基于故障特征时段识别的间歇过程故障诊断方法.docVIP

 n 基于故障特征时段识别的间歇过程故障诊 断方法# 赵珍1,2，王姝1,3，常玉清1,3，王福利1,3* 5 10 15 20 25 30 35 40 45 （1. 东北大学信息科学与工程学院 辽宁省沈阳市 110819； 2. 辽宁石油化工大学信息与控制工程学院 抚顺 113001； 3. 东北大学流程工业综合自动化教育部重点实验室 沈阳 110819） 摘要：针对间歇过程的故障诊断问题，本文提出了基于故障特征时段识别的 MFDA 故障诊 断方法。由于间歇过程的多时段操作特性，使得某一类型故障可能在一个或多个子操作时段 具有明显表征，而在其他时段没有故障表征，并且在各个子时段的故障表征程度也可能不一 致，即故障具有其相应的特征时段。故障的这种特征时段特性增加了故障诊断的复杂度。为 了准确地对缺乏先验知识的间歇过程进行故障诊断，首先必须自动识别出故障的特征时段。 由于过程数据的分布特性涵盖了系统运行状态的有效信息，据此通过对历史故障数据以及正 常数据质心的分析比较，识别历史故障的特征时段。继而，将识别后的各个时段综合在一起 划分出更细的“小时段”，利用多向 Fisher 判别分析(MFDA)方法分别建立对应的故障诊断 模型，从而将故障诊断的有哪些信誉好的足球投注网站空间深入到特定的特征时段，一方面避免了多模型 FDA 方法 模型过多的缺点，另一方面，避免了 MFDA 模型诊断精度不高、容易产生误判的缺点。仿 真实验验证了该方法的可行性和有效性。 关键词：间歇过程；多时段；故障特征时段；多向 Fisher 判别分析；故障诊断 中图分类号：TP277 MFDA Fault Diagnosis Method for Batch Process Based on Identification of Fault Feature Phases ZHAO Zhen1,2, WANG Shu1,3, CHANG Yuqing1,3, WANG Fuli1,3 (1. School of Information Science Engineering, Northeastern University, Shenyang 110819, China; 2. School of Information Control Engineering, Liaoning University of Petroleum Chemical Technology, Fushun, 113001; 3. Key Laboratory of Integrated Automation of Process Industry, Northeastern University, Ministry of Education, Shenyang 110819, China) Abstract: IThis paper proposes a MFDA fault diagnosis method for batch process based on identification of fault feature phases. Because of the multiplicity of operation phases in batch process, which have specific control objects, different dominant process variables and distinct process correlation characteristics, the faults may also have phase characteristic. It means that some type of fault may have obvious features in one or more specific operation phases, but do not show any features in other operation phases, at the mean time, the faulty features in different phases may also show different degrees. This phase characteristic of fault increases the complexity of fault diagnosis. To conduct fault diagnosis for batch process more precisely, it is needed to firstly identify fault feature phases. The dist