特征聚类在油田测试方案优化中的研究

发布时间：2018-06-17 13:15

本文选题：特征 + 聚类和分类　；参考：《计算机工程与应用》2017年21期

【摘要】：针对油田注产剖面动态测试在选井上没有一个合适参考标准的问题,提出利用基于特征的聚类将油井按照生产状况进行先聚类再分类的方法。首先对油井生产时间序列数据选择处理,然后提取序列特征进行聚类,并把聚类结果划分等级,最后制定油田测试优化方案。实验结果表明,提取的时间序列特征能很好地表征油井生产波动情况,获得较好地分类效果,对指导油田测试有重要意义。
[Abstract]:In order to solve the problem that there is no suitable reference standard for the dynamic testing of injection-production profile in oil fields, a method of clustering and classifying oil wells according to production conditions by feature-based clustering is proposed. Firstly, the production time series data are selected and processed, and then the sequence features are extracted for clustering, and the clustering results are classified. Finally, the oil field test optimization scheme is formulated. The experimental results show that the extracted time series features can well characterize the oil well production fluctuation and obtain a good classification effect. It is of great significance to guide oil field testing.
【作者单位】：中国石油大学地球物理与信息工程学院;中国石油大学油气数据挖掘北京市重点实验室;大庆油田测试技术服务分公司;
【分类号】：TE34;TP311.13

【相似文献】