基于距离异常点的机械比能评价储层物性研究  

Research on Mechanical Specific Energy Evaluation of Reservoir Physical Properties Based on Distance Anomaly Algorithm

作　　者：李龙[1] 管洁 LI Long;GUAN Jie(School of Computer and Information Technology,Northeast Petroleum University,Daqing 163318)

机构地区：[1]东北石油大学计算机与信息技术学院,大庆163318

出　　处：《计算机与数字工程》2023年第8期1924-1929,共6页Computer & Digital Engineering

摘　　要：在调研了大量国内外机械比能在石油钻探中的应用及储层物性评价模型的研究方法和实际应用的基础上,将结合塔东和松辽盆地的测录井数据以及相关物理实验,开展基于距离的异常点检测研究和应用,完善机械比能评价模型的研究方法,建立评价更加准确的机械比能评价储层物性模型,合理准确的评价该地区储层物性的好坏,为国家或石油公司开采石油的储层提前预测与评价的研究和行业发展趋势政策的选择提供决策性的依据。采用基于距离的异常点检测算法提取出了反映岩石储集层物性的特征参数(WOB,A_(b),T,RPM,ROP),通过确定标准化,和归一化后的特征参数,在实验中计算得到“物性指数”,来反映储层的物性,“物性指数”越大反映储层物性越差,相反,“物性指数”越小反应储集层的物性越好。最后实验结果表明,与普通算法相比,将基于距离的异常点检测算法应用到机械比能评价储层物性的模型中,储层物性评价结果准确率达到88.35%。On the basis of conducting extensive research on the application of mechanical energy ratio in oil drilling at home and abroad,as well as the research methods and practical applications of reservoir physical property evaluation models,the logging data from the Tadong and Songliao basins and relevant physical experiments are combined to carry out distance based anomaly detection research and application,the research methods of mechanical energy ratio evaluation models is improved,and more accurate mechanical energy ratio evaluation reservoir physical property models for evaluation are established.Reasonably and accurately evaluating the physical properties of the reservoirs in the region provides a decision-making basis for the research on early prediction and evaluation of reservoirs for national or oil companies to extract oil,as well as for the selection of industry development trends and policies.The distance based anomaly detection algorithm is used to extract feature parameters(WOB,A_(b),T,RPM,ROP)that reflect the physical properties of rock reservoirs.By determining the standardized and normalized feature parameters,the"physical property index"is calculated in the experiment to reflect the physical properties of the reservoir.The larger the"physical property index",the worse the physical properties of the reservoir.On the contrary,the smaller the"physical property index",the better the physical properties of the reservoir.The final experimental results show that compared with ordinary algorithms,applying the distance based anomaly detection algorithm to the mechanical specific energy evaluation model of reservoir physical properties achieves an accuracy of 88.35%in reservoir physical property evaluation results.

关 键 词：异常点检测算法 储层物性评价 物性指数 机械比能 

分 类 号：TE312[石油与天然气工程—油气田开发工程]