• Nuclear Engineering and Technology
• /
• v.55 no.9
• /
• pp.3423-3440
• /
• 2023

In this work, a multivariate time-series machine learning meta-model is developed to predict the transient response of a typical nuclear power plant (NPP) undergoing a steam generator tube rupture (SGTR). The model employs Recurrent Neural Networks (RNNs), including the Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid CNN-LSTM model. To address the uncertainty inherent in such predictions, a Bayesian Neural Network (BNN) was implemented. The models were trained using a database generated by the Best Estimate Plus Uncertainty (BEPU) methodology; coupling the thermal hydraulics code, RELAP5/SCDAP/MOD3.4 to the statistical tool, DAKOTA, to predict the variation in system response under various operational and phenomenological uncertainties. The RNN models successfully captures the underlying characteristics of the data with reasonable accuracy, and the BNN-LSTM approach offers an additional layer of insight into the level of uncertainty associated with the predictions. The results demonstrate that LSTM outperforms GRU, while the hybrid CNN-LSTM model is computationally the most efficient. This study aims to gain a better understanding of the capabilities and limitations of machine learning models in the context of nuclear safety. By expanding the application of ML models to more severe accident scenarios, where operators are under extreme stress and prone to errors, ML models can provide valuable support and act as expert systems to assist in decision-making while minimizing the chances of human error.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.460-474
• /
• 2023

A simple but accurate mathematical model is crucial for dynamic simulations and controller design of helical coil once-through steam generator (OTSG). This paper presents a three-region movable boundary dynamic model of the helical coil OTSG. Based on the secondary side fluid conditions, the OTSG is divided into subcooled region (two control volumes), two-phase region (two control volumes) and superheated region (three control volumes) with movable boiling boundaries between each region. The nonlinear dynamic model is derived based on mass, energy and momentum conservation equations. And the linear model is obtained by using the transfer function and state space transformation, which is a 37-order model of five input and three output. Validations are made under full-power steady-state condition and four transient conditions. Results show good agreements among the nonlinear model, linear model and the RELAP5 model, with acceptable errors. This model can be applied to dynamic simulations and controller design of helical coil OTSG with constant primary-side flow rate.

• Nuclear Engineering and Technology
• /
• v.55 no.5
• /
• pp.1651-1664
• /
• 2023

It is very difficult to capture the multi-dimensional phenomena such as asymmetric flow and temperature distributions with the one-dimensional (1D) model, obviously, due to its inherent limitation. In order to overcome such a limitation of the 1D representation, many state-of-the-art system codes have equipped a three-dimensional (3D) component for multi-dimensional analysis capability. In this study, a standard multi-dimensional analysis model of APR1400 (Advanced Power Reactor 1400) has been developed using TRACE (TRAC/RELAP Advanced Computational Engine). The entire reactor pressure vessel (RPV) of APR1400 has been modeled using a single 3D component. The fuels in the reactor core have been described with detailed and coarse representations, respectively, to figure out the impact of the fuel description. Using both 3D RPV models, a comparative analysis has been performed postulating a double-ended guillotine break at a cold leg. Based on the results of comparative analysis, it is revealed that both models show no significant difference in general plant behavior and the model with coarse fuel model could be used for faster transient analysis without reactor kinetics coupling. The analysis indicates that the asymmetric temperature and flow distributions are captured during the transient, and such nonuniform distributions contribute to asymmetric quenching behaviors during blowdown and reflood phases. Such asymmetries are directly connected to the figure of merits in the LBLOCA analysis. Therefore, it is recommended to employ a multi-dimensional RPV model with a detailed fuel description for a realistic safety analysis with the consideration of the spatial configuration of the reactor core.

• Nuclear Engineering and Technology
• /
• v.45 no.5
• /
• pp.605-612
• /
• 2013

Removal of decay heat from an operating reactor during a prolonged station blackout condition is a big concern for reactor designers, especially after the recent Fukushima accident. In the case of a prolonged station blackout condition, heat removal is possible only by passive means since no pumps or active systems are available. Keeping this in mind, the AHWR has been designed with many passive safety features. One of them is a passive means of removing decay heat with the help of Isolation Condensers (ICs) which are submerged in a big water pool called the Gravity Driven Water Pool (GDWP). The ICs have many tubes in which the steam, generated by the reactor core due to the decay heat, flows and condenses by rejecting the heat into the water pool. After condensation, the condensate falls back into the steam drum of the reactor. The GDWP tank holds a large amount of water, about 8000 $m^3$, which is located at a higher elevation than the steam drum of the reactor in order to promote natural circulation. Due to the recent Fukushima type accidents, it has been a concern to understand and evaluate the capability of the ICs to remove decay heat for a prolonged period without escalating fuel sheath temperature. In view of this, an analysis has been performed for decay heat removal characteristics over several days of an AHWR by ICs. The computer code RELAP5/MOD3.2 was used for this purpose. Results indicate that the ICs can remove the decay heat for more than 10 days without causing any bulk boiling in the GDWP. After that, decay heat can be removed for more than 40 days by boiling off the pool inventory. The pressure inside the containment does not exceed the design pressure even after 10 days by condensation of steam generated from the GDWP on the walls of containment and on the Passive Containment Cooling System (PCCS) tubes. If venting is carried out after this period, the decay heat can be removed for more than 50 days without exceeding the design limits.

• Nuclear Engineering and Technology
• /
• v.43 no.1
• /
• pp.25-44
• /
• 2011

KAERI has been operating an integral effect test facility, ATLAS (Advanced Thermal-Hydraulic Test Loop for Accident Simulation), for accident simulations of advanced PWRs. Regarding integral effect tests, a database for major design basis accidents has been accumulated and a Domestic Standard Problem (DSP) exercise using the ATLAS has been proposed and successfully performed. The ATLAS DSP aims at the effective utilization of an integral effect database obtained from the ATLAS, the establishment of a cooperative framework in the domestic nuclear industry, better understanding of thermal hydraulic phenomena, and an investigation of the potential limitations of the existing best-estimate safety analysis codes. For the first ATLAS DSP exercise (DSP-01), integral effect test data for a 100% DVI line break accident of the APR1400 was selected by considering its technical importance and by incorporating comments from participants. Twelve domestic organizations joined in this DSP-01 exercise. Finally, ten of these organizations submitted their calculation results. This ATLAS DSP-01 exercise progressed as an open calculation; the integral effect test data was delivered to the participants prior to the code calculations. The MARS-KS was favored by most participants but the RELAP5/MOD3.3 code was also used by a few participants. This paper presents all the information of the DSP-01 exercise as well as the comparison results between the calculations and the test data. Lessons learned from the first DSP-01 are presented and recommendations for code users as well as for developers are suggested.

• Nuclear Engineering and Technology
• /
• v.55 no.12
• /
• pp.4357-4366
• /
• 2023

MARS-KS, a domestic regulatory confirmatory code of Republic of Korea, had been developed by integrating RELAP5/MOD2 and COBRA-TF. The integration of COBRA-TF allowed to extend the capability of MARS-KS, limited to one-dimensional analysis, to multi-dimensional analysis. The use of COBRA-TF was mainly focused on subchannel analyses for simulating multi-dimensional behavior within the reactor core. However, this feature has been remained as a legacy without ongoing maintenance. Meanwhile, MARS-KS also includes its own multidimensional component, namely MULTID, which is also feasible to simulate three-dimensional convection and diffusion. The MULTID is capable of modeling the turbulent diffusion using simple mixing length model. The implementation of the turbulent mixing is of importance for analyzing the reactor core where a disturbing cross-sectional structure of rod bundle makes the flow perturbation and corresponding mixing stronger. In addition, the presence of this turbulent behavior allows the secondary transports with net mass exchange between subchannels. However, a series of assessments performed in previous studies revealed that the turbulence model of the MULTID could not simulate the aforementioned effective mixing occurred in the subchannel-scale problems. This is obvious consequence since the physical models of the MULTID neglect the effect of mass transport and thereby, it cannot model the void drift effect and resulting phasic distribution within a bundle. Thus, in this study, the turbulence mixing model of the MULTID has been improved by means of the inter-channel mixing model, widely utilized in subchannel analysis, in order to extend the application of the MULTID to small-scale problems. A series of assessments has been performed against rod bundle experiments, namely GE 3X3 and PSBT, to evaluate the performance of the introduced mixing model. The assessment results revealed that the application of the inter-channel mixing model allowed to enhance the prediction of the MULTID in subchannel scale problems. In addition, it was indicated that the code could not predict appropriate phasic distribution in the rod bundle without the model. Considering that the proper prediction of the phasic distribution is important when considering pin-based and/or assembly-based expressions of the reactor core, the results of this study clearly indicate that the inter-channel mixing model is required for analyzing the rod bundle, appropriately.