Robust artificial neural network for reliability and sensitivity analysis of complex non-linear systems

Oparaji, Uchenna Bright and Rong-Jiun, Sheu and Bankhead, Mark and Austin, Jonathan and Patelli, Edoardo (2017) Robust artificial neural network for reliability and sensitivity analysis of complex non-linear systems. Neural Networks, 96. pp. 80-90. ISSN 0893-6080

[thumbnail of Oparaji-etal-NN2017-Robust-artificial-neural-network-reliability-sensitivity-analysis-complex-non-linear-systems]
Text (Oparaji-etal-NN2017-Robust-artificial-neural-network-reliability-sensitivity-analysis-complex-non-linear-systems)
Accepted Author Manuscript
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo

Download (1MB)| Preview


    Artificial Neural Networks (ANNs) are commonly used in place of expensive models to reduce the computational burden required for uncertainty quantifcation, reliability and sensitivity analysis. ANN with selected architecture is trained with the back-propagation algorithm from few data representatives of the input/output relationship of the underlying model of interest. However, different performing ANNs might be obtained with the same training data as a result of the random initialization of the weight parameters in each of the network, leading to an uncertainty in selecting the best performing ANN. On the other hand, using cross-validation to select the best performing ANN based on the ANN with the highest R2 value can lead to biassing in the prediction. This is as a result of the fact that the use of R2 cannot determine if the prediction made by ANN is biased. Additionally, R2 does not indicate if a model is adequate, as it is possible to have a low R2 for a good model and a high R2 for a bad model. Hence in this paper, we propose an approach to improve the robustness of a prediction made by ANN. The approach is based on a systematic combination of identical trained ANNs, by coupling the Bayesian framework and model averaging. Additionally, the uncertainties of the robust prediction derived from the approach are quantified in terms of condence intervals. To demonstrate the applicability of the proposed approach, two synthetic numerical examples are presented. Finally, the proposed approach is used to perform a reliability and sensitivity analysis on a process simulation model of a UK nuclear effluent treatment plant developed by National Nuclear Laboratory (NNL) and treated in this study as a black-box employing a set of training data as a test case. This model has been extensively validated against plant and experimental data and used to support the UK effluent discharge strategy.

    ORCID iDs

    Oparaji, Uchenna Bright, Rong-Jiun, Sheu, Bankhead, Mark, Austin, Jonathan and Patelli, Edoardo ORCID logoORCID:;