基于机理仿真的抽油机有杆泵举升系统生产参数优化设计  

Optimization Design of Production Parameters of Rod pump LiftingSystem of Pumping Unit based on Mechanism Simulation

作　　者：腾卫卫 李杰[1] 马文元 张辉[1] 冯钢 檀朝东[2,4] 张文浩 TENG Weiwei;LI Jie;MA Wenyuan;ZHAGN Hui;FENG Gang;TAN Chaodong;ZHANG Wenhao(Engineering Technology Research Institute,PetroChina Xinjiang Oilfield Company,Karamay,Xinjiang 834000,China;China University of Petroleum(Beijing),Beijing 102249,China;Xi′an Supcon World Technology Development Co.,Ltd.,Xi′an,Shaanxi 710016,China;School of Information Science and Engineering,China University of Petroleum(Beijing),Beijing 102249,China)

机构地区：[1]中国石油新疆油田分公司工程技术研究院,新疆维吾尔自治区834000 [2]中国石油大学(北京)石油工程学院 [3]西安中控天地科技开发有限公司 [4]中国石油大学(北京)信息科学与工程学院

出　　处：《钻采工艺》2024年第1期108-114,共7页Drilling & Production Technology

基　　金：国家自然科学基金项目“基于大数据解析的大规模非集输油井群生产及拉运调度协同优化”(编号:51974327)。

摘　　要：基于传统经验法和预测模型法的抽油机有杆泵举升系统设计方法在实际应用中要么过于简单要么过于复杂,设计方案的可靠性还有待提升。因此,应用大数据驱动技术从海量抽油机井举升系统设计方案中探索规律,推荐以提高产量和节能降耗为目标的抽汲参数优化设计最优方案,将提升抽油机井举升系统设计方法的科学性和可操作性。文章分析了抽油机井产量主控因素,将软件仿真和现场获取的抽油机井举升系统设计样本进行聚类分析融合得到新的样本库,建立了抽油机井举升系统设计方案的深度学习模型,研究了基于匹配算法的方案相似度模型,开发了抽油机井有杆泵举升系统生产参数设计推荐系统。根据实例计算验证,该系统能够根据待设计井的地质特征和开发特征,从海量样本中匹配到与地质油藏条件相似度高且运行效果良好的举升设计方案,推荐方案的举升参数及举升效果与原有方案相比有较大的提升,证明了计算模型的实际可行性,为抽油机举升系统优化设计提供了新思路及方法。The design method of rod pump lifting system based on traditional experience method and prediction model method is either too simple or too complicated in practical application,and the reliability of the design scheme needs to be improved.The application of big data driven technology to explore the law of mass pumping well lifting system design scheme,and recommend the optimal swabbing parameter optimization design scheme with the goal of increasing production and reducing energy consumption,will effectively improve the scientific and operational design method of pumping well lifting system.This article analyzes the main control factors of pumping well production,and combines the software simulation and on-site collection of pumping well lifting system design samples for cluster analysis to obtain a new sample library.A deep learning model for pumping well lifting system design scheme is established,and a scheme similarity model based on matching algorithm is studied.A parameter design recommendation system for pumping well lifting system is developed.The calculation example verified that this system can match and find the lifting design scheme with high similarity of geological reservoir conditions and good operation effect from a large number of samples for recommendation,according to the physical properties and development characteristics of the geological reservoir to be designed.The analysis shows that the lifting parameters and lifting effect of the recommended lifting system design have been greatly improved in the original scheme.The feasibility of this algorithm is proved,which provides a new idea and method for the design of lifting system of pumping unit.

关 键 词：抽油机井 有杆泵 举升系统 机理仿真 生产参数 优化设计 

分 类 号：TE933.1[石油与天然气工程—石油机械设备]