基于小波分析和主元分析的连续工业过程的故障诊断-控制理论与控制工程专业论文.docxVIP

北京化工大学学位论文用纸基于小波分析和主元分析的连续工业过程的故障诊断 北京化工大学学位论文用纸 基于小波分析和主元分析的连续工业过程的故障诊断 摘要 随着现代化大生产的发展和科学技术的进步，现代工业系统的结 构越来越复杂，功能越来越完善，自动化程度越来越高。由于许多无 法避免的因素的存在，工业系统会出现各种各样的故障，这些故障引 起的直接后果，有些是降低或失去系统某些预定的功能，有些则造成 严重的甚至灾难性的事故。因此，大力开展过程故障诊断的研究，具 有重要的理论意义和实践价值。 本文针对国内外工业过程故障诊断技术的理论与工业应用的现 状，分析了故障诊断的过程与实质，研究了几种常用的故障诊断方法， 对其中较为重要的小波分析、自联想神经网络、主元分析等关键技术 进行了详细深入的探讨。针对小波分析和主元分析的优点，提出了一 种小波分析与主元分析相结合的连续过程故障诊断方案，并将此方案 应用于某乙烯厂的裂解炉故障诊断系统，取得了满意的诊断效果，证 实了该方案的有效性和可行性。最后指出了本课题中有待于进一步研 究解决的问题以及故障诊断领域的发展趋势。 关键词：小波分析，神经网络，主元分析，故障诊断 豁嗣煮钼全文公布 北京化工大学学位论文用纸THE 北京化工大学学位论文用纸 THE FAULrr DIAGNOSIS 0F CONTINUoUS INDUSTRIAL PRoCESS BASED oNⅥ(AvELET THEoRY AND PRINCIPAL COMPoNENT ANALYSIS ABSTRAT With the development of modernized wholesale manufacture and the progress of science and technology,the structures of continuous industrial process are more complex，the functions are more consummate and the automatic degree is higher．For the existence of inevitable factors，there are many faults in these industrial systems．The results of these faults may debase or invalidate the functions，or cause catastrophic accidents．So it has important academic and practical meanings to research industrial process fault diagnosis． According to the status pro of the theories and industrial applications of industrial process fault diagnosis technology at home and abroad，this paper studies the process and essential of fault diagnosis，analyzes some fault diagnosis technologies，and emphasizes on the methods of wavelet analysis，autoassociative neural network and principal component analysis， For the advantages of wavelet analysis and PCA，this paper proposes a II 北京化工大学学位论文用纸continuous 北京化工大学学位论文用纸 continuous industrial process fault diagnosis solution using these two methods．The solution has been applied to the fault diagnosis system of ethylene pyrolysis furnace．The validity and effectivity of the solution is proved and the fault diagnosis results are satisfied．At last，this paper points out the problems to be further researched and the