Industrial process monitoring by synthetic view

Effective monitoring of computerized industrial processes, for performance analysis and improvement, needs friendly human-computer interaction. This may be obtained preferably in a low-dimensional space where the human operator should easily identify the main features of the actual process behavior through some characteristics of a synthetic view of the process. For this purpose space reduction must keep the relevant and informative geometric characteristics of the original space, using proper metrics. In this work the reduction of n-dimensional space to bi or tri-dimensional one is developed through multidimensional scaling with a proposed iterative capability. In the 2-D or 3-D process map, named POM-Projected Operational Map, the operational regions of the process under specific conditions can be classified. This classification is made by clustering (hierarchical, k-means, etc). For on-line implementation a recursive clustering technique is required and Dignet algorithm was adopted, giving information to the monitoring system about the possible quality of the running operating conditions. For the multidimensional scaling several metrics are comparatively applied. This strategy is applied to the process of Hydro Desulfuration (HDS) from Refinery of Galp Energia at Sines, Portugal.