@article{xie_yaseen_wu_2024, title={Functional PCA and deep neural networks-based Bayesian inverse uncertainty quantification with transient experimental data}, volume={420}, ISSN={["1879-2138"]}, url={https://doi.org/10.1016/j.cma.2023.116721}, DOI={10.1016/j.cma.2023.116721}, abstractNote={This work focuses on developing an inverse uncertainty quantification (IUQ) process for time-dependent responses, using dimensionality reduction by functional principal component analysis (PCA) and deep neural network (DNN)-based surrogate models. The demonstration is based on the IUQ of TRACE physical model parameters using the FEBA benchmark transient experimental data on peak cladding temperatures during reflooding. The quantity-of-interest (QoI) is time-dependent peak cladding temperature (PCT) profiles. Conventional PCA can hardly represent the data precisely due to the sudden temperature drop at the time of quenching. As a result, a functional alignment method is used to separate the phase and amplitude information in the PCT profiles before conventional PCA is applied for dimensionality reduction. The resulting PC scores are then used to build DNN-based surrogate models to significantly reduce the computational cost in Markov Chain Monte Carlo sampling, while the code/interpolation uncertainty is accounted for using Bayesian neural networks. We compared four IUQ processes with different dimensionality reduction methods and surrogate models. The proposed approach has demonstrated the best performance in reducing the dimensionality of the transient PCT profiles. This approach also produces the posterior distributions of the physical model parameters with which the model predictions have the best agreement with the experimental data.}, journal={COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING}, author={Xie, Ziyu and Yaseen, Mahmoud and Wu, Xu}, year={2024}, month={Feb} }
 @article{yaseen_yushu_german_wu_2023, title={Fast and accurate reduced-order modeling of a MOOSE-based additive manufacturing model with operator learning}, volume={129}, ISSN={["1433-3015"]}, DOI={10.1007/s00170-023-12471-1}, number={7-8}, journal={INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY}, author={Yaseen, Mahmoud and Yushu, Dewen and German, Peter and Wu, Xu}, year={2023}, month={Dec}, pages={3123–3139} }
 @article{yaseen_wu_2022, title={Quantification of Deep Neural Network Prediction Uncertainties for VVUQ of Machine Learning Models}, volume={11}, ISSN={["1943-748X"]}, url={https://doi.org/10.1080/00295639.2022.2123203}, DOI={10.1080/00295639.2022.2123203}, abstractNote={Abstract Recent performance breakthroughs in artificial intelligence (AI) and machine learning (ML), especially advances in deep learning, the availability of powerful and easy-to-use ML libraries (e.g., scikit-learn, TensorFlow, PyTorch), and increasing computational power, have led to unprecedented interest in AI/ML among nuclear engineers. For physics-based computational models, verification, validation, and uncertainty quantification (VVUQ) processes have been very widely investigated, and many methodologies have been developed. However, VVUQ of ML models has been relatively less studied, especially in nuclear engineering. This work focuses on uncertainty quantification (UQ) of ML models as a preliminary step of ML VVUQ, more specifically Deep Neural Networks (DNNs) because they are the most widely used supervised ML algorithm for both regression and classification tasks. This work aims at quantifying the prediction or approximation uncertainties of DNNs when they are used as surrogate models for expensive physical models. Three techniques for UQ of DNNs are compared, namely, Monte Carlo Dropout (MCD), Deep Ensembles (DE), and Bayesian Neural Networks (BNNs). Two nuclear engineering examples are used to benchmark these methods: (1) time-dependent fission gas release data using the Bison code and (2) void fraction simulation based on the Boiling Water Reactor Full-size Fine-Mesh Bundle Tests (BFBT) benchmark using the TRACE code. It is found that the three methods typically require different DNN architectures and hyperparameters to optimize their performance. The UQ results also depend on the amount of training data available and the nature of the data. Overall, all three methods can provide reasonable estimations of the approximation uncertainties. The uncertainties are generally smaller when the mean predictions are close to the test data while the BNN methods usually produce larger uncertainties than MCD and DE.}, journal={NUCLEAR SCIENCE AND ENGINEERING}, author={Yaseen, Mahmoud and Wu, Xu}, year={2022}, month={Nov} }