Inversi Data Vertical Electrical Sounding Konfigurasi Schlumberger Menggunakan Algoritma Ensemble Kalman Inversion (EKI)

Insani, Wahyuning Fitri Mulyaning (2022) Inversi Data Vertical Electrical Sounding Konfigurasi Schlumberger Menggunakan Algoritma Ensemble Kalman Inversion (EKI). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 01111840000088-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Vertical Electrical Sounding (VES) konfigurasi Schlumberger merupakan salah satu teknik pengukuran metode geolistrik yang umumnya digunakan untuk menentukan model 1-D bawah permukaan bumi untuk kepentingan eksplorasi air tanah. Salah satu cara untuk menghasilkan model 1-D dari hasil data VES adalah dengan teknik inversi. Persamaan matematika yang menghubungkan antara data pengukuran VES dan model resistivitas 1-D bersifat non-linier. Sehingga, apabila dilakukan inversi dengan pendekatan lokal berpotensi menimbulkan konvergensi ke minimum lokal. Untuk proses inversi data VES, algoritma dengan pendekatan optimum global dibutuhkan. Salah satu algoritma pendekatan optimum global adalah algoritma Ensemble Kalman Inversion (EKI). EKI merupakan algoritma berbasis ensembel/populasi yang menggantikan matrik jacobian dalam algoritma Gauss-Newton dengan matrik kovarian dari ensembelnya. Algoritma EKI telah diterapkan pada inversi data VES sintetik serta data lapangan VES. Solusi optimum direpresentasikan oleh kandidat solusi dari sejumlah model yang juga cocok dengan data observasi dengan toleransi fungsi objektif tertentu. Setelah dilakukan uji coba pada beberapa tipe model bawah permukaan, diketahui bahwa EKI mampu mendapatkan harga parameter model dengan kondisi yang mendekati parameter model data sintetik. Sedangkan hasil inversi data VES yang terukur di tanggul lumpur Sidoarjo (LUSI) menunjukkan bahwa EKI mampu mengidentifikasi ketebalan tanggul dengan akurat.
=================================================================================================================================
Vertical Electrical Sounding (VES) with the Schlumberger configuration is one of the geoelectrical measurement techniques that is generally used to determine 1-D models of the earth's surface for the benefit of groundwater exploration. One way to generate a 1-D model from the VES data results is the inversion technique. The mathematical equation that relates the VES measurement data and the 1-D resistivity model is non-linear. Thus, if an inversion is carried out with a local approach, it may cause convergence to a local minimum. For the VES data inversion process, an algorithm with a global optimum approach is needed. One of the global optimum approach algorithms is the Ensemble Kalman Inversion (EKI) algorithm. EKI is an ensemble/population-based algorithm that replaces the Jacobian matrix in the Gauss- Newton algorithm with the covariance matrix of the ensemble. The EKI algorithm has been applied to the inversion of synthetic VES data as well as VES field data. The optimum solution is represented by a candidate solution from a number of models that also fit the observation data with a certain tolerance of the objective function. After conducting trials on several types of subsurface models, it is known that EKI is able to obtain model parameter values with conditions that are close to synthetic data model parameters. Meanwhile, the results of the inversion of VES data measured at the Sidoarjo mud embankment (LUSI) indicate that EKI is able to accurately identify the thickness of the embankment.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Ensemble Kalman Inversion, Vertical Electrical Sounding, Resistivitas, Ensemble Kalman Inversion, Vertical Electrical Sounding, Resistivity.
Subjects:	Q Science > QC Physics
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	08 Oct 2025 07:01
Last Modified:	08 Oct 2025 07:01
URI:	http://repository.its.ac.id/id/eprint/128513

Actions (login required)

View Item