Application of Multivariate Data Analysis Techniques for the Portable Isotopic Neutron Spectroscopy System

The Portable Isotopic Neutron Spectroscopy (PINS) is a commercialized system developed by Idaho National Laboratory to examine chemical compounds in munitions and containers non-destructively, utilizing the Prompt Gamma Neutron Activation Analysis (PGNAA) technique. The PINS system takes advantage of a germanium detector's high energy resolution, and gamma-ray peak analysis provides input to its chemical identification logic using a probabilistic decision tree. Multivariate analysis (MVA) techniques were contemplated with the expectation that they could supplement the current PINS algorithm. Principal Component Analysis (PCA) was selected to project gamma-ray spectra into the principal component domain. A PCA-based chemical identification algorithm was tested, and the results are presented in this study.