摘要
大型复杂工业系统具有众多过程变量、庞大信息数据、强耦合的特点。鉴于目前针对复杂系统采回的大量数据信息尚无有效的分析手段,提出了一种基于主成分分析的过程数据降维方法,该方法能有效的消除冗余信息,从而揭示数据的主要结构,然后对这种方法在水泥回转窑系统中进行了仿真研究,对系统繁多的过程变量进行降维筛选和分析利用。仿真结果证实了该方法的有效性及实用性。
Large-scaled complex industrial systems possess the features as lots of process variables, massive information data and strong couples. To the situation of no efficient analysis method to massive information data of complex system , a method of reducing process data dimensions based on principal component analysis (PCA) is proposed. This method can eliminate redundant information effectively, and reveal the main structure of original data. Simulation test for rotary cement kiln system was performed. The dimensions of various process variable are reduced, then it's easier to select and analysis key variables. Results on PCA method verify that it is effective and practicable.
出处
《硅酸盐学报》
EI
CAS
CSCD
北大核心
2002年第z1期131-134,共4页
Journal of The Chinese Ceramic Society
基金
浙江省自然科学基金资助项目(No.601112)。
关键词
主成分分析
水泥回转窑
因子载荷
最小二乘估计
principal component analysis
rotary cement kiln
factor loading
least squares estimation
