Schemes for Improving Efficiency of Pixel-Based Inversion Algorithms for Electromagnetic Logging-While-Drilling Measurements

We present an application of the two-and-half dimensional (2.5D) multiplicative-regularized Gauss-Newton inversion method for the interpretation of the directional resistivity measurements [electromagnetic logging-while-drilling (LWD) measurements]. In this work, an inversion algorithm is employed for obtaining a two-dimensional (2D) resistivity distribution (pixel-based) of the subsurface. Several modifications on both forward and inversion algorithms have been implemented to minimize the computational time and memory usage as well as to improve the accuracy of the algorithms. Numerical examples demonstrate the feasibility of using this 2.5D pixel-based inversion algorithm for electromagnetic LWD data.