Document Type : Research paper
Authors
1 Department of Radiation Application, Shahid Beheshti University, Tehran, Iran
2 Department of Radiation Application, Shahid Beheshti University G.C., Tehran, Iran
3 Faculty of Engineering, Shahed University, Tehran, Iran
Abstract
Concrete is commonly used to shield gamma rays and neutrons. The effectiveness of concrete shielding for neutrons depends on the moisture content in the concrete. Moreover, the strength and durability of concrete structures are influenced by the moisture content in the concrete specimen. Therefore, determination of the moisture content in concrete is very important. The gamma ray attenuation technique is a potentially attractive non-destructive method for determining the concrete moisture content due to its high accuracy and speed. In this study, gamma attenuation in concrete shields of different thicknesses and moisture levels was simulated using the Monte Carlo method. Two separate artificial neural networks (ANN) were trained with simulation data to accurately estimate results and decrease calculation time. The thickness of concrete is predicted in the first ANN. Then, the count in full energy peak and thickness is applied to the second ANN to determine the concrete moisture. The trained neural network can estimate the thickness of concrete with a main relative error (MRE) of 0.42%. The findings indicate that the suggested approach can accurately determine the concrete moisture with an MRE error of 4.6% for test data.
Keywords
Main Subjects
)