Tensor sequence component analysis for fault detection in dynamic process

Abstract

For dynamic processes, using sequence information to augment the data can improve fault detection performance. Traditional approaches transform raw data into augmented vectors, which leads to losses in structural information in the variables and increases the data dimension. This paper proposes a novel data dimension reduction algorithm called tensor sequence component analysis (TSCA) and applies it to dynamic process fault detection. The algorithm extends each sample into a matrix comprising current and past process data, and simultaneously reduces the dimensions of time delay and the variables for feature extraction, solving the problem of the curse of dimensionality. For the dimension reduction of time delay, in order to extract similar information from the samples, each sample is reconstructed with time neighbourhoods. For the dimension reduction of the variables, considering the information of different variables variance information of the latent variables is maximized for feature extraction. Finally, a numerical example and the Tennessee Eastman process are used to demonstrate the efficacy of the proposed method.

A novel imbalanced fault diagnosis method based on area identification conditional generative adversarial networks

“…Based on the actual chemical manufacturing process, the TE process simulation system was created, and it is ideal for building control schemes and validating fault diagnosis models. [22][23][24] The TE process monitors 11 operating variables and 41 measurement variables, and samples every 3 min. In addition, 21 types of fault samples can be obtained.…”