ONE-DIMENSIONAL MAGNETOTELLURIC INVERSION MODELLING USING BAT ALGORITHM

• Main Author: Arif Raihannur, Muhammad
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/65773
• Institution: Institut Teknologi Bandung

The magnetotelluric method is a passive geophysical method that can be used to obtain subsurface resistivity variations. Subsurface resistivity variation values can be modeled to determine the actual subsurface conditions. Magnetotelluric data modeling can be done by forward modeling and inversion modeling. This study was conducted to test the one-dimensional magnetotelluric inversion modeling method using the Bat Algorithm algorithm developed by Xin-She Yang. Bat Algorithm is a population-based algorithm that simulates the echolocation of micro bats to solve mathematical problems. The one-dimensional magnetotelluric inversion modeling using the Bat Algorithm was tested using sounding curves of types A, H, K, and Q. The test was carried out with three-layers synthetic model parameter data with two treatments, namely without noise and with 5% noise. Tests were also carried out by modeling three-layer magnetotelluric data with four-layer guesses and four-layer magnetotelluric data with three-layer guesses to test the accuracy of the onedimensional magnetotelluric inversion modeling method using the Bat Algorithm algorithm. The results of testing the method on all types of sounding curves without noise show an error value in the range of 0.01-0.08%. Meanwhile, testing the method on all types of sounding curves with noise shows an error value in the range of 4-5%. The error value obtained shows that the one-dimensional inversion method of magnetotelluric data using the Bat Algorithm has a high level of accuracy. This claim is supported by the success of the one-dimensional inversion of magnetotelluric data using the Bat Algorithm in modeling three-layer magnetotelluric data with four-layer guesses and four-layer magnetotelluric data with three-layer guessing.