Development of improved cubic spline method for synthesizing the axial power distribution in a nuclear reactor

Abstract

*The improved cubic spline method using new function sets and basis functions is developed for accurately predicting the Axial Power Distribution (APD) in a nuclear reactor. The improved function sets and basis functions are derived by analyzing 3,000 APDs from measured and calculated data for the OPR-1000 reactor (PWR). When comparing conventional and improved methods, APDs produced by improved method represent an average Root Mean Square (RMS) error of 1.68%, on the basis of reference data, whereas the other has an average RMS error more than 2.40%. Especially, the performance improvement is remarkably shown in APDs classified as the center peak and saddle types. The average RMS errors in the center peak and saddle types decrease by 21.03 and 51.36% respectively. The cubic spline method established in this study would complement the conventional method and enhance the accuracy to within the average RMS error of 1.80% in center peak, flat, and saddle types. It is expected that the unnecessary reactor trip by inaccurate prediction of power distributions would be reduced as well as improve the safety and economics of nuclear power plant by performance improvement of the cubic spline method in CPCS.