Inverting reservoir parameters in a two-phase fractured medium with a niche genetic algorithm

基于小生境遗传算法的双相裂隙介质储层参数反演(英文)

In this paper we calculate a synthetic medium surface displacement response that is consistent with real measurement data by applying the least-square principle and a niche genetic algorithm to the parameters inversion problem of the wave equation in a two-phase medium. We propose a niche genetic multi-parameter （including porosity, solid phase density and fluid phase density） joint inversion algorithm based on a two-phase fractured medium in the BISQ model. We take the two-phase fractured medium of the BISQ model in a two- dimensional half space as an example, and carry out the numerical reservoir parameters inversion. Results show that this method is very convenient for solving the parameters inversion problem for the wave equation in a two-phase medium, and has the advantage of strong noise rejection. Relative to conventional genetic algorithms, the niche genetic algorithm based on a sharing function can not only significantly speed up the convergence, but also improve the inversion precision.

为了更好地寻找地下复杂介质中的油气,本文针对双相介质波动方程参数的反演问题,根据介质表面理论合成的位移响应应与实际测量数据相一致的原则,引入最小二乘原理和小生境遗传算法,建立起了基于BISQ模型双相裂隙介质的小生境多参数(孔隙度、固相密度和流相密度)联合反演算法。最后,以基于BISQ模型的二维半空间双相裂隙介质模型为例,进行了数值反演分析。数值分析结果表明:本文方法对双相介质波动方程参数反演问题求解方便,且具有较强的抗"噪声"干扰性能;相对常规遗传算法,基于共享函数的小生境遗传算法不仅可以明显加快收敛速度,而且还可以有效提高反演精度。