Amalgamation of anomaly-detection indices for enhanced process monitoring

Abstract

Accurate and effective anomaly detection and diagnosis of modern industrial systems are crucial for ensuring reliability and safety and for maintaining desired product quality. Anomaly detection based on principal component analysis (PCA) has been studied intensively and largely applied to multivariate processes with highly cross-correlated process variables; howver conventional PCA-based methods often fail to detect small or moderate anomalies. In this paper, the proposed approach integrates two popular process-monitoring detection tools, the conventional PCA-based monitoring indices Hotelling's T 2 and Q and the exponentially weighted moving average (EWMA). We develop two EWMA tools based on the Q and T 2 statistics, T 2-EWMA and Q-EWMA, to detect anomalies in the process mean. The performances of the proposed methods were compared with that of conventional PCA-based anomaly-detection methods by applying each method to two examples: a synthetic data set and experimental data collected from a flow heating system. The results clearly show the benefits and effectiveness of the proposed methods over conventional PCA-based methods.

Anomaly Detection Based on Zone Partition for Security Protection of Industrial Cyber-Physical Systems

“…Hu X. et al [21] employed the hotelling's T 2 statistic to handle multivariate anomaly detection problem in control systems. Harrou F. et al [22] integrated Principle Component Analysis (PCA) and EWMA to develop two process-monitoring detecting tools, T 2 -EWMA and Q-EWMA, which exhibited an effective approach to balance the false negative rate and false positive rate. Similarly, system state equation was regarded as another useful method.…”

Improved nonlinear fault detection strategy based on the Hellinger distance metric: Plug flow reactor monitoring

“…In these cases, data-based monitoring techniques are more commonly used [6]. Moreover, data-based techniques provide efficient tools for extracting useful feature for design of monitoring schemes based on empirical models derived from the available process data [6,7,8,9]. Such methods require a minimal a prior knowledge about process physics, but depends on the availability of quality input data.…”

A measurement-based fault detection approach applied to monitor robots swarm

“…By extracting useful data from the original dataset using PCA modeling and then using monitoring indices faults in the monitored swarm robotics can be detected. Unfortunately, conventional PCAbased monitoring indices such as T 2 and Q charts are less efficient in detecting incipient changes in the mean of process data [14]- [16].…”