http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Bayram Bilmez,Ozan Toker,Selçuk Alp,Ersoy Oz,Orhan _ Içelli 한국원자력학회 2022 Nuclear Engineering and Technology Vol.54 No.1
The mass attenuation coefficient is the primary physical parameter to model narrow beam gamma-rayattenuation. A new machine learning based approach is proposed to model gamma-ray shieldingbehavior of composites alternative to theoretical calculations. Two fuzzy logic algorithms and a neuralnetwork algorithm were trained and tested with different mixture ratios of vanadium slag/epoxy resin/antimony in the 0.05 MeVe2 MeV energy range. Two of the algorithms showed excellent agreementwith testing data after optimizing adjustable parameters, with root mean squared error (RMSE) valuesdown to 0.0001. Those results are remarkable because mass attenuation coefficients are often presentedwith four significant figures. Different training data sizes were tried to determine the least number ofdata points required to train sufficient models. Data set size more than 1000 is seen to be required tomodel in above 0.05 MeV energy. Below this energy, more data points with finer energy resolution mightbe required. Neuro-fuzzy models were three times faster to train than neural network models, whileneural network models depicted low RMSE. Fuzzy logic algorithms are overlooked in complex functionapproximation, yet grid partitioned fuzzy algorithms showed excellent calculation efficiency and goodconvergence in predicting mass attenuation coefficient.