Use of Principal Components in Parallel Coordinates for Process Monitoring

Mark J. Nixon, Ricardo Dunia, Willy K. Wojsznis, and Thomas Edgar

Keywords

Process Monitoring, Parallel Coordinates, Principal Component Analysis, Fault Detection

Abstract

This work demonstrates that parallel coordinates is a recognized visualization technique suitable for process monitoring applications in industrial facilities, where a significant number of sensors are used to detect and identify abnormal operating conditions. Principal Component monitoring methods implemented in Parallel Coordinates, named PC², permits to compare faulty events and determine the frequency of false alarms. The monitoring visualization technology presented by PC² was successfully applied to column flooding using actual process data.

Important Links:

• DOI: 10.2316/P.2013.794-006
• From Proceeding (793) Artificial Intelligence and Applications / 794: Modelling, Identification and Control / 795: Parallel and Distributed Computing and Networks / 796: Software Engineering / 792: Web-based Education - 2013

Go Back