地质统计学反演与常规反演在储层预测中的应用效果对比分析

Geostatistical Inversion and Conventional Inversion Application Effect Contrastive Analysis in Reservoir Prediction

针对砂泥岩薄互层非均质性强、单层砂体薄、储层识别困难的工区,进行了地质统计学反演,并与常规约束稀疏脉冲反演结果进行了对比分析,结果表明:常规约束稀疏脉冲波阻抗反演只能识别大套砂体,无法识别薄互层小砂体,而基于储层参数地质统计分析和随机模拟技术的地质统计学反演方法充分发挥了测井数据的垂向高分辨率和地震数据的横向高分辨率优势,可实现砂泥岩薄互层储层砂体分布的精细描述,变差函分析及马尔科夫链-蒙特卡罗随机模拟技术对于厚度在5m以下的小砂体也能较清晰地刻画。

Aimed at prospecting areas with mainly interbedded thin sandstone and mudstone, thin single sand mass and difficult to identify reservoirs carried out geostatistical inversion and conventional constrained sparse spike inversion contrastive analysis. The result has shown that the conventional constrained sparse spike inversion can only identify large set of sand mass and unable to identify interbedded thin and small sand mass. While the geostatistical inversion based on reservoir parameters geostatistical analysis and stochastic simulation technique can give full play to the predominance of well logging data vertical high resolution and seismic data transverse high resolution and realize accurate description of interbedded thin sandstone and mudstone reservoir sand mass distribution. The variograms analysis and Markov chain-Monte Carlo stochastic simulation technique can clearly describe small sand mass with thickness less than 5m.