Application of Unsupervised Chemometric Analysis and Self-organizing Feature Map (SOFM) for the Classification of Lighter Fuels

摘要

A variety of lighter fuel samples from different manufacturers (both unevaporated and evaporated) were analyzed using conventional gas chromatography-mass spectrometry (GC-MS) analysis. In total 51 characteristic peaks were selected as variables and subjected to data preprocessing prior to subsequent analysis using unsupervised chemometric analysis (PCA and HCA) and a SOFM artificial neural network. The results obtained revealed that SOFM acted as a powerful means of evaluating and linking degraded ignitable liquid sample data to their parent unevaporated liquids.