(371e) Nonlinear and Non-Gaussian Process Monitoring Based on Vine Copula and Robust Auto-Encoder Model

Recently, modern industrial processes are pushed towards complicate, automatic and digital direction. However, process safety is still the core target of process control systems, and process monitoring is the key part of process control. Meanwhile, with the development of distributed control system and sensor network, the process control system collects huge number of process data. Therefore, data driven process monitoring has become a popular research subject in both academic research and industrial applications. Due to the conventional Vine Copula-Based Dependence Description (VCDD) model is not robustness and cannot handle the big data, this paper proposes a VCDD model based on Robust Auto-encoder (RAE) named RAE-VCDD method. The target of RAE-VCDD model is making the model output robust to change in input during training procedure and improve the sensitivity of model reconstruction errors to the faults. Moreover, theoretical analysis shows that the RAE-VCDD model has better performance in fault detection area. Because of the combination of RAE, the proposed method can deal with big data very well. The effectiveness and benefits of the proposed model are illustrated with a numerical examples and the Tennessee Eastman benchmark process for fault detection. The results show that the proposed method can achieve good performance in chemical process fault monitoring.