PROJECTED ORIENTATION MAPS BY SPACE REDUCTION FOR INTELLIGENT MONITORING OF INDUSTRIAL PROCESSES

Authors

Abstract

The simplified visualization of operational data in industrial plants might guarantee to the process operators take quick decisions for convenient supervision and monitoring of complex industrial systems. A space reduction technique is implemented to visualize and identify different work zones in the production of multiple grade products. The analysis of a historical database helps to create a bi-dimensional (2-D) representation of the operational regions of a Hydro Treating Process which works in 7 different modes. The space reduction technique keeps the relevant and informative geometric characteristics of the original space and builds a 2-D map, called Projected Operational Map (POM), where the operational regions of the process under a specific feed rate or load can be identified. In the POM can be detected some points where the system works away of its traditional work zone, indicating a probable abnormal operation of the process. A strategy to monitor dynamic systems based on POM is proposed in order to increase the productivity, quality and safety of the industrial plant.

Keywords

supervision and monitoring; space reduction; data mining; synthetic visualization; refinery

Related Project

CLASSE - Classificação Sintética para Supervisão Industrial (Synthetic Classification for Industrial Supervision)

Conference

IFAC-IEEE MCPL Management and Control of Production and Logistics, November 2004