碎屑岩储层随机建模的基本理论与实践

A Basic Theory of Clastic Reservoir Stochastic Modeling and Its Practice

简要论述了储层随机建模的基本思想和方法体系 ,其核心是将空间上任一点的储层属性作为随机变量 ,进行后验条件累积分布函数的计算和输出值的蒙特卡罗模拟实现。根据实际地质条件和对模型要求的不同 ,模拟的方法包括高斯模拟、指示模拟、序贯模拟、布尔模拟、退火模拟等 5种类型。提出了一个适用于碎屑岩储层的建模流程 ,主要分为 3个基本过程 :1建立储层原型模型或储层地质知识库 ;2建立储层的随机模型 ;3进行储层的随机模拟。模拟时又采用了岩相模型建立和岩相控制下的储层参数模型建立两个基本步骤。通过一个油田的实例 ,具体介绍了上述过程和步骤的实现方法。建模的结果说明了该方法的可行性和必要性。

A basic thought and a method system are briefly described for stochastic reservoir modeling The key of the method is to take the reservoir attributes on an arbitrary point in space as random variables for calculating the post conditional accumulative distribution function and realizing Monte Carlo simulation of output values Based on the practical geologic conditions and different needs for a model, the simulation methods include 5 types, such as Gauss simulation, indicator simulation, sequential simulation, Boolean simulation and annealing simulation A procedure for clastic reservoir modeling is put forward, which includes:①establishing prototype reservoir model or a reservoir geologic knowledge base;②establishing a stochastic reservoir model;③stochastic reservoir modeling Two basic procedures for building a lithofacies model and a reservoir parameter model by controlling the lithofacies are deployed during the simulation The method for realizing the procedures and process mentioned above is introduced specifically through an oilfield case study The result of modeling shows that the method is feasible and necessary