Get started
Work

Thermal-Hydraulic Analysis of Passive Safety Systems Performance in Advanced Korean-Designed Reactors During SBLOCA Sequences Using the TRACE Code

Year: 2025
Type: dissertation
Abstract: The resurgence of nuclear energy as a safe, reliable and low-carbon power source is driving the development of advanced reactor technologies equipped with innovative passive safety systems. These systems rely on natural physical phenomena such as gravity, condensation and natural circulation rather than active components or external power. Their successful implementation promises to enhance safety, simplify plant design and increase resilience against complex accidents, such as Station Blackouts or Small Break Loss-Of-Coolant Accidents (SBLOCAs). In this context, South Korea’s APR1400 reactor design and its evolutionary successors (APR+and i-POWER) represent some of the most advanced Generation III+ pressurized water reactors, integrating both active and passive safety features. The Advanced Thermal-hydraulic Test Loop for Accident Simulation (ATLAS), built at the Korea Atomic Energy Research Institute (KAERI), provides a unique integral-effect facility to experimentally study these systems and to generate benchmark-quality data for the validation of system codes. This Master’s Thesis addresses the thermal-hydraulic analysis of passive safety systems in APR1400, APR+ and i-POWER reactors during SBLOCA sequences by means of TRACE, a safety code developed by the U.S. NRC. The primary focus is the ATLAS C2.1 test, a challenging transient that simulates simultaneous top and bottom breaks in the reactor pressure vessel, while the Passive Emergency Core Cooling System (PECCS) is available. Complementary work includes a simulation of a D2.1-like scenario, which incorporates the Passive Auxiliary Feedwater System (PAFS), showing how different passive systems work better when combined. The work begins with an extensive review of passive safety system concepts, thermal-hydraulic phenomena and thei categorization according to IAEA definitions. Natural circulation, condensation in the presence of non-condensable gases, thermal stratification, flashing and directcontact condensation are identified as key mechanisms determining system reliability. Special attention is paid to the South Korean designs, as these integrate solutions such as fluidic devices in SITs, Hybrid Safety Injection Tanks (HSITs) and advanced gravity-driven cooling systems. This theoretical framework establishes the basis for the modeling and interpretation of ATLAS experiments. Also, the ATLAS facility is described and the UPM-developed TRACE model for the ATLAS C2.1 test —whose primary side is shown in figure 0.1— is detailed. This model includes great detail in the representation of the reactor pressure vessel, steam generators, pressurizer, loops, pumps and, crucially, the passive safety systems. In the model, several control blocks are included to replicate the logic of valve actuation, heater power trips and depressurization criteria. Steady-state simulations reproduce the initial test conditions with deviations lower than 2% for most parameters, validating the model for transient use. The C2.1 post-test simulation results are then compared against the experimental data obtained at ATLAS. Key phases of the accident are successfully captured: (i) the pressure plateau, where primary pressure oscillations are driven by the on/off cycling of the main steam safety valves, (ii) the ineffective injection of the HPSITs at their 10 MPa actuation setpoint, (iii) the actuation of ADVs triggered by peak cladding temperature excursions, (iv) the delayed injections from both HPSITs and SITs, and (v) the final clad temperature re-excursion and test termination. The great agreement between the experimental and simulated results —as can be seen in figure 0.2— demonstrates TRACE’s robustness in replicating real accidental scenarios. Sensitivity analyses confirm that break size and location strongly affect depressurization dynamics, with bottom breaks producing the most severe consequences due to gravity-driven drainage. A central finding of this thesis is the failure of HPSITs to inject effectively at high pressure. Both experimental data and TRACE simulations showed that these tanks require either a very high nitrogen cover-gas pressure (≥ 16 MPa) or significant depressurization by ADVs to provide sustained coolant delivery. The analysis performed revealed the lack of high enough pressure differences to effectively inject water. Alternative configurations, such as relocating the pressure balance line or elevating the tanks, were also explored but provided no substantial improvement. These insights are highly relevant for the design of future passive systems. To extend the analysis, the TRACE model adapted to simulate a D2.1-like scenario is described. This model incorporates the Passive Auxiliary Feedwater System. Results demonstrate that PAFS, through natural circulation between steam generators and a Passive Condensate Cooling Tank, accelerates depressurization of the primary system after ADV opening, thereby enabling earlier long-term cooling injection and delaying the second peak clad temperature excursion. This highlights the potential of hybrid passive configurations, where systems complement each other to overcome the limitations observed when acting alone. An additional contribution of this work is the comparison of different TRACE patches in these specific simulations. While newer patches such as Patch 9 employed extremely small timesteps, leading to execution times of over four days, Patch 5 achieved accurate results with far greater efficiency, requiring only 13 hours for the same transient. This finding is important for both academic research and industrial safety analyses, where computational cost is a critical factor. This thesis leads to three main conclusions. First, the TRACE code, when properly nodalized and calibrated, is capable of reproducing complex SBLOCA phenomena observed in ATLAS with very good fidelity. Second, the current configuration of HPSITs presents fundamental thermal hydraulic limitations that hinder their reliable high-pressure operation, raising the need for design improvements. Third, the incorporation of additional systems such as PAFS proves ben eficial, demonstrating that the combination of passive features can extend accident management margins and strengthen overall reactor safety. In summary, this thesis not only achieves its initial objectives but also contributes valuable lessons for the future of passive safety design, experimental validation and system code application. By combining rigorous modeling, comparison with experiments and forward-looking reflections, it provides a foundation upon which both academic research and industrial practice can build. Finally, it highlights that continuous safety innovation is essential for nuclear power to remain a reliable pillar of the global clean energy transition.
Language: English
Open Access status: green
OpenAlex ID: W7113087329
Cites:
Cited by:
Related to: