摘要
本文系统地介绍了当前油井含水率在线检测方法,分析了油井含水率测量的影响因素,综述了其研究进展。为了提高油井含水率测量的准确性和稳定性,总结了近年来出现的基于多传感器融合技术的含水率检测方法,即应用偏最小二乘法(PLS)、神经网络、多维回归分析、支持向量机等方法来构建数据驱动模型。研究表明,结合传统检测方法和多传感器信息融合的原油含水率在线测量技术将是未来油井含水率测量的主要发展方向。
This paper systematically introduces the current on-line detection method of oil well water content, analyzes the influencing factors of oil well water content measurement, and summarizes its research progress. In order to improve the accuracy and stability of oil well water content measurement, the water content detection method based on multi-sensor fusion technology in recent years is summarized, namely partial least squares (PLS), neural network, multi-dimensional regression analysis and support vector machines and other methods to build a data-driven model. Research shows that the online measurement technology of crude oil moisture content combined with traditional detection methods and multi-sensor information fusion will be the main development direction of oil well water content measurement in the future.
作者
张紫琴
檀朝东
吴浩达
宋健
ZHANG Ziqin;TAN Chaodong;WU Haoda;SONG Jian(China University of Petroleum, Beijing Changping, 102249, China;Beijing Yandan Petroleum Technology Development Co, Ltd, Beijing Changping, 102200, China)
出处
《数码设计》
2018年第4期61-63,共3页
Peak Data Science
关键词
油井
含水测试方法
数据融合
进展
oil well
water content
test method
data fusion
progress