ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

A New Insight into the Risk of Gaseous Release Assessment Based on a Minimax Approach

C. I. Anghel and A. Ozunu

Department of Chemical Engineering, Faculty of Chemistry and Chemical Engineering, University “Babes-Bolyai”, RO-3400 Cluj-Napoca

E-mail: canghel@chem.ubbcluj.ro

Abstract: An alternative method of the risk assessment for pollutant gas release based on artificial intelligence approach is presented. In principle the method is a pattern recognition developed as a linear discriminative classification problem related as a Minimax approach. This procedure has the ability to estimate both the dispersion of pollutant emission and the probabilistic certainty of these values. A simplified numerical experiment, prediction of the sulphur dioxide concentration from a smokestack, was carried out either using the proposed algorithm implemented in MATLAB package, or a specialized software package SLAB View. Both analyses produced reasonably accurate results. However, the proposed method was able to estimate both dispersion of the pollutant emission and a probabilistic certainty of this value.

Full paper in Portable Document Format: 596ba469.pdf

Chemical Papers 59 (6b) 469–475 (2005)