文摘
The wavelet theory and multiscale method has generated an interest for fault monitoring andcontrol in petrochemical processes. Principal component analysis (PCA) has been used successfully as a multivariate statistical process tool for detecting faults by extracting feature informationfrom complex petrochemical data. The traditional linear PCA (LPCA) is restricted to complicatednonlinear systems; therefore, an adaptive nonlinear PCA (NLPCA) that is based on an improvedinput training neural network (IT-NN) is presented. A momentum factor and adaptive learningrates are added into the learning algorithm, to improve the training speed of the IT-NN. Anovel method of wavelet-based adaptive multiscale nonlinear PCA (MS-NLPCA) is proposedfor process signal monitoring. It can effectively monitor the slow and feeble changes of faultsignals that cannot be monitored by conventional PCA, and yet detect early faults to yield aminimum rate of false alarms. The validity of the proposed approach has been proved byexperimental simulations and practical application.