基于演化模型偏好多目标优化的智能采油辅助决策支持

发布时间：2018-01-05 06:28

本文关键词：基于演化模型偏好多目标优化的智能采油辅助决策支持　出处：《机械工程学报》2017年13期 　论文类型：期刊论文

【摘要】：通过智能采油系统自主分析与决策获取油田机采过程最佳决策参数,对解决机采系统效率低、能耗大等问题具有重要意义。受机械、地层、人为等不确定因素影响,智能采油系统难以构建生产参数、环境变量与系统性能、设定生产方式之间的机理关系并优化决策。为此,提出基于动态演化建模的偏好多目标优化方法,以实现采油系统的自主决策。利用无迹卡尔曼滤波神经网络(Unscented Kalman filter neural network,UKFNN)挖掘机采系统潜在规律,建立其动态模型;构建产液量偏好多目标优化目标函数,并利用非改进支配排序遗传算法(Non-dominated sorting genetic algorithm 2,NSGA2)获取相应的最佳决策参数。某油田试验结果表明:该方法使得系统日耗电量降低15.87%,系统效率提高4.9%。可见,所提方法可行且有效。
[Abstract]:Through independent analysis and decision for oil recovery process machine optimal decision parameter intelligent production system, to solve the pumping system efficiency is low, energy consumption and other issues is of great significance. The formation, mechanical, man-made and other uncertain factors affecting the production parameters of oil production system to build intelligent, environment variables and system performance, the relationship between setting mechanism the mode of production and optimization decision. Therefore, put forward a method of dynamic multi objective optimization evolutionary modeling based on preference, in order to realize the autonomous decision of oil system. By using the unscented filter neural network (Unscented Kalman Calman filter neural network, UKFNN) excavator production system latent rule, establishes the dynamic model; constructing preference multi-objective optimization objective function. The fluid volume, and use the improved non dominated sorting genetic algorithm (Non-dominated sorting genetic algorithm 2, NSGA2) the best decision to obtain the parameters accordingly. The experimental results of an oil field show that the method makes the daily power consumption of the system lower by 15.87%, the system efficiency is improved by 4.9%., and the proposed method is feasible and effective.

【作者单位】：重庆科技学院电气与信息工程学院;四川理工学院人工智能重点实验室;西安石油大学电子工程学院;新疆华隆油田科技股份有限公司;
【基金】：国家自然科学基金(51375520,51404051) 重庆市基础与前沿研究计划重点(cstc2015jcyj BX0089) 重庆市教委科学技术研究(KJ1401309,KJ1501304) 重庆科技学院校内科研基金(CK2016Z16)资助项目
【分类号】：TE938;TP18
【正文快照】： 0前言1智能制造装备是先进制造技术、信息技术和智能技术的集成和深度融合。它是高端装备制造业的重点发展方向、信息化与工业化深度融合的重要体现,也是有效提升生产效益、技术水平和产品质量,降低能源资源消耗重要手段[1-2]。油田主要采油系统,即抽油机采油系统面临效率低下

【相似文献】