%0 Journal Article
%T Neural Network Modeling for Evaluating Sodium Temperature of Intermediate Heat Exchanger of Fast Breeder Reactor
%J Advances in Computing
%@ 2163-2979
%D 2012
%I 
%R 10.5923/j.ac.20120202.03
%X This Sodium temperature estimation in Intermediate Heat Exchanger is very significant for nuclear power generation in fast breeder test reactor (FBTR). Hence accurate evaluation of sodium temperature is a major concern both in case of offline and online operation of nuclear power plant (NPP). This paper addresses the training of artificial neural network model to precisely estimate the sodium temperature of Sodium-Sodium (Na-Na) Intermediate Heat exchanger and studying its behaviour at transient conditions. Severely unbalanced flow conditions in addition to steady state condition are investigated to generate sufficient number of dataset. Based on the in house data gathered from Quadratic Upstream Interpolation for Convective Kinetics code (QUICK), a three layer neural network model is developed for training and subsequent validation. The back propagation (BP) algorithm is used for training the network. Further a model based on Radial Basis Function (RBF) neural network is developed and trained and the results are compared with standard back propagation algorithm. From the comparison studies, it is found that the network trained with RBF converges faster than BP network. Training and testing results show the successful modelling of plant dynamics of the reactor with improved accuracy. ANN can be an alternative to the conventional model as it predicts the physical parameters without much complex calculations as used in conventional model.
%K Artificial Neural Network
%K Na-Na Heat Exchanger
%K Multi Layer Perceptron
%K Back Propagation Algorithm
%K Radial Basis Function
%K Fast Reactors
%U http://article.sapub.org/10.5923.j.ac.20120202.03.html