Change point and fault detection using Kantorovich Distance. (August 2019)

• Record Type: Journal Article
• Authors: Kammammettu, Sanjula; Li, Zukui
• Is Part Of: Journal of process control. Volume 80(2019)
• Page Start: 41
• Page End: 59
• Publication Date: 2019-08
• Subjects: Fault detection -- Kantorovich Distance -- Principal component analysis -- Change Point detection -- Linear programming; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2019.05.012
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: Proposed a novel algorithm for fault detection using PCA modelling with the Kantorovich Distance metric. Evaluates the effect of sampling distribution of noise on the fault detection performance of the KD metric. Proposed method for online monitoring has been evaluated on various case studies through simulation. Proposed method demonstrates superior performance than conventional PCA metric based method. Abstract: The automation of real-time process monitoring in the industry is an ongoing challenge. Chief among the objectives of monitoring are change point detection and the detection of faults in process variables and sensor measurements. In this paper, we propose a novel algorithm for change point and fault detection using Kantorovich Distance (KD), a metric induced from optimal mass transport theory. To evaluate the performance of the proposed method, we first evaluate the change point detection capability of the KD metric for data sampled from various probability distributions. Next, the fault detection performance of the KD metric is evaluated for three cases of faults – sustained bias, drift, and multiple intermittent biases – and contrasted against that of the traditional PCA-based metrics, Q and T 2 statistics. The algorithm is tested on several case studies including a synthetic data, a simulated continuous stirred tank heater system and the benchmark Tennessee Eastman process. The results obtained showcase the superiority of the proposed algorithm over the conventional scheme.