Research On Electromagnetic Induction And The Propagation Of Low Frequency Electromagnetic Wave In An Inhomogeneous Medium

With more and more oilfields being developed in the middle and later period, in order to detect subtle reservoir of oil and gas, the technology of drilling and well logging has undergone rapid progress. However, some new challenges also arise, such as the response characteristic analysis of induction logging tools in highly deviated or horizontal wells, the electric anisotropic effect on the response of well logging, the measurement of the anisotropic resistivity of formation, the evaluation and design of new types of well logging sonde, etc. It has been shown that the numerical simulation and experiment research of logging tool can serve as an efficient tool to solve these problems.This paper is just carried out in the above-mentioned background, whose content includes forward modeling method and inversion method, and also the experiment research of multicomponet induction logging tool. The forward modeling method includes three-dimensional numerical method and also one-dimensional analytic method and two-dimensional semi-analytical and semi-numerical method. These three methods have its own advantages and disadvantages, and we could selectively use them in accordance with the reality in numerical simulation analysis.Firstly, Green's Functions for planar and cylindrical layered electric anisotropic media are derived in this paper, and they are applied to the modeling of induction logging problems in one-dimensional layered formation. Its advantage is fast and accurate, and is able to model excited source of any direction, and is especially applicable to deviated anisotropic formatiom. In addition, we introduce the numerical mode method (NMM) to simulate the response of induction and propagation logging in two-dimensional inhomogeneous media.In this paper, we study two kinds of numerical methods used to simulate the response of induction and electromagnetic propagation logging tools in three-dimensional complex media: finite element method and volume integral equation method. Considering the characteristics of electromagnetic logging model, some innovations are made in the modeling, which is helpful for decreasing the unknowns and accelerating the solution speed. Mixed order basis function is applied in three-dimensional finite element modeling, and a fast method of solving linear equation is incorporated into the implementation of the FEM algorithm. The results of multiple measure points can be got by solving linear equation only once. A fast and effective numerical computational method is developed to deal with the volume integral equation used in the modeling of three-dimensional electromagnetic logging problems.Based on the above studies, we simulate and analyze two kinds of representative electromagnetic logging tools: Logging-while-drilling (LWD) electromagnetic wave propagation logging tool and multicomponent induction logging tool. Especially, the response of multicomponent induction logging tool is simulated and analyzed comprehensively. Differences of response in axial and transversal coils as well as the detective characteristics of axial and transversal coils are investigated. Some methods for decreasing the borehole and shoulder effects on the response of multicomponent induction logging are gived. The above analysis is useful for the explanation of logging data and optimal design of new induction logging tools.We preliminarily study the inversion technology of multicomponent induction logging data, and put forward the fast inversion method based on the forward modeling with the analytic formulae. Linear search is incorporated into the inversion method to improve the speed of the convergence rate of inversion.Based on the above theories, the experiment of multicomponent induction logging is designed, which includes the designs of well model, coils-array configuration, method of measurement and system of measurement. Finally, the comparision of experimental result and simulated result illustrates that the experiment design is correct and feasible. In addition, the inversion method above is applied to experimental data inversion, and the results show that the horizontal and vertical resistivity of anisotropic formation can be obtained by data inversion of multicomponent induction logging.