An Intelligent Inversion Method for Azimuth Electromagnetic Logging While Drilling Measurements

Azimuth electromagnetic (EM) Logging While Drilling (LWD) tools play an important role in geological steering and reservoir evaluation. Its measuring response is amplitude ratio (ATT) phase difference (PS), which can’t directly reflect formation information. To obtain direct information such as resistivity boundary, accurate efficient inversion method essential. However, the existing methods (i.e., iterative method) have some problems, slow computation speed, dependence on initial value selection easy to be trapped by local minimum. Therefore, this paper proposes intelligent for azimuthal EM LWD measurements based U-net deep learning network framework. Firstly, analytical solution used generate amounts of data. Those samples are divided into training test sets a 9:1 ratio, testing network, respectively. Then, parameters constantly adjusted during ensure its performance. Finally, trained utilized invert sets, results compared with forward model. The study’s demonstrate that capable efficiently precisely inverting both isotropic anisotropic formations, single sample being inverted under 0.05 seconds. noise layer can improved successfully noisy data, leading good robustness. In addition, has applicability complex formations. These consequences highlight significant potential use applications.