Online Fault Diagnosis of Chemical Processes Based on Attention-Enhanced Encoder–Decoder Network

Abstract

The data of chemical processes often contain dynamic timing characteristics, and traditional fault detection has low usage of dynamic information, which limits the fault diagnosis performance. To address this problem, this paper proposes a new chemical process fault diagnosis method based on an attention-enhanced encoder–decoder network model (AEDN). The long short-term memory (LSTM)-based encoding part is used to extract the feature information of the process data and combine it with the attention mechanism to utilize the dynamic information among the process data more effectively, the decoding part uses the LSTM and combines the context vector provided by the attention mechanism to provide more accurate state information for the softmax regression, and finally, the softmax regression is used to obtain the probability value of the fault category for each sample data. The attention mechanism improves the model’s efficiency in processing dynamic information in the time domain. The proposed method is experimented using Tennessee Eastman (TE) process data and compared with the standard vanilla long short-term memory with batch normalization (LSTM-BN) results. The results show that the proposed method is more effective in diagnosing faults.