储层预测中的非线性反演方法(为《岩性油气藏》创刊而作)

Nonlinear inversion of the reservoir prediction——example of Seimpar Inversion

储层预测中非线性反演方法的研究是地球物理学中一个非常重要的研究领域。在对人工神经网络、分形分维等非线性方法研究的基础上,阐述了Seimpar测井参数反演技术,它是一种非线性反演方法,既可以避开褶积公式进行波阻抗反演,又可以在井约束下用地震记录反演测井参数曲线。能够解决常规地震反演难以解决的复杂地质问题,实际应用中取得了较好的地质效果。

Nonlinear inversion in the reservoir prediction is a very important field of geophysics. On the base of nonlinear methods, such as artificial neural network, fractal and fractal dimension etc., the authors expound that, as a method of nonlinear inversion, Seimpar Inversion could carry out wave impedance inversion without using convolution equation, and also inverse logging parameter by seismic records under the logging constrained. This method of inversion could resolve complex geological problems which are hardly to be solved by routine inversion, and it is applied well in practice.