Prediction of radiation shielding properties for concrete by artificial neural networks

With discovering of radioactivity in the last century, the radiation started to be used in different fields, and due to its hazardous effect for human cell, radiation shielding became one of the most important topics for researcher. Besides using conventional materials such as concrete- and lead-based materials, improvement of their radiation shielding properties has also been popular. Theoretical calculation and setting up model to obtain linear attenuation coefficients to predict radiation shielding properties are the main ways for this purpose. In this study, the linear attenuation coefficients of concrete produced by adding ulexite in different rates have been predicted using the artificial neural network (ANN) model for 1 keV to 100 GeV photon energies. The main input for the ANN model was photon energy, density and ulexite rate in concrete, and the results were obtained by XCOM. The obtained ANN results were compared with the results obtained by XCOM calculations, and %99 linear correlations have been observed.