One-dimensional magnetotelluric (MT) data inversion modeling using convolutional neural network

Abstract The magnetotelluric method is a geophysical commonly used to map subsurface resistivity. subsurface’s true resistivity generated by inversion of the data. Inversions carried out using conventional methods such as linear and global approaches have several limitations including need for an initial model, models trapped in local minima, large number iterations long computation time. To overcome drawbacks, this paper proposes invert one-dimensional data one deep learning methods, convolutional neural network, which heavily inspired human nervous system. This starts training network with amounts trained then receiving input form apparent generating thickness values instantly. has been tested on synthetic curves type A, H, K, Q. results show that could approach fairly small error extremely fast time without model guess iteration.