Keywords

1 Introduction

The proposed advanced heavy-water reactor is a 920 MWth thorium-based vertical pressure tube-type boiling light-water-cooled and heavy-water-moderated reactor [1]. One of the important passive design features of this reactor is that the heat removal is achieved through natural circulation of primary coolant at all allowed power levels with no primary coolant pumps (Fig. 1).

Fig. 1
figure 1

Schematic of AHWR MHT

Full size image

Station blackout leads to feed water pump trip and loss of condenser vacuum which in turn results in turbine trip on low condenser vacuum signal. The bypass flow is not available due to loss of condenser vacuum. This leads to bottle up of the system, leading to pressure rise in main heat transport (MHT) system. Reactor trips on high MHT pressure signal. As the pressure continues to rise, IC valve starts opening at a pressure of 7.65 MPa to limit the pressure rise. The relief valve may open in the case of IC’s capacity which is not adequate. This analysis highlights the thermal-hydraulic conditions following station blackout for long duration and highlights the design mitigating provisions to limit the pressure and clad temperature rise and intercode comparison of CATHARE2 predictions with RELAP5 [2] findings.

2 Modeling of AHWR

2.1 Modeling of AHWR in CATHARE

The analysis has been done using French best estimate code CATHARE. CATHARE-2/V2.5_2 [3] is a thermal-hydraulic system code that solves the conservation laws for water and steam for a wide variety of single- and two-phase flow conditions. The AHWR model includes reactor core, 452 feeders and 452 tailpipes, 4 steam drums (SDs), 16 downcomers, inlet header, and steam pipes up to turbine in CATHARE code. The ECC system comprising of four segmental ECC headers is modeled with axial volume. The accumulators are modeled with CATHARE-specific gadget ACCU along with discharge line. The IC inlet pipe, IC inlet header, further 1440 tubes of H-X, outlet header, and IC outlet pipe are modeled. At the steady-state condition, core flow, drum pressure, core inlet temperature, steam drum level, and total steam flow are found to be 2143 kg/s, 6.999 MPa, 259 °C, 2.2 m, and 414.8 kg/s, respectively. These values of different important parameters are matching well with nominal design values.

Isolation condenser (IC) loops have been modeled in CATHARE as shown in Fig. 2. The IC inlet pipe is modeled with axial element ICINLPIP having five control volumes. The IC inlet header is modeled with axial element ICINLHD. Further, 1440 tubes of H-X are modeled with ICHEX having seven control volumes. The five control volumes are passing heat to GDWP. The outlet header and IC outlet pipe are modeled with axial element ICOUTHD and ICOUTPIP, respectively. The exchanger model is used to model the heat transfer to the GDWP. The HSPV valve is simulated using CATHARE-specific control valve. This modeling of IC path is combined with an already developed model of AHWR MHT and ECC system, and proper steady state is obtained.

Fig. 2
figure 2

AHWR MHT and IC nodalization in CATHARE

Full size image

2.2 Modeling of AHWR in RELAP5

The nodalization scheme for the main heat transport (MHT) system for AHWR in RELAP5 is shown in Fig. 3. Out of 452 channels in the core, 20 high-power channels are clubbed as hot channel and remaining 432 channels are clubbed as average channels. The average channel path is represented by component 100, and hot channel path is represented by component 101. Volume 4 to volume 8 of each component represents core channels. Volume 1 and volume 2 of these parallel paths represent vertical and horizontal portions of inlet feeders, while core bottom end is represented by volume 3 of these pipe components. Volume 9 of these components simulates core top end, while volumes, 10, 11, and 12 represent tailpipes. The effect of nodalization was evaluated in a separate study, and it has been insured that the present nodalization scheme yields accurate and converged results.

Fig. 3
figure 3

AHWR MHT and IC nodalization in RELAP5

Full size image

Four steam drums (SDs) are modeled by a single equivalent SD, i.e., component 102. The saturated volume of steam drum 102 is modeled separately using ten control volumes. The bottom part of the drum (subcooled portion) in between two baffles is represented by subcooled volume component 103. The feed water along with recirculation flow enters into the subcooled part of the drums. Flow from this portion of steam drums enters the downcomers. Feed water conditions are simulated using component 600. The suction header of feed pumps is modeled by component 653. Two feed pumps are modeled by components 656 and 657. Component 660 models discharge header. Individual feed lines are modeled by component 675.

Four downcomers lines, from each steam drum, are modeled by a single equivalent downcomer, i.e., 104. Inlet header is modeled by component 105. Steam lines emerging from the steam drums are represented by volumes 351. Two steam lines from each bank are joined together by volumes 353. Bypass valve is modeled by component 673. Combined isolation emergency stop (CIES) valves have been simulated by component 665. Governor valves have been modeled by component 671. Safety relief valves SRV 1 and SRV 3, and SRV 2 and SRV 4 mounted on steam lines are modeled by components 313 and 413, respectively. Appropriate core power modeling has been done considering power distributions (radial and axial), and point kinetics model has been used to simulate the reactor power. The thermal-hydraulic feedbacks are also modeled. Decay heat has been modeled using ANS 79 [2].

Each drum is connected to two isolation condensers (ICs), modeled by component 301 and 302, and outlet from IC is connected to subcooled volume 103 of steam drum. Each IC representing 70 tubes is divided into three axial volumes. Inlet cylindrical headers of each IC path are simulated with a single component (803, 804). Outlet cylindrical headers of each IC path are simulated with a single component (805, 806). The pipes in IC circuit are simulated by component 801, 802, 807, 808 as shown in Fig. 3.

The steady-state condition results of RELAP5 and CATHARE are compared with design nominal conditions. The comparison is given in Table 1.

Table 1 Steady-state comparison
Full size table

3 Results and Discussions

The station blackout event begins at time t = 0 s. System pressure rises due to the bottling of MHT as shown in Fig. 4. The reactor trip is credited on high system pressure signal once system pressure reaches 7.6 MPa.

Fig. 4
figure 4

SD pressure transient during SBO

Full size image

Further passive valve in the IC line opens at MHT pressure of 7.65 MPa. The heat removal from the ICs is initially small compare to reactor power. This results in initial pressure rise as shown in Fig. 4. Once the reactor power comes to decay heat level and comparable to IC heat removal, the system pressure rise is arrested. The system pressure remains around 7.65 MPa due to the controlled opening of the IC valve based on system pressure. The initial pressure rise trend following initiation of the transient is well captured by both codes, and peak pressure value is comparable. Overall, the pressure response predicted by both the codes is having a good comparison.

The heat removed from the isolation condensers is transferred to GDWP large pool water having inventory of 6000 tonnes. Heat is removed passively by GDWP initially by sensible heating and later by boil off once the temperature reaches to saturation temperature. The water inventory in GDWP is sufficient to effectively remove the decay heat for more than grace period of 7 days. Figure 5 shows the core flow behavior during station blackout scenario. As the reactor trips on the high system pressure signal, the core flow reduces as shown in Fig. 5. The natural circulation flow depends on the heat coming from the core. As the core power keeps on decreasing over the time due to reduction in decay heat, the core flow also keeps on reducing over the transient as shown in Fig. 5. The core flow predictions by RELAP5 and CATHARE analyses match very well in the initial part of transient; however, CATHARE code predicts slightly higher flow than RELAP5 in the later part of transient. The clad temperature initially rises marginally corresponding to saturation temperature. In later part of transient, as heat removal is same as decay heat coming to the system, the clad temperature remains stable as shown in Fig. 6. The effect of slight difference in core flow does not reflect in clad temperature prediction as shown in Fig. 6.

Fig. 5
figure 5

Core flow transient during SBO

Full size image
Fig. 6
figure 6

Clad temperature transient during SBO

Full size image

4 Conclusions

The prolonged SBO has been analyzed for AHWR using best estimate code CATHARE, and results are compared with already available results of RELAP5 code. The ICs are able to remove decay heat with the help of GDWP and maintain core temperatures well within the limit. Decay heat is removed passively by GDWP initially by sensible heating and later by boil off as seen in long-term calculation done with RELAP5. It has been found in analysis that IC system is capable to remove decay heat for more than 7 days. The decay heat removal through IC path along with passive moderator and end shield cooling keeps the integrity of different systems and maintains the core temperature well below the acceptance limit. Both the codes predict similar pressure response following SBO. The core flow prediction shows good match in the initial part of the transient; however, the core flow predicted by CATHARE code is little higher than the RELAP5 which results in later part of transient. The difference in core flow does not result in difference in clad temperature. Overall predictions made by CATHARE are in good agreement with RELAP5 results.