Simplex optimization of artificial neural networks for the prediction of minimum detectable activity in gamma-ray spectrometry
Abstract
A three-layer feed-forward artificial neural network (ANN) with a back-propagation learning algorithm was used to predict the minimum detectable activity (AD) of radionuclides (Ra-226, U-238, U-235, K-40, Th-232, Cs-134, Cs-137 and Be-7) in environmental soil samples as a function of measurement time. The ANN parameters (learning rate, momentum, number of epochs, and the number of nodes in the hidden layer) were optimized simultaneously employing a variable-size simplex method. The optimized ANN model revealed satisfactory predictions, with correlation coefficients between experimental and predicted values 0.9517 for 232 Th (sample with U-238/Th-232 ratio of 1.14) to 0.9995 for K-40 (sample with U-238/Th-232 ratio of 0.43). Neither the differences between the measured and the predicted A(D) values nor the correlation coefficients were influenced by the absolute values of AD for the investigated radionuclides.
Keywords:
ANN / radionuclides / minimum detectable activity / simplex / soilSource:
Nuclear Instruments & Methods in Physics Research Section A-Accelerators Spectrometers Detectors And, 2006, 564, 1, 308-314Publisher:
- Elsevier, Amsterdam
Funding / projects:
- Nove metode i tehnike za separaciju i specijaciju hemijskih elemenata u tragovima, organskih supstanci i radionuklida i identifikaciju njihovih izvora (RS-MESTD-MPN2006-2010-142039)
DOI: 10.1016/j.nima.2006.03.047
ISSN: 0168-9002
WoS: 000239669500039
Scopus: 2-s2.0-33745924967
Institution/Community
Tehnološko-metalurški fakultetTY - JOUR AU - Dragović, Snežana AU - Onjia, Antonije AU - Bacić, Goran PY - 2006 UR - http://TechnoRep.tmf.bg.ac.rs/handle/123456789/970 AB - A three-layer feed-forward artificial neural network (ANN) with a back-propagation learning algorithm was used to predict the minimum detectable activity (AD) of radionuclides (Ra-226, U-238, U-235, K-40, Th-232, Cs-134, Cs-137 and Be-7) in environmental soil samples as a function of measurement time. The ANN parameters (learning rate, momentum, number of epochs, and the number of nodes in the hidden layer) were optimized simultaneously employing a variable-size simplex method. The optimized ANN model revealed satisfactory predictions, with correlation coefficients between experimental and predicted values 0.9517 for 232 Th (sample with U-238/Th-232 ratio of 1.14) to 0.9995 for K-40 (sample with U-238/Th-232 ratio of 0.43). Neither the differences between the measured and the predicted A(D) values nor the correlation coefficients were influenced by the absolute values of AD for the investigated radionuclides. PB - Elsevier, Amsterdam T2 - Nuclear Instruments & Methods in Physics Research Section A-Accelerators Spectrometers Detectors And T1 - Simplex optimization of artificial neural networks for the prediction of minimum detectable activity in gamma-ray spectrometry EP - 314 IS - 1 SP - 308 VL - 564 DO - 10.1016/j.nima.2006.03.047 ER -
@article{ author = "Dragović, Snežana and Onjia, Antonije and Bacić, Goran", year = "2006", abstract = "A three-layer feed-forward artificial neural network (ANN) with a back-propagation learning algorithm was used to predict the minimum detectable activity (AD) of radionuclides (Ra-226, U-238, U-235, K-40, Th-232, Cs-134, Cs-137 and Be-7) in environmental soil samples as a function of measurement time. The ANN parameters (learning rate, momentum, number of epochs, and the number of nodes in the hidden layer) were optimized simultaneously employing a variable-size simplex method. The optimized ANN model revealed satisfactory predictions, with correlation coefficients between experimental and predicted values 0.9517 for 232 Th (sample with U-238/Th-232 ratio of 1.14) to 0.9995 for K-40 (sample with U-238/Th-232 ratio of 0.43). Neither the differences between the measured and the predicted A(D) values nor the correlation coefficients were influenced by the absolute values of AD for the investigated radionuclides.", publisher = "Elsevier, Amsterdam", journal = "Nuclear Instruments & Methods in Physics Research Section A-Accelerators Spectrometers Detectors And", title = "Simplex optimization of artificial neural networks for the prediction of minimum detectable activity in gamma-ray spectrometry", pages = "314-308", number = "1", volume = "564", doi = "10.1016/j.nima.2006.03.047" }
Dragović, S., Onjia, A.,& Bacić, G.. (2006). Simplex optimization of artificial neural networks for the prediction of minimum detectable activity in gamma-ray spectrometry. in Nuclear Instruments & Methods in Physics Research Section A-Accelerators Spectrometers Detectors And Elsevier, Amsterdam., 564(1), 308-314. https://doi.org/10.1016/j.nima.2006.03.047
Dragović S, Onjia A, Bacić G. Simplex optimization of artificial neural networks for the prediction of minimum detectable activity in gamma-ray spectrometry. in Nuclear Instruments & Methods in Physics Research Section A-Accelerators Spectrometers Detectors And. 2006;564(1):308-314. doi:10.1016/j.nima.2006.03.047 .
Dragović, Snežana, Onjia, Antonije, Bacić, Goran, "Simplex optimization of artificial neural networks for the prediction of minimum detectable activity in gamma-ray spectrometry" in Nuclear Instruments & Methods in Physics Research Section A-Accelerators Spectrometers Detectors And, 564, no. 1 (2006):308-314, https://doi.org/10.1016/j.nima.2006.03.047 . .