BDPCA在线过程监测方法  

A Method of On-line Process Monitoring Based on BDPCA

作　　者：肖应旺[1] 姚美银[1] 

机构地区：[1]华南师范大学南海校区计算机工程系,广东佛山528225

出　　处：《控制工程》2009年第2期133-137,147,共6页Control Engineering of China

基　　金：科技部"863"基金资助项目(2003AA241160)

摘　　要：针对基于多向主元分析(Multiway Principal Component Analysis,MPCA)的方法在批过程故障监测中以样本观测相互独立作为假设前提条件,没有考虑到时间序列相关性的影响及需要对新批次未反应完的数据进行预估的缺陷,提出一种批过程动态主元分析(Batch Dynamic PCA,BDPCA)在线监测方法。该方法采用时滞变量将过程的静态和动态特征相结合,有效地去除了测量变量时间序列的自相关关系,并通过时滞窗口提供了在线监测方案,避免了对新批次未反应完的数据进行预估的需要,提出确定时滞变量的算法。将BDPCA应用于β-甘露聚糖酶发酵批过程的仿真监测,与移动窗多向主元分析(Moving Window MPCA,MWMPCA)法相比,仿真结果表明该方法能够更精确地对过程故障行为进行描述,具有良好的准确性和实时性。To the limitations of batch process monitoring approach based on muhiway principal component analysis （ MPCA ） an batch dynamic PCA （BDPCA） on-line monitoring approach is proposed. BDPCA incorporates both static and dynamic process characteristic by using the time-lagged variable. So the auto-correlation in time series of the process variables is effectively removed. The on-line monitoring technique is provided by the time-lagged window, and the estimation of the future value of each new batch is no longer required. The algorithm of the time-lagged variable is proposed. BDPCA is applied to β-mannanase fermentation batch process simulation monitoring. The simulation results show that, compared with moving window MPCA, the BDPCA can describe the process fault behavior more precisely, and has better accuracy and real-time characteristic.

关 键 词：批过程动态主元分析 时滞变量 在线监测 β-甘露聚糖酶发酵批过程 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]