基于大数据的油气集输系统生产能耗时序预测模型

发布时间：2018-06-14 10:34

本文选题：集输系统 + 混沌时序　；参考：《石油学报》2016年S2期

【摘要】：针对集输系统组成关系多、系统行为复杂、子系统之间以及系统与环境之间的关联程度高、耦合性强、易产生故障和能耗高等特点,基于油气集输生产过程中积累的温度、压力、流量、设备工作制度、能耗等海量数据,建立了集输数据粒度模型,实现了基于热能利用率、单位液量能耗等多目标、多变量时序的集输系统生产能耗预测。针对不同时间粒度(如日、月、年等)、不同空间粒度(如井组、区块、油田等)、不同集输方式粒度(如单相输、油-气-水混输),建立了多变量时序混沌能耗预测模型;构造了粒关联规则模式挖掘算法。以大港油田A集输系统为例,研究了集输生产系统的能耗因素粒之间的关联关系;预测了集输生产参数调整对系统未来能耗变化,获得集输系统效率和能耗的预警。
[Abstract]:In view of the characteristics of the gathering and transportation system, such as multi-component relations, complex system behavior, high degree of correlation between subsystems and between systems and environment, strong coupling, easy to produce faults and high energy consumption, etc., based on the accumulated temperature in the process of oil and gas gathering and transportation production. The granularity model of gathering and transportation data is established, such as pressure, flow rate, equipment working system, energy consumption and so on. The prediction of energy consumption of gathering and transportation system based on multi-objective and multi-variable time series, such as thermal energy utilization ratio and energy consumption per unit liquid quantity, is realized. According to different time granularity (such as day, month, year and so on), different spatial granularity (such as well group, block, oil field, etc.), different granularity (such as single phase transportation, oil-gas-water mixed transportation), the chaotic energy consumption prediction model of multivariable time series is established. A mining algorithm of granular association rules pattern is constructed. Taking A gathering and transporting system in Dagang Oilfield as an example, the correlation between energy consumption factors of gathering and transportation production system is studied, and the change of energy consumption in the future by adjusting gathering and transportation production parameters is predicted, and the early warning of gathering and transportation system efficiency and energy consumption is obtained.
【作者单位】：中国石油大学石油工程学院;中国石油大港油田公司采油工艺研究院;
【分类号】：TE863

【参考文献】