Predicting Gamma Ray Linear Attenuation Coefficient for Different Nano-Concrete Types Using Artificial Intelligence

Fathy, Islam N.; El-Sayed, Alaa A.; Sufe, Waleed H.

doi:10.21608/fuje.2021.205144

Predicting Gamma Ray Linear Attenuation Coefficient for Different Nano-Concrete Types Using Artificial Intelligence

Document Type : Original Article

Authors

¹ Civil Engineering Department, October High Institute For Engineering and Technology

² Civil Engineering Department, Faculty of Engineering, Fayoum University

³ Housing and Building National Research Center, Egypt

10.21608/fuje.2021.205144

Abstract

Fire in buildings is nearly always man- made, i.e. resulting from negligence or error, which can cause immense damage in terms of lives and property [1]. But when we deal with nuclear constructions (like nuclear power plants NPP), the dangers of fire do not stop only at the potential damage that the concrete structure is exposed to, but rather extends to the risk of a radiation leak that may cause serious damage to the human life and all living creatures. For this reason, designers of nuclear constructions (which are mostly reinforced concrete) give special attention for making the concrete structure capable of resisting the effects of fire or thermal leakage, as well as having a high ability to resist all types of radiation (specially gamma ray radiation). On the other hand, incorporation of nano additives into concrete structures components become a promising field of research these days. The current study tries to investigate the effect of using different nano materials (Nano silica, Nanoclay, and hybrid mix of both materials) as a cement replacement into the concrete radiation resistance ability (in the term of linear attenuation coefficient μ). Results showed remarkable enhancement on the values of μ at all temperature degrees. For the conduct of reliable estimate and prediction of the values μ, this study adopts the fuzzy logic models as powerful tools of artificial intelligence to model the non-linear cause and effect relationships. Prediction results was superior when compared with traditional linear regression analysis.

Keywords

Main Subjects