Fast hierarchical inversion for borehole resistivity measurements in high-angle and horizontal wells using ADNN-AMLM

With the rapid development of deep learning, intelligent scheme is gradually introduced to solve various nolinear inverse problems. In this paper, we combine an efficient adaptive neural network (ADNN) framework with modified Levenberg-Marquardt (AMLM) algorithm based on three-layer inversion model exact formation resistivity and invasion depth from array laterolog measurements. ADNN presented in paper can realize 2D/3D fast forward modeling laterolog. AMLM hierarchical are adopted improve anti-noise ability convergence complex logging environments, which realizing accurate reconstruction longitudinal profile HA/HZ wells. The numerical simulation shows that only takes 0.021s for each point, maximum relative error less than 2%. Three-layer eliminate effect surrounding bed accuracy thinly layered formation. between inverted results truth 3%. effectively suppress influence noise, 10 steps achieve convergence.