基于支持向量机和聚类分析理论的钻具失效分析方法

Failure analysis of drill stem based on support vector machine and cluster analysis theory

针对钻具失效影响因素比较复杂的问题,建立了基于支持向量机和聚类分析理论的钻具失效原因分析模型。应用该模型可收集钻具失效样本数据,并对数据进行机器学习及优化,得到最佳的内部结构,从而可以计算和分析出钻具失效的原因。对大庆油田某区块的钻具失效数据进行了处理和分析,确定了该区块钻具发生失效的根本原因,并且有针对性地提出了6点预防钻具失效措施。通过计算证明,这种新模型的计算和分析结果真实、可靠。

Considering the complex influence factors for the failure of drill stem,an analytic model for failure of drill stem based on support vector machine and cluster analysis theory was established. This model can be used to collect the data of trouble drill stem samples. Then,the data of samples were optimized,and the best inner structure of the data was obtained. Furthermore,the real failure reasons of drill stem could be obtained with the model. The failure data of drill stem collected from a district of Daqing Oilfield were pretreated and analyzed,according to the requirements of the model. Good results have been obtained successfully. Six measures for preventing failure of drill stem were presented. The new model is more accurate and reliable to calculate and analyze the failure of drill stem in drilling engineering.