Optimal artificial neural network architecture selection for performance prediction of compact heat exchanger with the EBaLM-OTR technique

Journal Article · Fri Jul 01 04:00:00 EDT 2011 · Nuclear Engineering and Design

DOI:https://doi.org/10.1016/j.nucengdes.2011.04.045· OSTI ID:1029332

Wijayasekara, Dumidu; Manic, Milos; Sabharwall, Piyush; Utgikar, Vivek

Artificial Neural Networks (ANN) have been used in the past to predict the performance of printed circuit heat exchangers (PCHE) with satisfactory accuracy. Typically published literature has focused on optimizing ANN using a training dataset to train the network and a testing dataset to evaluate it. Although this may produce outputs that agree with experimental results, there is a risk of over-training or overlearning the network rather than generalizing it, which should be the ultimate goal. An over-trained network is able to produce good results with the training dataset but fails when new datasets with subtle changes are introduced. In this paper we present EBaLM-OTR (error back propagation and Levenberg-Marquardt algorithms for over training resilience) technique, which is based on a previously discussed method of selecting neural network architecture that uses a separate validation set to evaluate different network architectures based on mean square error (MSE), and standard deviation of MSE. The method uses k-fold cross validation. Therefore in order to select the optimal architecture for the problem, the dataset is divided into three parts which are used to train, validate and test each network architecture. Then each architecture is evaluated according to their generalization capability and capability to conform to original data. The method proved to be a comprehensive tool in identifying the weaknesses and advantages of different network architectures. The method also highlighted the fact that the architecture with the lowest training error is not always the most generalized and therefore not the optimal. Using the method the testing error achieved was in the order of magnitude of within 10{sup -5} - 10{sup -3}. It was also show that the absolute error achieved by EBaLM-OTR was an order of magnitude better than the lowest error achieved by EBaLM-THP.

• Research Organization: Idaho National Laboratory (INL)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC07-05ID14517
• OSTI ID: 1029332
• Report Number(s): INL/JOU-11-23922
• Journal Information: Nuclear Engineering and Design, Journal Name: Nuclear Engineering and Design Journal Issue: 7 Vol. 241; ISSN NEDEAU; ISSN 0029-5493
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS
ACCURACY
ALGORITHMS
COMPUTER ARCHITECTURE
FORECASTING
HEAT EXCHANGERS
NEURAL NETWORKS
PERFORMANCE
PRINTED CIRCUITS
TESTING
TRAINING
VALIDATION
artificial neural network