【摘要】 作为统计学的一个分支,聚类分析己经被广泛研究了多年,并形成了系统的方法体系,利用基于聚类分析方法的数据挖掘在实践中己取得了较好效果。IMS算法是N。ASA根据智能监控系统以及故障诊断技术的发展而提出的一种利用正常数据库来监控异常数据的可实时监控算法,已经被NASA使用在各个方面,收获了令人满意的效果。在分析美国航天飞机主发动机(SSME)IMS算法的基础上,提出一种基于液体火箭发动机稳态运行故障数据的夹角余弦IMS聚类分析研究方法,并对其进行仿真验证。更多还原

【Abstract】 As a branch of statistics,clustering analysis has been widely studied for many years,and formed a system of the systematic method.The data mining based on clustering analysis method has achieved a good effect in practice.IMS algorithm presented by NASA according to the development of the intelligent monitoring system and fault diagnosis technology is a real-time monitoring algorithm to use normal database to monitor the abnormal data,which has been used by NASA in many aspects and gained a satisfactory effect.The included angle cosine IMS clustering analysis method based on steady process fault data of LRE is proposed in this article on the basis of analysis of the IMS algorithm of SSME.It was verified by simulation analysis.更多还原