摘要
马尔可夫链蒙特卡罗(MCMC)方法是一种基于全局最优化技术进行地质统计学求解的方法。综合考虑地震数据、测井数据、岩性数据,对波阻抗、岩性概率分布的马尔科夫链进行蒙特卡罗模拟,从初始模拟结果出发,利用MCMC算法,模拟出多个波阻抗曲线,通过合成地震记录与实际地震记录剖面相关性,确定合理的储层参数,反演出空间上的波阻抗体及岩性体。最重要的质控手段是抽井检验,以保证模型的稳定性。MCMC方法在大芦湖油田樊18-3区块沙三中、沙三下亚段浊积砂体储层预测中取得了良好的效果。
The method of Monte Carlo Simulation of Markov Chain(MCMC) was a kind of geostatistical method containing multi-information and mainly based on the global optimum taking account of seismic data,logging,lithological data and probability distribution concerning AI and lithology. Starting from the preliminary simulation result,the MCMC method simulated several AI curves,and got the reasonable reservoir parameters by means of the correlation between synthetic seismogram and seismic section. The inversion of the AI and lithologic cube are obtained. The most important means of quality controlling is the inspection of selected well to guarantee the stability of the model. The method of MCMC performs better in the application of the reservoir prediction of turbidity sand body in the middle and lower members of Es3 Block Fan 18-3 of Daluhu Oilfield.
出处
《石油天然气学报》
CAS
CSCD
北大核心
2010年第2期249-252,266,共5页
Journal of Oil and Gas Technology
关键词
马尔可夫链蒙特卡罗模拟
岩性分析
储层标定
稀疏脉冲反演
直方统计
变差分析
Markov Chain Monte Carlo simulation
lithologic analysis
reservoir prediction
constrained sparse pulse inversion
hisotogram
statistics
variation analysis
