Prediction of peak-to-background ratio in gamma-ray spectrometry using simplex optimized artificial neural network

An artificial neural network (ANN) model was used for the prediction of peak-to-background ratio (PBR) as a function of measurement time in gamma-ray spectrometry. In order to make the ANN model with good predictive power, the ANN parameters were optimized simultaneously employing a variable-size simplex method. Most of the predicted and the experimental PBR values for eight radionuclides (226Ra, 238U, 235U, 40K, 232Th, 134Cs, 137Cs, and 7Be) commonly detected in soil samples agreed to within ±19.4% of the expanded uncertainty and 2.61% of average bias.