Isotopic correlation for 242PU composition prediction: Multivariate regresssion approach

Multivariate regression calibration using multiple linear regression (MLR), principle component regression (PCR) and partial least squares regression (PLSR) algorithm was performed on Pu-238, Pu-239, Pu-240 and Pu-241 atom% abundances to predict (PU)-P-242 isotopic abundance. The MLR algorithm was found to be the best among these three algorithms. The effect of (PU)-P-238 composition on the (PU)-P-242 abundance prediction was found to be small but significant especially for achieving high accuracy of < 0.5%. PCR and PLSR generated nearly identical results and were inferior to the MLR results. A comparison of MLR results with those obtained by employing seven previously reported empirical methods revealed far superior prediction capability of MLR model. Among the seven empirical models, the best prediction capability was found for Bignan correlation containing (PU)-P-238 isotopic data. The study clearly demonstrates that the production of Pu-238 and Pu-242 has some small correlation and the use of Pu-238 in isotopic correlation for Pu-242 prediction is important to get accurate results. (C) 2014 Elsevier Ltd. All rights reserved.