Kyung Doo Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Kyung Doo Kim has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Hae-Yong Jeong
Department of Quantum and Nuclear Engineering, Sejong University, 209, Neungdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea
2 shared publications - Mai The Vu
Department of Artificial Intelligence and Robotics, Sejong University, Seoul, Republic of Korea
2 shared publications - Nguyen Huu Tiep
Department of Quantum and Nuclear Engineering, Sejong University, 209, Neungdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea
2 shared publications - Nguyen Ngoc Anh
Phenikaa University
2 shared publications - Nguyen Xuan-Mung
Faculty of Mechanical and Aerospace Engineering, Sejong University, 209, Neungdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea
2 shared publications - Van-Khanh Hoang
PHENIKAA Institute for Advanced Study, PHENIKAA University, Hanoi 12116, Viet Nam
2 shared publications
Latest Publications
- Open Access
- 10 December 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (24) , 3892
A New Hyperparameter Tuning Framework for Regression Tasks in Deep Neural Network: Combined-Sampling Algorithm to Search the Optimized Hyperparameters
This paper introduces a novel hyperparameter optimization framework for regression tasks called the Combined-Sampling Algorithm to Search the Optimized Hyperparameters (CASOH). Our approach enables hyperparameter tuning for deep learning models with two hidden layers and multiple types of hyperparameters, enhancing the model’s capacity to work with complex optimization problems. The primary goal is to improve hyperparameter tuning performance in deep learning models compared to conventional methods such as Bayesian Optimization and Random Search. Furthermore, CASOH is evaluated alongside the state-of-the-art hyperparameter reinforcement learning (Hyp-RL) framework to ensure a comprehensive assessment. The CASOH framework integrates the Metropolis-Hastings algorithm with a uniform random sampling approach, increasing the likelihood of identifying promising hyperparameter configurations. Specifically, we developed a correlation between the objective function and samples, allowing subsequent samples to be strongly correlated with the current sample by applying an acceptance probability in our sampling algorithm. The effectiveness of our proposed method was examined using regression datasets such as Boston Housing, Critical heat flux (CHF), Concrete compressive strength, Combined Cycle Power Plant, Gas Turbine CO, and NOx Emission, as well as an ‘in-house’ dataset of lattice-physics parameters generated from a Monte Carlo code for nuclear fuel assembly simulation. One of the primary goals of this study is to construct an optimized deep-learning model capable of accurately predicting lattice-physics parameters for future applications of machine learning in nuclear reactor analysis. Our results indicate that this framework achieves competitive accuracy compared to conventional random search and Bayesian optimization methods. The most significant enhancement was observed in the lattice-physics dataset, achieving a 56.6% improvement in prediction accuracy, compared to improvements of 53.2% by Hyp-RL, 44.9% by Bayesian optimization, and 38.8% by random search relative to the nominal prediction. While the results are promising, further empirical validation across a broader range of datasets would be helpful to better assess the framework’s suitability for optimizing hyperparameters in complex problems involving high-dimensional parameters, highly non-linear systems, and multi-objective optimization tasks.
- 10 December 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (24) , 3892
Open AccessA New Hyperparameter Tuning Framework for Regression Tasks in Deep Neural Network: Combined-Sampling Algorithm to Search the Optimized Hyperparameters
This paper introduces a novel hyperparameter optimization framework for regression tasks called the Combined-Sampling Algorithm to Search the Optimized Hyperparameters (CASOH). Our approach enables hyperparameter tuning for deep learning models with two hidden layers and multiple types of hyperparameters, enhancing the model’s capacity to work with complex optimization problems. The primary goal is to improve hyperparameter tuning performance in deep learning models compared to conventional methods such as Bayesian Optimization and Random Search. Furthermore, CASOH is evaluated alongside the state-of-the-art hyperparameter reinforcement learning (Hyp-RL) framework to ensure a comprehensive assessment. The CASOH framework integrates the Metropolis-Hastings algorithm with a uniform random sampling approach, increasing the likelihood of identifying promising hyperparameter configurations. Specifically, we developed a correlation between the objective function and samples, allowing subsequent samples to be strongly correlated with the current sample by applying an acceptance probability in our sampling algorithm. The effectiveness of our proposed method was examined using regression datasets such as Boston Housing, Critical heat flux (CHF), Concrete compressive strength, Combined Cycle Power Plant, Gas Turbine CO, and NOx Emission, as well as an ‘in-house’ dataset of lattice-physics parameters generated from a Monte Carlo code for nuclear fuel assembly simulation. One of the primary goals of this study is to construct an optimized deep-learning model capable of accurately predicting lattice-physics parameters for future applications of machine learning in nuclear reactor analysis. Our results indicate that this framework achieves competitive accuracy compared to conventional random search and Bayesian optimization methods. The most significant enhancement was observed in the lattice-physics dataset, achieving a 56.6% improvement in prediction accuracy, compared to improvements of 53.2% by Hyp-RL, 44.9% by Bayesian optimization, and 38.8% by random search relative to the nominal prediction. While the results are promising, further empirical validation across a broader range of datasets would be helpful to better assess the framework’s suitability for optimizing hyperparameters in complex problems involving high-dimensional parameters, highly non-linear systems, and multi-objective optimization tasks.
- Open Access
- 29 December 2023
- Published by MDPI AG in Applied Sciences
- Vol. 14 (1) , 324
Machine Learning Applications and Uncertainty Quantification Analysis for Reflood Tests
The reflooding phase, a crucial recovery process after a loss of coolant accident (LOCA) in reactors, involves cooling overheated fuel rods with subcooled water. Its complex nature, notably in its flow regime and heat transfer, makes prediction challenging, resulting in high uncertainty and computation cost. In this study, we utilized the data assimilation (DA) technique to enhance the prediction of reflooding phenomena and subsequently deployed machine learning models to predict the accuracy of the safety and performance analysis code (SPACE) simulation. To generate the dataset for the machine learning model, we employed the sampling method for highly nonlinear system uncertainty analysis (STARU), providing a high-quality dataset for a complex problem such as a reflooding simulation. In this dataset, the physical models were assimilated under their selected uncertainty bands and utilized the effective sampling approach of STARU, generating the high-quality output and efficient enhancement of SPACE predictions. Consequently, the implemented machine learning model can be used to enhance model development and uncertainty quantification (UQ) analysis using the system code.
- 29 December 2023
- Published by MDPI AG in Applied Sciences
- Vol. 14 (1) , 324
Open AccessMachine Learning Applications and Uncertainty Quantification Analysis for Reflood Tests
The reflooding phase, a crucial recovery process after a loss of coolant accident (LOCA) in reactors, involves cooling overheated fuel rods with subcooled water. Its complex nature, notably in its flow regime and heat transfer, makes prediction challenging, resulting in high uncertainty and computation cost. In this study, we utilized the data assimilation (DA) technique to enhance the prediction of reflooding phenomena and subsequently deployed machine learning models to predict the accuracy of the safety and performance analysis code (SPACE) simulation. To generate the dataset for the machine learning model, we employed the sampling method for highly nonlinear system uncertainty analysis (STARU), providing a high-quality dataset for a complex problem such as a reflooding simulation. In this dataset, the physical models were assimilated under their selected uncertainty bands and utilized the effective sampling approach of STARU, generating the high-quality output and efficient enhancement of SPACE predictions. Consequently, the implemented machine learning model can be used to enhance model development and uncertainty quantification (UQ) analysis using the system code.
- Open Access
- 1 December 2021
- Published by Elsevier in Nuclear Engineering and Design
- Vol. 385
Development of singular value decomposition based solution scheme for a numerical simulation of multiphase flow
- 1 December 2021
- Published by Elsevier in Nuclear Engineering and Design
- Vol. 385
Open AccessDevelopment of singular value decomposition based solution scheme for a numerical simulation of multiphase flow
- 22 June 2018
- Published by Taylor & Francis in Nuclear Technology
- Vol. 204 (2) , 162-171
Improvement of One-Dimensional Two-Fluid Momentum Conservation Equations for Vertically Stratified Flow
This paper deals with numerical challenges associated with simulating thermal-hydraulic phenomena in nuclear reactors with one-dimensional system analysis codes. The main focus of this research is directed toward assessment of the pressure gradient in vertically stratified flow, particularly the separate pressure drop effects for gas and liquid phases along the control cell. The pressure drop term in momentum conservation currently being developed based on the assumption of gas and liquid combined pressure drop was redefined such that two different pressures were imposed for gas and liquid separately. The verification of the proposed momentum equation for a vertically stratified flow was completed through simulations of the liquid velocity in a U-shaped manometer. Sensitivity analysis was also performed by increasing liquid mass in the pipe leading to different positions of the liquid-vapor interface from the bottom of each manometer pipe when the flow oscillation is stopped; i.e., the interfaces are not only cell boundaries but also various positions between cell edges. As a result, improved simulation results were obtained using the modified equations as it was indicated that the oscillation of fluid decays over time while the original solution for the large pipe does not converge to zero due to a mainly incorrect pressure drop term.
- 22 June 2018
- Published by Taylor & Francis in Nuclear Technology
- Vol. 204 (2) , 162-171
Improvement of One-Dimensional Two-Fluid Momentum Conservation Equations for Vertically Stratified Flow
This paper deals with numerical challenges associated with simulating thermal-hydraulic phenomena in nuclear reactors with one-dimensional system analysis codes. The main focus of this research is directed toward assessment of the pressure gradient in vertically stratified flow, particularly the separate pressure drop effects for gas and liquid phases along the control cell. The pressure drop term in momentum conservation currently being developed based on the assumption of gas and liquid combined pressure drop was redefined such that two different pressures were imposed for gas and liquid separately. The verification of the proposed momentum equation for a vertically stratified flow was completed through simulations of the liquid velocity in a U-shaped manometer. Sensitivity analysis was also performed by increasing liquid mass in the pipe leading to different positions of the liquid-vapor interface from the bottom of each manometer pipe when the flow oscillation is stopped; i.e., the interfaces are not only cell boundaries but also various positions between cell edges. As a result, improved simulation results were obtained using the modified equations as it was indicated that the oscillation of fluid decays over time while the original solution for the large pipe does not converge to zero due to a mainly incorrect pressure drop term.