Advanced data-driven fault detection in gas-to-liquid plants

Fault detection is a critical part of process monitoring, where the objective is to flag unexpected operating behavior quickly and accurately. In this paper, a novel extension of the Generalized Likelihood Ratio charts is proposed, denoted as the Maximum Multivariate GLR charts. Linear and nonlinear data-driven models, namely principal component analysis and its kernel extension and neural networks , are combined with different statistical charts towards the detection of multiple fault types in three distinct case studies: synthetic, Tennessee Eastman process, and Gas-to-Liquid process. The results show that the MMGLR charts have a better detection accuracy than conventional charts, and that neural networks are more robust modeling techniques than PCA and KPCA for the sake of fault detection. • Robust fault detection is crucial for operational safety and maintenance. • A maximum multivariate Generalized Likelihood Ratio chart (MMGLR) is proposed. • Interval data analysis is used to improve its sensitivity to different fault types. • MMGLR is validated through both simulated and real chemical process data.