【摘要】 针对基于动态主元分析的故障检测方法存在的主元个数较多以及计算效率低等问题,本文提出基于混合动态主元分析(Hybrid Dynamic Principal Component Analysis,HDP-CA)的复杂过程故障检测方法。该方法采用分步策略消除数据之间的自相关和互相关性,提高了故障检测的精度和效率。对TE过程典型故障和热连轧过程中断带故障检测结果表明:HDPCA方法提取的主元个数少于DPCA方法提取的主元个数。并且,基于HDPCA的T2和SPE统计量的检测性能和检测精度都由于基于DPCA的统计量。因此,本文提出的方法可以准确有效地检测出故障。更多还原

【Abstract】 In dynamic principal component analysis(DPCA)for fault detection,there are some drawbacks such as an excess of the number of principal components(PCs),low computational efficiency and etc.For dealing with the problem,this paper presents a hybrid dynamic principal component analysis(HDPCA)method for fault detection.This method can remove cross-correlation and serial correlation by divide-and-conquer algorithm instead of parallel processing strategy,which can detect individual fault accurately and efficiently.The fault 4 of Tennessee Eastman process(TE)and the strip breaking fault in steel rolling process are used to demonstrate the presented performance of the presented method in comparison with DPCA-method fault detection.The simulation shows that:the extracted number of Principal components using HDPCA algorithm is fewer than DPCA.Moreover,the presented method can detect fault more accurately and effectively than DPCA algorithm by T2 and SPE statistic.It can be perceived that the presented method has the better performance for fault detection and computational efficiency.更多还原