最大熵在感应测井反演中的应用

The Application of Maximum Entropy to the Inversion of Induction Log Data

常规的反演方法是对感应测井资料进行线性反褶积。由于感应测并获得的响应曲线是地层非线性时变滤波的结果,感应测井响应随地层电导率发生非线性变化。本文把熵的概念引入到感应测井反演中。用最大熵方法分别设计出实部信号和虚部信号所对应的反滤波因子,两条反映测井响应随电导率非线性变化的非线性函数也被拟合出来。文章最后用设计出的滤波因子及被拟合出的非线性函数对模拟相量感应测井资料进行了相量反褶积。处理结果表明,围岩效应和趋肤效应基本被消除。本方法与常规的三点反褶积法相比,克服了过聚焦现象,提高了精度。

The traditional invesion is the linear deconvolution of induction log data. The induction log responsecan be nonlinearly changed due to it is result of nonlinear and time-variable filter. This paper introduces theconcept of entropy to the inversion of induction log data. The inversion filter factor correspoding to inphaseand quadrature signals are designed by the method of the meximum entropy respectively, and then two non-linear functions are fitted so as to accomodate nonlinear tool response. The phasor deconvolution of the sim-ulated induction log proves the shoulder effect and skin effect can be substantively eliminated. Compared withtraditional three-point deconvolution, it doesn't have high resistivity streaks and is of high precision.