.its.ac.id/id/eprint/137239

Actions (login required)

View Item