摘要
随着油田智能化发展的需要,传统的故障诊断、动液面、产量计量算法已不能满足油井问题在实时性和准确性方面的要求。针对目前国内外油井监测与分析系统中存在的问题,对系统核心算法故障诊断、动液面、产量计量进行修正。其中,故障诊断采用基于关联度函数修正的方法,并给出一种专家知识库实时更新完善策略;动液面算法采用物元分析的方法,通过建立物元模型确定油井的最优多相流算法;产量计算算法采用均值滤波的方法,通过消除示功图中复杂的频率部分精确求取示功图有效冲程。对以上修正算法进行了仿真,提高了油井问题诊断的准确性和诊断效率,为智能油田提供理论基础。
With the development of intelligent oilfield, the traditional fault diagnosis, dynamic surface and yield measurement algorithm could not meet the requirements of real-time and accuracy in the oil well problem. Aimed at the problems existing at home and abroad oil well monitoring and analysis system, this paper corrected the system core algorithm of fault diagnosis, dy- namic surface and yield measurement. Among them, the fault diagnosis method was based on correlation function correction, and presented a real time updating strategy of expert knowledge base. q＇he method of matter element analysis was used to determine the optimal multiphase flow in oil wells by using matter-element model The yield measurement algorithm used the method of mean fiher to get the effective stroke of the indicator diagram by eliminating the complex frequency part of the indicator dia- gram. And it simulated the above correction algorithm. They improve the accuracy and efficiency of the diagnosis of oil well problems, provide theoretical basis for intelligent oilfield.
出处
《计算机应用研究》
CSCD
北大核心
2018年第2期428-431,共4页
Application Research of Computers
基金
国家"863"计划资助项目(2015AA043901)
中国科学院先导专项资助项目(XDA06020500)
关键词
智能油田
故障诊断
动液面
产量计量
intelligent oilfield
fault diagnosis
dynamic surface
yield measurement
