时序关联规则在钻井事故中的应用  

Application of temporal association rules in drilling accident

作　　者：肖斌[1] 肖亚飞[1] 

机构地区：[1]西南石油大学计算机科学学院,成都610500

出　　处：《计算机应用》2017年第A01期308-311,共4页journal of Computer Applications

摘　　要：针对关联规则用于分析石油钻井事故的因果关系未考虑时间属性的问题,提出了一个分析石油钻井事故属性的时序关联规则方法。该方法是基于粗糙集、时间序列、关联规则的划分算法。首先,在粗糙集的基础上过滤数据库中冗余属性来获得核属性;然后,设定连续时间长度和子时间长度,连续时间长度被子时间长度划分为若干个等时间间隔;最后,基于改进Apriori算法在满足支持度和置信度的前提下生成时序关联规则。实验结果表明,时序关联规则算法可以准确预测钻井事故发生的趋势。Association rules are used to analyze causality of oil drilling accidents without considering time attributes. To resolve the issue, a temporal association rule method was proposed. This method is based on rough sets, time series and association rules. First of all, redundant attributes in the database were filtered on the basis of rough set to obtain core attributes; then, the continuous time length and time sublength were set, and continuous time length was divided into some equal time intervals; finally, the improved Apriori algorithm in the premise of support degree and confidence was used to generate temporal association rules. The experimental results show that the temporal association rules can accurately predict the trend of drilling accidents.

关 键 词：石油钻井 关联规则 粗糙集 时序关联规则 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]