Filters
Results 1 - 1 of 1
Results 1 - 1 of 1.
Search took: 0.017 seconds
[en] In this paper is presented an Artificial Neural Network (Ann) that has been designed, trained and validated to determinate the effective dose e, ambient dose equivalent h(10) and personal dose equivalent hp(10,θ) fluence-to-dose conversion coefficients at different positions, having as only input data 7 count rates obtained with a Bonner Sphere Spectrometer (Bss) system. A set of 211 neutron spectra and the fluence-to-dose conversion coefficients published by the International Atomic Energy Agency were used to train and validate the Ann. This set was divided into 2 subsets, one of 181 elements to train the Ann and the remaining 30 to validate it. The Ann was trained using Bss count rates as input data and the fluence-to-dose conversion coefficients as output data. The network was validated and tested with the set of 30 elements that were not used during the training process. Good results were obtained proving that Ann are a good choice for calculating the fluence-to-dose conversion coefficients having as only data the count rates obtained with a Bss. (Author)
Source
Oct 2012; 5 p; Sociedad Mexicana de Irradiacion y Dosimetria; Mexico D. F. (Mexico); 13. International Symposium; 13. Conferencia Internacional; Ocoyoacac, Estado de Mexico (Mexico); 15-19 Oct 2012; 23. National Congress on Solid State Dosimetry; 23. Congreso Nacional sobre Dosimetria de Estado Solido; Ocoyoacac, Estado de Mexico (Mexico); 15-19 Oct 2012; ISBN 978-607-00-6167-7; 