论文部分内容阅读
流动单元的准确划分不仅是建立精细地质、搞清剩余油分布的必要条件,而且关系到后期挖潜剩余油措施的选择.本文应用BP神经网络算法,以取芯井参数聚类分析结果为学习样本,对非取芯井进行流动单元划分,实例表明与常规的判别分析方法相比,在划分结果上基本一致,但是BP神经网络算法具有更准确的样本学习能力,不存在回判错误的可能,为单井流动单元划分提出了新的思路.
The accurate division of flow units is not only the necessary condition to establish the fine geology and to find out the remaining oil distribution, but also relates to the choice of remaining oil tapping potential in the later stage.Based on the BP neural network algorithm and the cluster analysis results of coring well parameters, , The flow cell division of non-coring wells is demonstrated. The examples show that the results are basically the same as those of the conventional discriminant analysis. However, the BP neural network algorithm has a more accurate sample learning ability, A new idea is put forward for dividing single flow unit.