Pregled bibliografske jedinice broj: 323286
Machine Learning of the Reactor Core Loading Pattern Critical Parameters
Machine Learning of the Reactor Core Loading Pattern Critical Parameters // Proceedings of the International Conference Nuclear Energy for New Europe 2007 / Jenčić, Igor ; Lenošek, Melita (ur.).
Ljubljana: Nuclear Society of Slovenia, 2007. str. 113.1-113.10 (poster, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 323286 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Machine Learning of the Reactor Core Loading Pattern Critical Parameters
Autori
Trontl, Krešimir ; Pevec, Dubravko ; Šmuc, Tomislav
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the International Conference Nuclear Energy for New Europe 2007
/ Jenčić, Igor ; Lenošek, Melita - Ljubljana : Nuclear Society of Slovenia, 2007, 113.1-113.10
ISBN
978-961-6207-28-7
Skup
International Conference Nuclear Energy for New Europe 2007
Mjesto i datum
Portorož, Slovenija, 10.09.2007. - 13.09.2007
Vrsta sudjelovanja
Poster
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
Support Vector Regression; SVR; reactor physics; machine learning
Sažetak
The usual approach to loading pattern optimization involves high degree of engineering judgment, a set of heuristic rules, an optimization algorithm and a computer code used for evaluating proposed loading patterns. The speed of the optimization process is highly dependent on the computer code used for the evaluation. In this paper we investigate the applicability of a machine learning model which could be used for fast loading pattern evaluation. We employed a recently introduced machine learning technique, Support Vector Regression (SVR), which has a strong theoretical background in statistical learning theory. Superior empirical performance of the method has been reported on difficult regression problems in different fields of science and technology. SVR is a data driven, kernel based, nonlinear modelling paradigm, in which model parameters are automatically determined by solving a quadratic optimization problem. The main objective of the work reported in this paper was to evaluate the possibility of applying SVR method for reactor core loading pattern modelling. The starting set of experimental data for training and testing of the machine learning algorithm was obtained using a two-dimensional diffusion theory reactor physics computer code. We illustrate the performance of the solution and discuss its applicability, i.e., complexity, speed and accuracy, with a projection to a more realistic scenario involving machine learning from the results of more accurate and time consuming three-dimensional core modelling code.
Izvorni jezik
Engleski
Znanstvena područja
Elektrotehnika
POVEZANOST RADA
Projekti:
036-0361590-1579 - Gospodarenje gorivom standardnih i naprednih nuklearnih reaktora (Pevec, Dubravko, MZO ) ( CroRIS)
098-0982560-2565 - Postupci računalne inteligencije u mjernim sustavima (Marić, Ivan, MZOS ) ( CroRIS)
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb,
Institut "Ruđer Bošković", Zagreb
