Fast Algorithm Of The Explicity Sensitivity Matrix And Parametric Iterative Inversion Of LWD Azimuthal Electromagnetic Data In Layered Anisotropic Formation

In this paper,a new type of azimuth logging while drilling(LWD)technology that can monitor and realize geosteering at any time through drill collars,combined with the process of contemporary oil exploration and development,and is based on transmission line theory(TLM)and related The formula studies and establishes the analytical algorithm of the dyadic Green's function of the magnetic current source in the layered anisotropic medium,and obtains the analytical solution of the electromagnetic field in the uniform infinite medium,and then uses the Fourier transform method and the transmission line theory to analyze the current source and the magnetic field The calculation method of the flow source dyadic Green function derives the expression of the dyadic Green function in the frequency domain.The perturbation theory and Fréchet derivative are used to further derive the analytic relational expressions of the electromagnetic field changes of horizontal conductivity,vertical conductivity,formation interface depth,and micro dip perturbation of the formation,and to establish a fast calculation of the sensitivity matrix of the electromagnetic wave response of orthogonal azimuth while drilling Methods,the results of the orthogonal azimuth logging while drilling tool were obtained,and then the change rule of the sensitivity coefficient under different conditions was investigated to confirm the detection effect of the drilling while drilling tool on the formation boundary.The azimuth-while-drilling electromagnetic conductivity logging data determines the real-time formation conductivity around the drill collar,the position of the drill collar from the upper and lower formation boundaries,and the relative inclination angle of the borehole.At the same time,the theoretical synthetic logging data is calculated and these all results are displayed in real time,providing a reasonable basis for geosteering and reservoir evaluation.Due to the fact that the thickness of the formation is thin and the influence of the surrounding rock layer is still relatively large,it is necessary to use the inversion technology to eliminate the influence of the surrounding rock layer,so on this basis,a fast forward modeling of the azimuth LWD electromagnetic logging database The method is combined with one-dimensional multi-parameter regularized iterative inversion technology.The2N-dimensional data sequence vector composed of azimuth while drilling electromagnetic wave logging data is used to represent the conductivity,layer interface and relative dip in the layered anisotropic layered formation model And the azimuth angle,the minimum value of the parameterized inversion target function can be obtained,and then the inverse model apparent conductivity parameters can be obtained by comparing with the forward modeling database,so that the longitudinal conductivity and lateral conductivity of the several formations around the drill collar The physical parameters such as the layer interface position and azimuth angle can be determined,which can solve the difficulty of thin local layer thickness and the influence of surrounding rock layers.The electromagnetic front-of-view logging data while drilling is less affected by the drill collar due to the larger distance between receiving and transmitting.Based on the one-dimensional multi-parameter inversion method and the omnidirectional scanning of the formation interface,the theoretical synthesis data and electromagnetic while drilling The best fit of the forward-looking logging data,so it can detect a farther distance.The results of computer numerical simulation show that the sensitivity matrix numerical curve will change obviously in advance when it encounters the layer boundary,which proves that it has good edge detection ability.The numerical curve of the parametric iterative inversion theory can reflect that the extraction of horizontal conductivity at different inclination angles is relatively accurate,and the extraction of vertical conductivity is relatively biased,but these errors can be accepted in the actual application process.