Neutron spectrum unfolding using radial basis function neural networks.

Area of Science:

• Nuclear physics
• Computational methods
• Detector physics

Background:

• Neutron energy spectrum unfolding is crucial for applications like nuclear reactor monitoring and radiation shielding.
• Traditional methods include Bayesian theory, Monte Carlo simulations, and iterative approaches.
• Artificial neural networks (ANNs) offer advanced computational capabilities for complex data analysis.

Purpose of the Study:

• To compare the performance of different artificial neural network (ANN) architectures for neutron energy spectrum unfolding.
• To evaluate the effectiveness of Radial Basis Function (RBF) and Multilayer Perceptron (MLP) networks in this task.
• To identify the most suitable ANN method for neutron spectrum unfolding, with a focus on neutron dosimetry.

Main Methods:

• Utilized Radial Basis Function (RBF) and Multilayer Perceptron (MLP) artificial neural network (ANN) models.
• Trained and tested both ANN architectures using identical datasets derived from LiI(Eu) detector responses.
• Investigated the specific advantages of each ANN method in the context of neutron spectrum unfolding.

Main Results:

• Both RBF and MLP ANNs were successfully applied to neutron spectrum unfolding.
• The Radial Basis Function (RBF) neural network demonstrated particular effectiveness.
• The RBF network showed strong performance, especially when neutron dosimetry was the primary objective.

Conclusions:

• Radial Basis Function (RBF) neural networks are a viable and effective tool for neutron energy spectrum unfolding.
• ANNs, specifically RBF networks, offer a promising approach for accurate neutron dosimetry.
• The study highlights the potential of advanced computational techniques in nuclear data analysis.