Fault detection in dynamic systems using the Kullback–Leibler divergence. (October 2015)

• Record Type: Journal Article
• Authors: Xie, Lei; Zeng, Jiusun; Kruger, Uwe; Wang, Xun; Geluk, Jaap
• Is Part Of: Control engineering practice. Volume 43(2015)
• Page Start: 39
• Page End: 48
• Publication Date: 2015-10
• Subjects: Kullback–Leibler divergence -- Auto-and cross correlation -- Subspace identification -- Non-Gaussian distribution -- Incipient fault conditions; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2015.05.010
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: This paper proposes detecting incipient fault conditions in complex dynamic systems using the Kullback–Leibler or KL divergence. Subspace identification is used to identify dynamic models and the KL divergence examines changes in probability density functions between a reference set and online data. Gaussian distributed process variables produce a simple form of the KL divergence. Non-Gaussian distributed process variables require the use of a density-ratio estimation to compute the KL divergence. Applications to recorded data from a gearbox and two distillation processes confirm the increased sensitivity of the proposed approach to detect incipient faults compared to the dynamic monitoring approach based on principal component analysis and the statistical local approach.