基于迭代主成分分析的过程监测方法的研究与实现

Research and implementation of process monitoring based on iterative principal component analysis(PCA)

利用迭代主成分分析 (PCA)算法 ,提出一种在线过程监测方法。根据实际生产过程经验 ,提供了由多元统计控制图判断过程是否正常的准则 ,实现了实时在线的 PCA建模和过程监测。

A new method based on iterative PCA is presented to realize on line modeling and process monitoring. According to the operation experiences in real processes, some rules are proposed by which operator can judge whether the process is under control or not when monitoring the multivariate statistical control charts. The simulation shows the validity and feasibility of this method.