石油学报 ›› 2013, Vol. 34 ›› Issue (1): 140-144.DOI: 10.7623/syxb201301017

• 油田开发 • 上一篇    下一篇

基于沉积过程建模算法Alluvsim的改进

李少华 1  刘显太 2  王 军 2  龚蔚青 2  卢文涛 3   

  1. 1.长江大学地球科学学院剩余资源研究组 湖北荆州 434023;2.中国石化胜利油田公司地质研究院 山东东营 257015; 3.中国石化江汉油田公司勘探开发研究院 湖北武汉 430223
  • 收稿日期:2012-05-04 修回日期:2012-08-27 出版日期:2013-01-25 发布日期:2013-04-09
  • 通讯作者: 李少华
  • 作者简介:李少华,男,1972年8月生,2003年获中国石油勘探开发研究院博士学位,现为长江大学地球科学学院教授、硕士生导师,主要从事地质统计学与储层建模方面的科研与教学工作。
  • 基金资助:

    国家重大科技专项(2011ZX05011-001)与国家自然科学基金项目(No.41272136)资助。

Improvement of the Alluvsim algorithm modeling based on depositional processes

LI Shaohua 1  LIU Xiantai 2  WANG Jun 2  GONG Weiqing 2  LU Wentao 3   

  • Received:2012-05-04 Revised:2012-08-27 Online:2013-01-25 Published:2013-04-09

摘要:

对基于沉积过程的河流相储层随机建模算法Alluvsim的基本概念及主要实现步骤进行了描述,与传统的储层随机建模方法相比,基于沉积过程的建模方法更有效地将与沉积过程有关的地质信息以及先验的地质知识整合到建模过程中,能够更加真实地再现储层构型要素,如河道、点坝、天然堤、决口扇等的几何形态和内在成因上的联系,进而建立更为真实的地质模型。针对建模算法Alluvsim无法刻画点坝砂体内部构型的不足,对该算法进行了改进,实现了点坝内部构型的模拟,改进后的算法能够灵活地控制点坝侧积层的倾角、延伸长度、频率等对流体运动起重要作用的关键参数,实现了河道的部分或全部废弃。并分析了基于沉积过程的随机建模算法存在的一些不足。

关键词: 沉积过程, 点坝, 随机建模, 算法改进, 河流

Abstract:

This paper introduced basic concepts, principles and modeling steps of the Alluvsim algorithm for the depositional-process-based reservoir stochastic modeling. Compared with traditional stochastic modeling methods, this modeling method, which efficiently integrates various kinds of data with experts’ knowledge based on depositional processes, enables a more geologically reproduction of the geometry and relationship of architectural elements of reservoirs, such as channel, point bar, levee and crevasse splay. Generally, the Alluvsim algorithm can not exactly describe internal architectural elements of a point bar, such as lateral-accretion units and associated mud drapes. Thus, an improvement was made to characterize the key parameters of mud drapes in a point-bar, including dip angle, extending length and frequency, which significantly affect the movement of fluids. In addition, channels could be partly or wholly abandoned in the new method. At last, some limits of the present depositional-process-based modeling algorithm were discussed.

Key words: depositional process, point bar, stochastic modeling, algorithm improvement, channel