Abstract— In this paper, subspace identification methods are compared regarding their capability to cope with process disturbances occurring in complex plants. The Tennessee Eastman process is considered to be a realistic simulation model of chemical processes. The MOESP, N4SID and ORT methods as well as some of their variants are applied to data gathered from the process while being subject to...

A hierarchical structure based on a Deep LSTM Supervised Autoencoder Neural Network (Deep LSTM-SAE NN) is presented for the detection and classification of faults in industrial plants. The proposed methodology has ability to classify incipient that are difficult detect diagnose with traditional many recent methods. Faults grouped into different subsets according degree difficulty them accuratel...