用人工神经网络模型获取采油指数的方法

APPROACH FOR PROVIDING RESERVOIR PRODUCTIVITY USING NEURAL NETWORKS

介绍了油藏工程中一种新的神经计算技术的实现方法——用人工神经网络模型获取开发初期分采时的采油指数；通过实例对该方法中的数据处理、模型确定及精度分析、预测效果进行了详尽地描述，实践表明该方法可对从事油田开发的现场研究人员提供借鉴。

In this paper,a model of artifical neural networks is used to get the productivity index in the initial period of development,which is a new neural calculating technique in oil reservoir engineering.A few example are provided to describe the processing of data,establishment of model,analysis of accuracy degree and the effect of prediction.It is tested that some lessons can be drawn from this way for the on the spot research workers working on oil field development.