基于MPCA与DTW的间歇反应过程的性能监视及故障诊断  

Monitoring and Fault Diagnosis of Batch Process Based on MPCA and DTW

作　　者：李元[1] 王纲[1] 曹锐[2] 

机构地区：[1]沈阳化工学院,辽宁沈阳110142 [2]北京化工大学,北京100029

出　　处：《沈阳化工学院学报》2003年第4期285-289,共5页Journal of Shenyang Institute of Chemical Technolgy

摘　　要：　研究了基于多方向主元分析(MPCA)理论和动态时间错位(DTW)理论对间歇反应过程进行性能监视及故障诊断的方法.多元统计分析理论是根据反应过程的历史批量数据,建立过程的统计数据库、统计数学模型;在此基础上确定过程的统计监控指标,并向各自的统计模型投影,判断其与模型的拟合程度来诊断是否有故障发生.由于间歇反应的各批次物理条件的不同、约束条件的不同,导致批次与批次之间的数据轨迹不同步.动态时间错位(DTW)理论通过搜寻两轨迹的相似特征,适时地对轨迹进行压缩和扩张,从而使其达到一致的程度.这为基于MPCA理论进行故障诊断提供了合理、可靠的数据依据.比较轨迹进行同步化处理前、后的故障诊断结果完全不同,因此,DTW理论的应用在基于MPCA理论进行性能监视及故障诊断中具有重要意义.This paper presents monitoring and fault diagnosis of batch process based on MPCA and DTW. Multi-way principle component analysis (MPCA) sets up statistical database and statistical model according to process historical data; Based on them, we can calculate statistical factors and map them onto respective model to diagnose fault by judging if they consist with models. Due to the presence of batch-batch disturbances and the existence of physical constraints, batch processes often are characterized by unsynchronized trajectories. Dynamic time warping (DTW) reconciles two trajectories by searching same parts between two trajectories and contracting or expanding them. This method can provide reasonable and reliable data for fault diagnosis of batch process based on MPCA. Fault diagnosis of synchronized trajectories is different with unsynchronized. So it is very important to apply DTW to the monitoring and fault diagnosis of batch process based on MPCA.

关 键 词：多方向主元分析(MPCA) 动态时间错位(DTW) 故障诊断 同步化 

分 类 号：O235[理学—运筹学与控制论]