An Artificial Neural Network-Based Estimation of Bremsstarahlung Photon Flux Calculated by MCNPX

Abstract

Bremsstrahlung has an important place in the field of experimental physics, especially for description of photon-matter interaction and for characterization and analysis of materials. Bremsstrahlung photon is created by a high-energy electron, deflected in the electric field of atomic nucleus. Bremsstrahlung is also important for experimental studies, not only in the field of nuclear physics and particle physics but also in the fields of solid state physics, applied physics and astrophysics. In recent years, Monte Carlo simulation has become a widely used method for calculations related to bremsstrahlung. On the other hand, predictions by using artificial neural network can be performed with high accuracy. This study aims at observing variation in the photon flux as unction of target thickness and at processing output data by using an artificial neural network. We achieved a high degree of compatibility between two different methods. This study suggests that artificial neural network is a powerful tool for prediction of Bremsstrahlung and for other scientific problems.