基于混合遗传算法的叠前随机反演方法

发布时间：2018-06-17 03:41

本文选题：混合遗传算法 + 叠前随机反演　；参考：《中国石油大学学报(自然科学版)》2017年04期

【摘要】：针对常规随机反演方法计算效率低的问题,提出一种基于混合遗传算法的叠前随机反演方法。该方法充分利用测井资料中的高频信息,并以地震数据作为约束,首先通过快速傅里叶滑动平均(fast Fourier transform-moving average,FFT-MA)谱模拟算法进行随机模拟得到基于地质统计学的初始模型信息,随后结合提出的混合遗传算法对模拟结果进行快速优化,得到符合地下地质结构的反演剖面,实现对叠前弹性参数的反演。混合遗传算法避免了一般遗传算法常见问题,如收敛速度慢以及产生"早熟"现象,与模拟退火相结合能够快速收敛达到全局最优,保证了反演精度。数值试验结果表明,融入混合遗传算法的叠前随机反演方法,在充分利用叠前信息的同时可以保证反演结果有效收敛,并且与模型数据吻合较好,与传统的叠前反演方法相比具有较高的分辨率,在储层识别和油藏描述中起到了重要作用。
[Abstract]:A prestack stochastic inversion method based on hybrid genetic algorithm is proposed to solve the problem of low computational efficiency of conventional stochastic inversion methods. This method makes full use of the high frequency information in the logging data and takes seismic data as the constraint. Firstly, the initial model information based on geostatistics is obtained by random simulation using the fast Fourier transform-moving averaging FFT-MAA spectrum simulation algorithm. Then the simulation results are quickly optimized by using the hybrid genetic algorithm, and the inversion profiles consistent with the underground geological structure are obtained, and the inversion of prestack elastic parameters is realized. Hybrid genetic algorithm avoids the common problems of general genetic algorithm, such as slow convergence speed and "premature" phenomenon. Combined with simulated annealing, the hybrid genetic algorithm can quickly converge to the global optimum and ensure the inversion accuracy. The numerical results show that the prestack stochastic inversion method incorporating the hybrid genetic algorithm can make full use of prestack information and ensure that the inversion results are convergent effectively and are in good agreement with the model data. Compared with the traditional prestack inversion method, it has higher resolution and plays an important role in reservoir identification and reservoir description.
【作者单位】：中国石油大学地球科学与技术学院;海洋国家实验室海洋矿产资源评价与探测技术功能实验室;
【基金】：国家自然科学基金-石油化工基金联合重点项目(U1562215) 国家自然科学基金项目(41204085)
【分类号】：P618.13;P631.4

【相似文献】