Evaluation of the suitability of neural network method for prediction of uranium activity ratio in environmental alpha spectra

• The suitability of RBF network method for the prediction of 234U/238U activity ratio was evaluated.
• The average of 2 consecutive channels of a partial uranium raw spectrum was plotted as the average curve.
• The points that their slopes are of the order of 0–1% per 10 channels, were used as inputs to the RBF.
• The network was trained by the simulated spectra library, could accurately predict the activity ratio.
• The network was trained by the real spectra library, could reasonably predict the activity ratio.

Applying Artificial Neural Network to an alpha spectrometry system is a good idea to discriminate the composition of environmental and non-environmental materials by the estimation of the 234U/238U activity ratio. Because it eliminates limitations of classical approaches by the extraction the desired information from the average of a partial uranium raw spectrum. The network was trained by an alpha spectrum library which was developed in this work. The results indicated that there was a small difference between the target values and the predictions. These results were acceptable, because the thickness of samples and the inferring elements were different in the real library.