| 基于Metropolis优化的叠前全局迭代地质统计学反演方法 |
| |
| 引用本文： | 赵晨,张广智,张佳佳,郗诚,肖亚楠.基于Metropolis优化的叠前全局迭代地质统计学反演方法[J].地球物理学报,2020,63(8):3116-3130. |
| |
| 作者姓名： | 赵晨 张广智 张佳佳 郗诚 肖亚楠 |
| |
| 作者单位： | 1. 中国石油大学(华东)地球科学与技术学院, 青岛 266580;2. 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, 青岛 266071;3. 中国石油西南油气田公司勘探开发研究院, 成都 610041 |
| |
| 基金项目： | 国家科技重大专项（2016ZX05027004-001），国家自然基金项目（41674130）和中央高校基本科研业务费专项资金（18CX06020A）联合资助. |
| |
| 摘 要： | 叠前地质统计学反演将随机模拟与叠前反演相结合,不仅可以反演各种储层弹性参数,还提高了反演结果的分辨率.基于联合概率分布的直接序贯协模拟方法可以在原始数据域对数据进行模拟,不需要对数据进行高斯变换,拓展了地质统计学反演的应用范围;而联合概率分布的应用确保了反演参数之间相关性,提高了反演的精度.本文将基于联合概率分布的直接序贯协模拟方法与蒙特卡洛抽样算法相结合,参考全局随机反演策略,提出了基于蒙特卡洛优化算法的全局迭代地质统计学反演方法.为了提高反演的稳定性,我们修改了局部相关系数的计算公式,提出了一种新的基于目标函数的优化局部相关系数计算公式并应用到协模拟之中.模型测试及实际数据应用表明,该方法可以很好的应用于叠前反演之中.
|
| 关 键 词： | 地质统计学反演 直接序贯模拟 优化局部相关系数 叠前AVA反演 蒙特卡洛优化算法 |
| 收稿时间： | 2018-12-05 |
Prestack global iteration geostatistical inversion method based on metropolis sampling algorithm |
| |
| ZHAO Chen,ZHANG GuangZhi,ZHANG JiaJia,XI Cheng,XIAO YaNan.Prestack global iteration geostatistical inversion method based on metropolis sampling algorithm[J].Chinese Journal of Geophysics,2020,63(8):3116-3130. |
| |
| Authors: | ZHAO Chen ZHANG GuangZhi ZHANG JiaJia XI Cheng XIAO YaNan |
| |
| Institution: | 1. China University of Petroleum(East China), Qingdao 266580, China;2. Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266071, China;3. Research Institute of Exploration and Development, PetroChina Southwest Oil and Gas Field Company, Chengdu 610041, China |
| |
| Abstract: | The prestack geostatistics inversion method combining stochastic simulation with prestack seismic inversion can invert for various elastic properties and improve the resolution of inversion results. The direct sequential co-simulation method based on the joint probability distribution can simulate data in the original data domain without Gaussian transformation, which expands the application scope of geostatistical inversion. In addition, the application of the joint probability distribution ensures the correlation between the inverted properties and improve the accuracy of the inversion. In this paper, combining the metropolis sampling algorithm with the direct sequential co-simulation method, the global iteration geostatistical inversion method based on Metropolis sampling algorithm is proposed by referring the global stochastic inversion strategy. In order to improve the stability of inversion, we modified the calculation formula of local correlation coefficient, and proposed a new calculation formula of optimized local correlation coefficient based on the objective function. Model testing and application to actual data demonstrate that this method works well in prestack inversion. |
| |
| Keywords: | Geostatistical inversion Direct sequential simulation Optimized local correlation coefficient Prestack AVO inversion Metropolis sampling algorithm |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《地球物理学报》浏览原始摘要信息 |
| 点击此处可从《地球物理学报》下载免费的PDF全文 |
|
