摘要
在油气集输系统中,采用机器学习技术与油气集输过程相结合的方式,进行集输系统的能耗预测研究。分析了集输过程主要耗能参数,定量化描述了各耗能参数与能耗之间的变化关系,研究建立耗能定量化预测大数据模型,实现了集输系统能用能定量化预测分析。以胜利油田A集输系统为例,对建立的模型进行了验证。结果表明:采用此方法可以大大降低建模难度,缩短设计周期,降低工作量,进而为油田集输方案的调整与节能优化运行提供合理的参考依据。
In the oil and gas gathering and transportation system,the combination of machine learning technology and the oil and gas gathering and transportation process was adopted to conduct the energy consumption prediction research of the gathering and transportation system.The main energy consumption parameters of the gathering and transmission process were analyzed,and the relationship between the energy consumption parameters and the energy consumption was described quantitatively.A large data model for quantitatively predicting the energy consumption was researched and established.Taking A gathering and transportation system in Shengli oilfield as an example,the established model was verified.The results showed that,using this method could greatly reduce the modeling difficulty,shorten the design cycle,and reduce the workload.The prediction model can provide reasonable reference for the adjustment of the oilfield gathering and transportation scheme and the optimization of energy saving operation.
作者
李振泉
张丁涌
周长敬
王兴武
安学先
高华
孙东
刘文聪
闫恩祥
李红强
孙秀玲
杨文辉
张腾
梁莹
王增光
LI Zhen-quan;ZHANG Ding-yong;ZHOU Chang-jing;WANG Xing-wu;AN Xue-xian;GAO Hua;SUN Dong;LIU Wen-cong;YAN En-xiang;LI Hong-qiang;SUN Xiu-ling;YANG Wen-hui;ZHANG Teng;LIANG Ying;WANG Zeng-guang(Sinopec Shengli Oilfield Company,Dongying 257000,China)
出处
《当代化工》
CAS
2020年第12期2818-2821,共4页
Contemporary Chemical Industry
基金
中国石化科技攻关项目(2018—2020),油田能源优化管控关键技术研究(项目编号:318016-11)
关键词
集输系统
机器学习
能耗
大数据
预测模型
Gathering and transportation system
Machine learning
Energy consumption
Big data
Prediction model
