【摘要】 针对传统单模态故障检测方法对多模态工业过程故障检测准确率和效率低的问题,提出将变分模态分解(VMD),独立主元分析(ICA)和核主成分分析KPCA相结合的联合故障检测方法 VMD-IKPCA应用于多模态故障检测。首先,在对样本数据进行模态聚类之后,应用VMD对多模态过程数据进行滤波、降噪处理,通过ICA对处理过后的数据进行主元提取并应用KPCA对提取的主元变量进行故障检测。该方法的有效性通过多模态TE过程的故障检测进行验证,并与传统KPCA方法进行比较。实验结果表明,VMD-IKPCA对多模态过程故障检测有效性好,准确率高。更多还原

【Abstract】 In order to solve the low accuracy and efficiency problem of traditional single mode fault detection methods for multimode processes,a novel IKPCA method is proposed based on variational mode decomposition( VMD). Firstly,after the modal clustering of the sample data and for the sake of filtering the data better, the VMD algorithm is introduced.Then,with the application of independent component analysis( ICA), it can extract the principal components from the processed data. Finally, the extracted main variables are sent to KPCA for fault diagnosis. The feasibility and validity of the proposed strategies are demonstrated through the Tennessee Eastman( TE) process. Compared with the conventional KPCA,simulation results show that VMD-IKPCA is an effective and accurate method in multimode process fault detection.更多还原