利用神经网络建立储层宏观参数动态模型--以胜坨油田二区为例

建立描述和表征储层宏观参数随油田开发过程发生动态变化的物理和数学模型,可有效提高油田管理水平及最终采收率。以长期注水开发的胜坨油田二区沙河街组二段第8砂层组第3小层三角洲相储层为例,在研究了储层参数变化规律的基础上,利用神经网络建立了表征储层宏观参数变化的动态模型及数学表达式。该模型可有效地预测不同开发阶段储层宏观参数的变化过程和变化规律,为油田开发提供了科学依据。

Dynamic model of reservoir macro - parameters built by neural network-taking the second block of Shengtuo Oilfield as an example

The physical and mathematical model,which were built up to describe and characterize the dynamic variety of reservoir macro-parameters along with oilfield development, will advance oilfield management level and ultimate recovery factor effectively. Change rule of delta reservoir parameters were researched in layer 8 of Es2 in the second block of Shengtuo Oilfield exploited by waterflooding for a long time. Dynamic model and mathematical expression characterized the reservoir macro-parameters variety were built up using neural network. This model can be used to effectively predict the process and rule of reservoir parameter changes at different development stages, which could provide scientific basis for oilfield development.